Colorfront Logo
Transkoder Logo

Transkoder 2018 Introduction

About Transkoder

Transkoder (TKD) is based on COLORFRONT’s award winning On-Set Dailies architecture and uses cost-effective, commodity hardware to deliver faster-than-real-time processing to transcode, decode, convert, wrap and process files to various high-end formats.

Transkoder 2018 is the latest release of COLORFRONT’s mastering system for digital cinema and high-end UHDTV production. COLORFRONT Transkoder is the ultimate tool for DCP and IMF mastering, offering the industry’s highest performance JPEG2000 encoding and decoding, 32-bit floating point processing on multiple GPUs, MXF wrapping, accelerated checksums, encryption & decryption, IMF/IMP and DCI/DCP package mastering, editing and versioning.

Transkoder 2018 adds features such as Dolby Atmos® audio and Dolby Vision™ HDR encoding, per frame metadata to OpenEXR, Civolution NexGuard forensic watermarking, audio level meter and vectorscope, CIE xy HDR analyzer, automatic calculation of MaxFALL / MaxCLL values to be inspected in a graph window for SMPTE ST 2086:2014.

Advanced DCP/IMF mastering features include Time Text format (SMPTE-TT), closed caption and subtitles, including keyframe animated PNG/text based subtitles, insert editing, bitrate and PSNR analysis, two pass encoding with PSNR and bitrate targeting, package validation with logging and reporting and international language support.

Running on the latest PC and Mac platforms, such as the latest Supermicro and HP Z840 workstations and the Mac Pro, Transkoder 2018 allows online editing of RAW camera data, converting to various output formats, color grading and mastering for HDR and rendering various 2K/4K/UltraHD high-end delivery and mezzanine formats.

Key Features of Transkoder 2018

  • DCP, IMF

    Industry-leading IMF and DCP mastering, authoring, validation, and QC tool. Supports the latest specifications, including HDR.

  • Performance

    4K DCP/UHD IMF and HEVC encoding at 100+FPS

  • RAW Support

    Highest quality RAW deBayering from and camera including ALEXA65, Varicam 35, Red Weapon 8K

  • COLORFRONT Engine

    32-bits per channel managed look pipeline built on ACES 1.0

  • Watermarking

    NexGuard forensic watermarking from Civolution with extreme performance even in 4K

  • High Dynamic Range

    Dolby Vision™ support and industry-leading HDR mastering tools

  • Dolby Atmos®

    Transkoder is certified to create DCP’s with Dolby Atmos® audio

  • Dual UHD/4K Output

    2x simultaneous real-time 4K video outputs for concurrent HDR / SDR mastering

  • PC, Mac And Cloud Platforms

    Runs on off-the-shelf PC and Mac workstations, not on obsolescence-prone, proprietary hardware. Transkoder also runs on cloud instances to be used with the REST API.

Features available only in the Windows version
  • Second Head Analyzer

  • Flexible GPU configuration for best performance

  • AS-02 and some flavours of Op1a creation

  • DirecTV deliverables

  • Fox Archive MXF encoding

  • Dolby Vision mastering features

  • Cut timeline based on light level

  • HDR-10plus light level statistics

  • Hardlink support on network storages (for IMF and DCP mastering workflows)

  • Avid Interplay Checkin

  • Dailies feature set (On-Set Dailies features) for Transkoder users

  • ARRI De-Bayer Algorithms (ADA)

  • Dual Video Output

  • NexGuard forensic watermarking

The Mac HW platform performance is subpar compared to the Windows configuration.

Accessing the Transkoder Website

All information with regard to COLORFRONT Transkoder software application can be found on the COLORFRONT Transkoder website at http://www.transkoder.com and in the Help Center at http://support.colorfront.com/hc.

Contacting Customer Support

In order to provide you support in the most efficient way, please collect the following items and attach them to your request. (Details on where to find these files are explained here.)

  • Transkoder log file, saved right after the issue

  • Crash dump files

  • Export of the problematic project (use the File menu to export it to a .zip file)

  • Any possibly related non-sensitive metadata or sample file

COLORFRONT Transkoder Support contact points are the following:

Web:

http://support.colorfront.com

Help Center:

http://support.colorfront.com/hc

Client Downloads:

http://www.colorfront.com/osd

Email:

support@colorfront.com

Phone (US):

+1 310 736 1036

Phone (EU):

+36 1 880 3970

Skype:

colorfrontdev

About COLORFRONT

COLORFRONT, based in Budapest, Hungary, is one of Europe’s leading post production facilities. The company was founded by brothers Mark and Aron Jaszberenyi, who together played a pivotal role in the emergence of non-linear digital intermediate. The company’s R&D team earned an Academy Award for the development of Lustre, Autodesk’s DI grading system, and a Primetime Engineering Emmy for COLORFRONT On-Set Dailies. Combining this in-depth expertise with a pedigree in the development of additional cutting-edge software, COLORFRONT offers today’s most advanced technologies for scanning and recording, DI grading, conforming, digital dailies, VFX, online and offline editing, cinema sound mixing, mastering and deliverables. For more information please visit http://www.colorfront.com.

1. Installation and Configuration

This chapter describes the setup of the supported hardware platforms, the installation procedures of necessary driver components, third party software, and the Transkoder application itself.

1.1. System Requirements

Transkoder runs on Apple OS X and Microsoft Windows and supports either OpenCL™ or CUDA® for rendering.

The Supermicro 7049 workstations with four processing GPUs, with a minimum of 96GB RAM are the supported PC configurations.

The HPZ840 workstations with two processing GPUs and an optional display GPU, with a minimum of 64GB RAM are also supported PC configurations where no real time playback of high bitrate HDR or mathematically lossless J2K video is required.

Graphics: Currently, we only support 8+ GB cards from the last three generation, thus Maxwell Titan-X, M4000, M5000, M6000, M60, and Pascal 1080, 1080 Ti, Titan X Pascal, Titan Xp, P4000, P5000, P6000, Turing 2080, 2080 Ti.

Processor: Dual 14 core Intel Xeon E5-2697 V3 processors.

Graphics: Two NVIDIA 1080Ti or 2080Ti GPUs are required. For stable playback on the UI, an additional P4000 is also recommended.

For UHD IMF work, use the more powerful 4 GPU Supermicro configuration. It is more capable of high performance GPU processing (e.g. UHD / 4K workflows, J2K processing).

Video: Video output via HD-SDI requires a video card. Currently, AJA Corvid 88, Kona3G, Kona4 and Blackmagic Design video cards are supported.

Monitor: A screen resolution of at least 1920 x 1080.

Storage: Transkoder uses open storage for media. The system should have fast enough storage for real-time playback and for the best rendering speed.

Operating system: 64-bit version of either Windows 7 SP1 Professional, Windows Server 2008 R2, or Windows 10 Pro.

The recommended PCI card order in the HP Z840 workstation is the following:

Slot# Slot Type Card

1

PCI-E3.0 x4

empty

2

PCI-E3.0 x16

NVIDIA 1080Ti / 2080Ti

3

PCI-E3.0 x8 in physical x16

This slot is covered by the NVIDIA 1080Ti / 2080Ti GPU installed in slot 2.

4

PCI-E3.0 x16

Storage controller card [Optional]

5

PCI-E3.0 x8

Video card

6

PCI-E3.0 x16

NVIDIA 1080Ti / 2080Ti

7

PCI 32/33

This slot is covered by the NVIDIA 1080Ti / 2080Ti GPU installed in slot 6.

Please make sure to use NVIDIA 1080Ti / 2080Ti cards with active cooling.

Supermicro SM-7049GP-TRT Recommended Configuration

Processor: Dual 14 core Intel Xeon Gold 6132 processors.

Graphics: Four NVIDIA 1080Ti or 2080Ti GPUs are required.

Video: Video output via HD-SDI requires a video card. Currently, AJA Corvid 88, Kona3G, Kona4 and Blackmagic Design video cards are supported.

Memory: 12 RAM modules, either 12x 8GB or 12x 16GB

Monitor: A screen resolution of at least 1920 x 1080.

Storage: Transkoder uses open storage for media. The system should have fast enough storage for real-time playback and for the best rendering speed.

Operating system: 64-bit version of Windows 10 Pro.

The recommended PCI card order in the Supermicro SM-7049GP-TRT workstation is the following:

Slot# Slot Type Card

JPCIE11

CPU2 SLOT11 PCI-E3.0 x4 in physical x8

Empty

JPCIE10

CPU2 SLOT10 PCI-E3.0 x16

Storage controller card [Optional]

JPCIE9

CPU1 SLOT9 PCI-E3.0 x16

Video card

JPCIE8

CPU2 SLOT8 PCI-E3.0 x16

NVIDIA 1080Ti / 2080Ti

JPCIE6

CPU2 SLOT6 PCI-E3.0 x16

NVIDIA 1080Ti / 2080Ti

JPCIE4

CPU1 SLOT4 PCI-E3.0 x16

NVIDIA 1080Ti / 2080Ti

JPCIE2

CPU1 SLOT2 PCI-E3.0 x16

NVIDIA 1080Ti / 2080Ti

Windows 10 systems using four GPUs need to be upgraded to at least 96GB system memory. This is essential for the GPU-based JPEG2000 encoding and decoding.

Please make sure to use NVIDIA 1080Ti / 2080Ti cards with active cooling.

Currently, the new Apple Mac Pro 6.1 with Dual AMD D700 GPUs with a minimum of 32GB RAM is the supported Mac configuration.

D700 Apple Mac Pro 6.1
Mac Pro 6.1

Workstation: The new Mac Pro 6.1 D700 with 2.7GHz 12-core processor with 30MB L3 cache, 32GB RAM or above.

Processor: 2.7GHz 12-Core Intel Xeon E5-2697v2 processor.

Graphics: Dual AMD FirePro D700 12GB GPUs for display and image processing.

Storage: Transkoder uses open storage for media. The system should have fast enough storage for real-time playback and for the best rendering speed.

Video: Video output via HD-SDI requires AJA Io 4K and AJA Io 4K Plus.

Operating system: Mac OS X 10.10 or higher.

Supported Thunderbolt external devices: AJA Io 4K, AJA Io 4K Plus, AJA T-TAP, Sonnet Echo Express III-D Supported external devices inside an external box such as a Sonnet Echo Express III-D: RED ROCKET-X, ATTO, or Myricom cards.

1.2. Video Output

The supported AJA devices are the following:

  • Corvid88

  • Kona4

  • Kona3G

  • Io 4K

  • Io 4K Plus

  • IoXT

  • T-Tap

Transkoder uses all Blackmagic devices through the same API, so the following supported devices may not complete:

  • Ultra Studio Mini Monitor

  • Decklink Extreme 4K.

A few video I/O specific features are only available on AJA devices, so in general this is the recommended brand. The AJA only features are the following:

  • 12-bit color depth 2

  • modes above UHD 30 fps 2

  • halved HFR mode 2

  • Dual video out modes 1

  • UHD on two cables 2

  • Dual 4K or UHD video at up to 60 fps3

  • 2SI support

  • RGB over 3G Level A mode

  • RS422 control

  • SDI embedded timecode

  • More than 8 SDI embedded audio channels

  • HDR10 / HLG HDMI Signaling

  • Dolby Vision HDMI Tunneling

  • 8K video out

  • Stereo mode

  • External sync

  • LTC

1: Corvid88 or dual Kona4 system only
2: Kona4, Corvid88, Io 4K or Io 4K Plus system only
3: Dual Kona4 system only

1.3. Installation Step-by-Step

Log in to the Help Center at http://support.colorfront.com/hc and navigate to the Download section. In this section (https://support.colorfront.com/hc/en-us/categories/115001258983-Download) there are step by step instructions and download links to all necessary components.

1.3.1. Installation Instructions for Windows

Installation Components to Download
Windows Installation Components (applies to the latest published revision)

Windows All Versions

Installer Package Name

Comment

osdxxxxx_2018_beta2.msi

Transkoder 2018 software installer

AJA_DesktopSoftware_Win_v15.2.zip

AJA installer for Kona3G / Kona4 / Corvid88

Avid_Codecs_LE_2.3.9_Win.zip

Avid QuickTime codec installer

Blackmagic_Desktop_ Video_Windows_10.4.1.zip

BlackMagic installer for BMD video cards

NVIDIA driver version 417.35 required.

NVIDIA driver installer for NVIDIA GFX cards

QuickTimeInstaller_7.79.exe

Quick Time codec installer

RED_Rocket-X_Installer_v2.1.34.zip (This driver includes the recommended 1.4.22.18.3 firmware.)

RED Rocket-X driver installer

xampp-win32-7.1.7.exe for new Transkoder 2018 installations, and xampp-win32-1.7.2.exe for Transkoder 2016 builds.

XAMPP installer

Installation Prerequisites

Make sure that Hyper-Threading is enabled in the BIOS.

On Supermicro configurations with four GPUs, please adjust the following BIOS settings:

  • Numa: Disabled

  • VT-x, PCIe bootrom: Disabled

Driver and Third Party Application Installation
  1. Download and install the appropriate NVIDIA GPU driver for the specific windows version as listed in the table above titled Windows Installation Components

  2. Adjust the NVIDIA parameters in the NVIDIA Control Panel by navigating to 3D Settings > Manage 3D Settings > Global Settings tab and make the following changes:

    • CUDA - GPUs - All

    • Multi-display/mixed-GPU acceleration - Single display performance mode

    • OpenGL rendering GPU - Quadro M6000

    • Power management mode - Prefer maximum performance

    • Vertical Sync - Use the 3D application setting

  3. Adjust the NVIDIA parameters in the NVIDIA Control Panel navigating to 3D Settings > Configure Surround, PhysX > PhysX Settings

    • Change the Processor setting to CPU

  4. Install QuickTime.

  5. Install the Avid Codec Pack.

  6. Install the XAMPP package if it is not installed already.

Important! Please read!
The reinstallment of XAMPP will likely destroy all databases causing project and/or metadata loss. Please make sure to make a backup of your data ahead of time.
For systems acting as the database server, please install the XAMPP package. This will install PHP, MySQL, Apache on your system. Make sure the Apache and the MySql services are running in the XAMPP control panel.
  1. Install the appropriate video drivers if using any of these add-on cards:

    • AJA Kona3G

    • AJA Kona4

    • BMD

Install Transkoder on Windows

To install or reinstall Transkoder on Windows, double click the osdxxxxx_2018_beta2.msi file and follow the prompts.

The installer creates both an On-Set Dailies and a Transkoder icon on the Desktop. This makes it easy to launch either On-Set Dailies or Transkoder without changing licenses or setting up custom shortcuts.

Icons installed on the Desktop
Icons installed on the Desktop
The Transkoder executable file osd.exe is in C:\Program Files\OSD2018\bin\ folder.

1.3.2. Install Transkoder on OS X

Installation Components to Download
OS X Installation Components

Installer Package Name

Comment

osdxxxxx_2018_beta2.pkg

Transkoder 2018 software installer

Avid_Codecs_LE_2.3.9_Mac.zip

Avid Codec installer

AJA_Mac_Retail_13.0.x.pkg.zip

AJA Io 4K and Io 4K Plus driver

Blackmagic_Desktop_Video_Macintosh_10.4.1.zip

Blackmagic Decklink driver

RED_Rocket-X_Installer_v2.1.34.zip (This driver includes the recommended 1.4.22.18.3 firmware.)

RED Rocket-X driver installer

xampp-macosx-1.7.3.dmg

XAMPP installer, please use this version for new Transkoder 2018 installations. Transkoder 2016 builds have it built into the Transkoder software installer.

Installation Prerequisites
  1. Update OS X to at least the version 10.10, or above

  2. Once updated to OS X 10.10 or greater, set the Allow applications downloaded from field value to Anywhere under OS X System Preferences  Security & Privacy.

Driver and Third Party Application Installation
  1. Install the Avid codecs.

  2. Install the appropriate drivers if using any of these add-on cards:

    • AJA card: install the AJA Kona 3G driver.

    • Blackmagic DeckLink: install the DeckLink driver.

    • Red Rocket-X: both the latest driver as well as the firmware need to be installed.

Make sure to install or upgrade the firmware first and the driver afterwards using the versions mentioned above.
  1. Make sure the Use all F1, F2, etc. keys as standard function keys checkbox under the Keyboard tab of the OS X System Preferences  Keyboard settings is checked.

  2. Make sure the Web Proxy Server checkbox under the Proxies tab of the Advanced.. window of the OS X System Preferences  Network settings is unchecked.

  3. Install the XAMPP package if it is not installed already.

Important! Please read!
The reinstallment of XAMPP will likely destroy all databases causing project and/or metadata loss. Please make sure to make a backup of your data ahead of time.

In order to run the database server must be reachable, by default it is localhost. Make sure your firewall settings do not prevent this.

For systems acting as the database server, please install the XAMPP package. This will install PHP, MySQL, Apache on your system. Make sure the Apache and the MySql services are running in the XAMPP control panel.

Installing Transkoder

To install or reinstall Transkoder on OS X, double click the osdxxxxx_2018_beta2.pkg file and follow the prompts.

The installer creates both an On-Set Dailies and a Transkoder icon in the Applications/OSD2018/bin folder. This makes it easy to launch either On-Set Dailies or Transkoder without changing licenses or setting up custom shortcuts.

The Transkoder executable file, OSD.app is in the /Applications/OSD2018/bin/ folder.

Please make sure that the mysql and the Apache daemons are running in the XAMPP controller.

1.4. Reinstallation or Upgrade

1.4.1. Reinstallation and Upgrade

To upgrade or reinstall Transkoder, use the desired installation file and follow the regular installation process detailed in Install Transkoder on Windows and Install Transkoder on OS X chapters.

When downgrading, please remove the later version of Transkoder first, then start the regular installation process.
Upgrading Database Scripts on a Centralized DB Server

To upgrade the database scripts when the database server is running on a centralized Linux database server, first upgrade Transkoder on the workstation. Then copy all the directories and scripts from the htdocs folder to the /var/www/html folder on the Linux database server.

For Example
Windows

C:\xammp\htdocs\

Mac OS X

/Applications/XAMMP/htdocs

For running projects, copy scripts from \htdocs\baseproject\ to the folder of the project under ..\htdocs and overwrite them.

1.5. Background Render Installation

The Transkoder Background Render application is installed along with the Transkoder installation package into the render sub-folder of the installation folder.

The Transkoder Background Render application is launched from the bin sub-folder of the render folder:

For Example
Windows

C:\Program Files\OSD2018\render\bin\rendertray.exe

OS X

/Applications/OSD2018/render/bin/rendertray.app

The rendertray application is configured to start and stop automatically when the user logs on and off.

The background service can also be controlled manually. Launching the tray application starts the background render service, and quitting the tray application stops the background render service.

The tray application is called "OSD Background Render" in the service list.

1.5.1. Configuring Background Render

Transkoder Background Render can be configured as follows.

You may need to modify the parameters of CommandLine section in MonitorService.ini under "render\bin" folder.

You can find the OSD Background Render configuration file on Mac OS X as /Library/LaunchAgents/com.colorfront.osd.renderqueue.plist This file is the same as 'monitorservice.ini' on Windows.

  • Make sure .plist file has -rw-r—​r-- permission. If not, then set it as follows: sudo chmod 644 /Library/LaunchAgents/com.colorfront.osd.renderqueue.plist

  • If the .plist file is not loaded, you can load it using this command: launchctl load \~/Library/LaunchAgents/com.colorfront.osd.renderqueue.plist

  • Set the address of your database server as the -h parameter and also check the database user name and password using the -u and -p options.

For Example

-h 192.168.1.1 -u root -p password -d projects -t renders -r C:/Program Files\OSD2018 -h localhost -u root -p password -d projects -t renders -r C:/Program Files\OSD2018

Similarly to the main TKD application, the IP address (or host name) of the database server must be set in the startup_template.xml of TKD Background Render application under "render\config" folder. This template will be used to generate the startup.xml for the background render jobs. Enable "BackgroundRenderCompatible" under Settings/Startup tab, if you want to use Background Render on the same system as the Master application.

If the Background Render is to be used on the same machine as the Master Application, the system memory should be upgraded to at least 32GB!

1.6. Application Removal

1.6.1. Removing Transkoder on Windows

You can remove Transkoder from "Control Panel\All Control Panel Items\Programs and Features" of Windows 7 or Windows 10.

1.6.2. Removing Transkoder on OS X

Make sure to create a backup of your lut, and dvd folders before removing Transkoder.

If using a software license, which is not stored on a dongle, make sure to create a backup of your license folder as well.

Move the /Applications/OSD2018 folder to Trash.

1.7. Control Panels

Transkoder will work with the Tangent Element, Wave, or Ripple control panels.

Tangent Element TK
Tangent Element TK Panel
Tangent Ripple Panel
Tangent Ripple Panel
The Tangent Element Tk and Tangent Element Mf is a nice combination for grading and transport control.

1.8. Licensing

1.8.1. Licensing Overview

Transkoder can be licensed with either a software license or with a Hasp Sentinal dongle.

The software license can be used only on the machine that it was created for, while the dongle can be moved between systems.

Software licenses need to be renewed once every quarter (3 months), while licenses used together with the Hasp Sentinal dongle need to be renewed only after a full year.

Licensing Method Quick Comparison

Comparison of the Licensing Methods

License Method ID Format Machine Restriction License Renewal Installation Method

Software

Host ID with Numbers and Letters

Valid Only for a specific Machine

Every Quarter (3 Months)

Copy to the license sub-folder of your installation folder.

Please refer to the the section titled License Installation: Software Method for details.

Hasp Sentinal Dongle

Dongle ID Number with Numbers Only

May be moved between systems using the dongle

Once a Year

The license is loaded on to the Hasp Sentinal dongle using Colorfront License Utility.

Please refer to the the section titled License Installation: Hasp Sentinal Dongle Method for details.

1.8.2. Obtaining the Host ID

  • If Transkoder is launched without a license present, it will display NO VALID LICENSE FOUND! and the host ID.

    Host ID
  • Transkoder will then exit and automatically copy the host ID to the clipboard.

  • If a dongle is present, but does not contain a valid Transkoder license, the dongle’s ID number will be used instead of the machine’s host ID and will be made up of numbers only.

  • If no dongle is present, the machine’s host ID will be used and be made up of numbers and letters.

1.8.3. Requesting a Valid License

Please email either the Hasp Sentinal dongle’s ID number or the machine’s host ID to support@colorfront.com to receive a valid license. The license file will be sent as an email attachment named osd.lic.

Demo or Evaluation License

The procedure for obtaining a demo or evaluation license is the same as for a normal license. Please contact support@colorfront.com for more information.

1.8.4. License Installation

License Installation: Software Method

The Colorfront Transkoder software license file is a plain text file containing the license string. It must be named osd.lic and placed in the license sub-folder of the Transkoder installation folder.

For Example
OS X

/Applications/OSD2018/license/osd.lic

Windows

C:\Program Files\OSD2018\license\osd.lic

Please refer to the the Appendix Troubleshooting section titled Software License is Received as Text in case the license is sent as text instead of an attachment.
License Installation: Hasp Sentinal Dongle Method
  1. Plug the Hasp Sentinal (USB) dongle into the workstation where you have a valid license.

  2. Start the Colorfront License Utility executable.

  3. Load a license file or copy-and-paste a license code that you received via e-mail.

  4. Double check that the dongle’s Host ID and the license code’s first three (or two) characters are the same, otherwise the licensing will not work.

  5. Upload the license to the dongle by pressing the Upload to dongle button and use it in any machine where you want to.

If the workstation doesn’t see the Dongle, please click on the Refresh button. Please refer to the the Appendix Troubleshooting section titled Issues with the Dongle if the problem persists.
Transkoder automatically reads the license from an inserted dongle and ignores the computer’s osd.lic file.

The license is loaded on to the Hasp Sentinal dongle using Colorfront License Utility which is found in the bin sub-folder of the Transkoder installation folder.

For Example
OS X

/Applications/OSD2018/bin/

Windows

C:\Program Files\OSD2018\bin\

Colorfront License Utility
The first 12 numbers of the license should match the dongle’s ID number.

1.8.5. License Renewal

Starting with Transkoder 2018, there is a simple one button solution which automatically refreshes the license string on the dongle.

1.8.6. Understanding the license key code text

The first three character section of the license code is the host ID (of the computer or of the dongle). This is followed by the start and the end date of the license validity. Next is the ID of the software package. This is followed by the machine ID and the start date again, and finally the key code string.

1.8.7. Background Render Licensing

If you have valid Transkoder license, then you can use TKD Background Render application with the same license on that system.

Render only systems, nodes in a local render farm or virtual machines in the cloud may have a render only licenses. In case of local render nodes either license files, or dongles can be used. Transkoder will not run with the user interface if a render only license is installed.

1.9. Display Calibration

Transkoder has a built-in Display Calibration feature that lets you adjust the system settings by applying a filter to the screen to display a color-proof image while using the software. You can choose already existing presets, initiate a new calibration process to create a new preset or 3D Mesh LUT files for further use, or execute certain customized measurements.

Calibration only works with the X-rite i1 Pro spectral radiometers.

The View  Calibration menu has the following options:

Calibrate Display

It initiates a new calibration process and creates a preset from the results. For the detailed process and instructions, check the Calibrate Display chapter.

Verify Calibration

It verifies the accuracy of the native (Off) or the currently applied preset of the GUI monitor. This command does not create a new preset, but simply checks if the display needs further calibration or not.

Measure Single Patch

This option allows you to measure a single custom color patch.

Measure Custom Patches

This option lets you inspect custom color patches defined in an external descriptor file. See the Measure Custom Patches chapter for instructions.

Off

With this option selected, Transkoder does not apply any display calibration presets.

Display calibration options and presets
Display calibration options and presets

1.9.1. Calibrate Display

Do the following steps to calibrate your display and create a preset and/or a 3D Mesh LUT file for further use:

  1. Connect the device to the workstation.

  2. Install the device if it has not been installed yet.

    • On Windows, navigate to the Program Files{app-installation-folder}\installer\i1prodriver folder, right click the i1_pro.inf file and choose the Install option.

  3. Choose the View  Calibration  Calibrate Display menu option to bring up the Calibration Setup window.

    Display calibration setup
    Display calibration setup
  4. Adjust settings and press Start.

    GUI Monitor

    Choose this option to calibrate your GUI display, and create a new preset and calibration files from the results. The preset will appear in the menu for further use and will also be applied automatically to the GUI.

    Video 1

    This option is for calibrating the Video Output 1. Save the results in the OSD/lut/Video1CAL.3dmesh file for further use.

    Video 2

    Choose this option to calibrate the Video Output 2 and save the results in the OSD/lut/Video2CAL.3dmesh file for further use.

    3D LUT

    This option is for creating a 3D mesh file from the measurement results and adding it to a custom folder for further use.

    Rec.709

    Choose this if the color space of the target display is Rec.709.

    DCI P3

    Choose this if the color space of the target display is P3.

  5. Follow the instructions on the screen.

    • Place the device onto the clean white tile. Thus the device initializes and calibrates itself before the measurement.

    • Place the device to the center of the screen for the measurement. Make sure not to exceed the allotted time for the calibration.

  6. Wait for the measurement to be made. A round white patch will appear on the screen, followed by other color patches such as red, green, blue, various gray shades, cyan, magenta and skin tone. The device measures the Delta E values for each one, which describes the difference from the required color.

  7. Calibration Presets and Files will be then created. If you have chosen the GUI display option, the GUI will also be automatically adjusted.

Display calibration patch
Display calibration patch

1.9.2. Verify Calibration

The Verify Calibration option verifies the accuracy of the native (Off) or the current preset of the GUI monitor. This command does not create a new preset, but simply checks if the display requires further calibration.

Display calibration verification failed
Display calibration verification failed

1.9.3. Measure Single Patch

This option allows you to measure a single custom color patch. Set the size and the color of the patch by its RGB coordinates, and inspect the results in the Yxy color space.

Display calibration on a single patch
Display calibration on a single patch

1.9.4. Measure Custom Patches

This option allows you to inspect more custom colors defined by their RGB coordinates in an external descriptor file. The file syntax is one set of R, G and B values in a row, separated by commas for a color. The results are in Yxy (three coordinates) color space, or throughout the whole spectrum (36 steps between 300 nm and 700 nm) depending on the option chosen. The output file with the results will be put in the same folder as the source file.

test.txt
1,0,0
1,1,0
1,1,1

1.9.5. Calibration Presets and Files

When initiating a calibration with the View  Calibration  Calibrate Display option for the GUI Display, the following preset files will be created in the Users/<user-name>/ folder:

.cpf

This file is made for the GUI adjustment, and it will also appear as a preset option in the View  Calibration menu for later use in all Colorfront products.

.3dmesh

This file can be loaded to a LUT node in the node pipeline, or as a Custom LUT for Video Output within Transkoder, and also to any external LUT box.

.log

This file includes the input, output, and Delta E values of the measurement in a human readable format.

Display calibration presets
Display calibration presets
To delete or rename a certain preset, search for the corresponding .cpf file in the file system, and delete or rename it.
To create a 3D mesh file for further use only, use the Video 1, Video 2 or 3D LUT option in the Calibration Setup window.

1.9.6. Custom LUTs for Video Outputs

To apply existing custom 3D mesh LUT files on the video outputs, choose the Calibration option in the Advanced Video Output Settings section of the Settings Page, and copy the required .3dmesh file(s) to the OSD2018/lut/ folder, and rename them according to the following table:

Setting Option Preset File for Video 1 Preset File for Video 2

Off

-

-

SDR

Video1SDR.3dmesh

Video2SDR.3dmesh

HDR

Video1HDR.3dmesh

Video2HDR.3dmesh

HDRC

Video1HDRC.3dmesh

Video2HDRC.3dmesh

Be careful to avoid accidentally applying a LUT twice. For example, when an SDR-HDR conversion is applied in the node pipeline plus a custom LUT is set on the video output by the Calibration option on the Settings Page.
After choosing a new video output LUT, restart Transkoder to apply the changes.
Only GUI video texture, GUI thumbnails, and video output is adjusted, while the render results remain untouched.
If file not found then no LUT will be added.
Shortcut Ctrl+F12 activates the SDR Output LUT for Video 1, while Ctrl+F10 activates the HDR Output LUT for Video 1.

1.9.7. Apply Custom LUT to Second Head Analyzer

To apply a custom LUT to the Second Head Analyzer display image, choose the color space of the target display in Analyzer  Analyzer Colorspace and set Analyzer  Analyzer display LUT to Auto.

Analyzer Display LUT
Analyzer display LUT

2. Projects

Work in Transkoder is done via projects. A project is used to manage a collection of assets, their metadata, timelines, bins and encoder settings. While there are a few global settings in Transkoder like tool presets or project templates, most of the configuration parameters are saved within the project. Transkoder projects are stored in a database - excluding saved high resolution stills, and can be exported and imported on another system. Projects do not contain a copy of any media, only links to files on a file system.

2.1. Create New Project

To create a new project:

  1. Launch the Transkoder

  2. Click on the current project’s name to enter the Project Load Page

  3. Click on the Create New Project button

  4. Enter a project name in the next screen shown:

    Transkoder Project Name Entry
    Transkoder Project Name Entry
  5. Press the ENTER key after having entered a project name.

  6. If needed, adjust the project name by double clicking on it. Projects can be relabeled after being created. Please see Renaming a Project After Project Creation chapter for more information.

  7. Set the desired default Resolution of the project. This drives the resolution of the default Reframe node and can be adjusted globally, or clip based later. Clicking on the number values on either side of the letter "x" will permit the entry of a custom value for the resolution, but this value will be validated by Transkoder and rejected if not valid.

  8. Click on the FPS selector to set default playback frame rate of the project.

  9. Click on the Color Correction parameter to define the desired default node pipeline. For more information on these pipeline presets please refer to the Color Science chapter.

  10. If needed choose a Template project to use the settings, node pipelines and encoder settings of a previously saved project template. This will override the parameters set above.

  11. Click on the Launch Project button.

    Transkoder Launch Project
    Transkoder Launch Project

Once a new project has been created and launched, an initial view will be presented:

Transkoder Initial View
Transkoder Initial View

The new project is empty, and an empty timeline and the Bin Window is shown. You can import media into the Bins and build Timelines, or go to the Timeline menu and import a CPL.

While hovering over blue text editable parameters turn yellow.

2.2. Project Load Page

The Load Page is where the project’s Timelines, Bins, Cards, etc. are listed, and can be launched, renamed, or deleted. For more information on these items see the Main Load Page chapter. By clicking on the project name, the Project Load Page opens, where one can rename, delete, create new projects or load another existing project. Clicking again the project name one can return to the main Load Page.

The Show Timelines Ordered field in the User Settings section of the Settings Page is used to specify sorting order of projects and timelines. They may be sorted alphabetically or by time last modified in ascending or descending order. Transkoder also displays the last modification date.

2.2.1. Load Existing Project

Click on the project name in the list on the right and click Load Project on the bottom, or simply double-click the project name in the list.

2.2.2. Delete Project

To delete a project:

  1. If it is not the current project to be deleted select the project in the list

  2. Click on the Delete Project label located on middle left hand side.

    • To confirm the deletion, click on the Yes, delete project label.

    • To cancel the deletion, click on the No, cancel delete label.

This will remove the project from the database and cannot be undone, but will not erase any of the project’s assets.
Transkoder Deleting a Project
Transkoder Deleting a Project

2.2.3. Rename Project

To rename a project after it has been created, or launched:

  1. Return to the Project Page.

  2. Click on the Rename Project label located on the middle left hand side.

  3. Modify the previously entered project name to the new one desired.

  4. Press the ENTER key.

A renamed project is different from a project created originally with the proper name. The URL of the project scrips, the location of the project specific PHP scripts on the database server and the database name in the - central or local - database is not going to be changed. By renaming the project a new "alias" will be created for it, which overrides the original project name within the application.
Transkoder Renaming a Project
Transkoder Renaming a Project

2.2.4. Project Info Page

The Project Info page displays an overview of the settings chosen during project creation. Some of the settings, such as default playback frame rate and resolution can be adjusted here, some settings such as the project path can be only inspected. To view the Project Info page go to the Project Page and click on the Project Info label located on the left hand side. To close the Project Info page click on the Done label located on the bottom left hand side to return to the Project Page.

Transkoder Project Info Page
Transkoder Project Info Page

2.2.5. Save Project as Project Template

To make it easier to set up new projects users can save project templates. By saving a project as a template, later on users can create new projects easily with matching settings. The following attributes will be stored:

  • Project settings from the Settings Page

  • Node pipeline templates for all camera formats

  • Global node template

  • List of encoders on the Encode Page

2.2.6. Promote Project Created in Earlier Version

To avoid compatibility issues the project list does not show projects created with an earlier major version. It is possible to promote a project though, so it appears in the new version, hiding it in the old one. The is because only forward compatibility is guaranteed. To promote the project please follow these steps:

  • Open the database through phpMyAdmin. On a local machine this is localhost/phpmyadmin

  • Open the Projects database on the left pane

  • Open the projects table within this database and locate the project in the table

  • Click on the pencil icon in the corresponding row

  • Increase the version number in the ProjectVersion field. For example replace OSD_V4 with OSD_V5.

2.3. Main Load Page

The main load page is an interface to create and access different timelines in the current project. For accessing the different timeline types, use the tabs on the top right side of the screen, where the list of the respective entries will be listed. Each tab has its own relevant context menus on the left-hand side.

Transkoder Project Page - Timeline list
Transkoder Project Page - Timeline list
It is possible to load custom timelines, CPLs or bins from withing the Bin Window as well. For more information see Bins.

2.3.1. Timelines tab

This tab includes custom timelines (in other words, cuts) created by the user, as well as conformed or imported timelines such as CPLs, EDLs or Final Cut XMLs. Upon initial import of EDLs, or FCP XMLs through the timeline menu, the corresponding entry in the list will be indicated as "NEW". Loading such a timeline will trigger the conform.

Under this tab the following operations are available:

Create New Timeline

Create a new timeline by clicking on the top left-hand side.

Import CPL

Imports a DCP or IMP CPL. The name of the timeline will be the Content Title.

Rename Timeline

Renames the currently selected timeline.

Delete Timeline

Deletes a timeline.

Load With Handle

For loading all the sources of a conformed timeline, such as an EDL or Final Cut edit with handles. This option is only available after the initial load. The handle length is defined in the Conform tab of the Settings Page. Loading an EDL with handles also eliminates retimes.

Load Timeline

Loads the selected timeline.

2.3.2. Bins tab

Bins are collections of clips that Transkoder is able to load and render. Please refer to the section titled Bins for more information. These are not "edits", so clips in a bin timeline cannot be trimmed or cut.

Under this tab the following operations are available:

Ingest Media

Opens Copy Central to copy media into the project.

Load Bin

For loading a bin by clicking on the bottom left-hand side. It becomes visible as a context menu, when a bin name is highlighted.

2.3.3. Cards/Dates tab

This tab is relevant for dailies workflows. If the user opens a project in Transkoder that was created either by On-Set Dailies (as a dailies project) or Express Dailies, this list will show the cards parsed into the database or found in the folder structure within the project.

By clicking on the Cards tab again, the list switches to the Date list. For more information consult the On-Set Dailies User Guide.

2.3.4. VTRolls tab

Similarly to Cards, this tab is relevant for dailies projects. In On-Set Dailies, in a dailies project, operators can organize media into virtual rolls called VTRolls as the unit of delivery. These can be accessed though this tab.

2.3.5. XML tab

Whenever a timeline is loaded in Transkoder, there is a temporary XML file generated. The project XML list displays these recent temp files. It is possible to load these temporary xml timeline files at a later time in XML load mode.

This feature is used primarily for engineering/support purposes. In this load mode, the timeline and grades are loaded from temporary cache files instead of the database.
These XML files may be out-of-date compared to the database, and they are not saved back to the database.

3. Bins

Bins are collections of clips that Transkoder is able to manage. There are auto-generated smart bins, such as the master bin with all media in the project, and user bins that the operator can create. All bins can be sorted by any metadata, and flexible search expressions can be used to find any media easily.

The Bin Window, particularly the master bin is the primary interface to manage video, audio and subtitles within the project. User created timelines, aka Cuts, and CPLs are also listed in this window. These are different than Bins in the sense that these timelines are not simple collections of media, but specific edits of events. One can inspect the contents of these timelines in all three characteristics: video, audio and titles. Any timeline can be loaded by selecting it and pressing Ctrl+o.

3.1. Bin Window

The Bin Window is made up of two main areas:

Selection pane

contains the Library, Generated, User, Cuts, CPL, Days and Looks entries, and is located on the left side of the Bin Window.

Asset list

contains the list of Video, Audio, and Text assets belonging to each entry in the selection pane, and is located on the right side of the Bin Window.

Press ALT+J to open the Bin Window, or use the Window menu.
Transkoder The Bin Window
Transkoder - The Bin Window

3.1.1. Add New Bin

  1. To add a new bin, first click on the small ⊕ (plus sign) button in the bottom left corner of the selection pane. A new bin with the name "Bin_2" will be created in the User bin list.

  2. Double click on the name of the new bin to rename it.

  3. Drag media from the timeline, from other bins, or from a file browser to load media into this bin.

3.1.2. Remove Bin

  1. Highlight the bin to be removed by clicking on its name in the selection pane, located on the left side of the Bin Window.

  2. Click on the small ⊖ (minus sign) button in the bottom left corner of the selection pane,

    or

    Press the Ctrl+Backspace key combination.

When removing bins, no clip is removed from the database. Clips are simply removed from the bin.

3.1.3. Rename Bin

  1. To rename a Bin, double click, or press ALT and click the bin name to be changed in the selection pane, located on the left side of the Bin Window.

    A text entry cursor will appear at the end of the currently entered bin name.
  2. Enter the desired name.

  3. Press the Enter key.

Transkoder Renaming a Bin
Transkoder - Renaming a Bin

3.2. Manage Assets within Bins

3.2.1. Add Assets to Bins via File System

To add assets, such as clips, to bins via the file system, click on the small ⊕ (plus sign) button in the bottom left corner of the Bin Window, and a file system browse dialog will appear on the screen. Browse to the asset to be added, and click on the OK button.

Transkoder - Adding assets via file system
Transkoder - Adding assets via file system

Assets can also be added by dragging them from any file system browser and dropping into the Bin Window.

3.2.2. Add Assets to Bins from Thumbnail Timeline

To add assets to bins from the thumbnail timeline, drag one or a selection of thumbnails from the bottom thumbnail timeline to the large empty area in the bin window.

To drag multiple thumbnails, hold CTRL and click the thumbnails for selecting them from the thumbnail timeline.
Transkoder - Adding assets from thumbnail timeline
Transkoder - Adding assets from thumbnail timeline

3.2.3. Remove Assets from Bins

To remove assets from bins, highlight the assets to be removed, and press Ctrl+Backspace while the mouse pointer is within the Bin Window, or click on the small ⊖ (minus sign) button in the bottom left corner of the Bin Window asset list.

3.2.4. Search Clips within Bin

Users can filter the contents of a bin by entering Google-like search queries in the search field, located along the top of the Bin Window.

For instance, one can easily find all DPX sequences in the project by simply typing *dpx in the search field of the master bin. To find all "A" camera clips, type CamRoll:A* in the search field.

For the list of available query keywords and rules refer to the Appendix section titled Bin Window Search Query Keyword List and Bin Window Search Query String Rules.

The search expression is evaluated in real time as the search string is being typed into the search field. Hence, when you start typing the desired expression, in case it matches the listed keywords, a suggestion keyword will appear. Press the Tab key, and it will be completed automatically.
Transkoder Search in the Bin Window
Transkoder - Search in the Bin Window

3.2.5. Sort Clips within Bin

The second input field located to the right of the search field is used to sort the results. Multiple sorting can be done by adding query keywords one by one without using a comma. One can use any metadata associated to the clips to order them. A sorted list can be later manually reorganized to fix any ordering issue.

Example Sort Expression

CamRoll Length AudioFramerate

Transkoder Sort Clips in a Bin
Transkoder - Sort Clips in a Bin

3.2.6. Copy Assets Between Bins

  1. To copy assets from one bin to another, first open the source bin by double clicking on the bin name.

  2. Select the assets to be copied by clicking on them. Make sure to hold CTRL when selecting multiple clips for copying.

  3. Drag and drop the asset(s) onto the name of the target bin.

Transkoder Bin Window - Dragging Video and Audio Between Bins
Transkoder - Bin Window - Dragging Video and Audio Between Bins

3.2.7. Reveal Clip Media Files in Browser

When working with previous projects, it may be helpful to easily find where source files of the clips are located in the file system.

  1. Select the clips you want to localize either in the Bin window or on the Timeline.

  2. Choose the File  Reveal Selected Clip in Browser menu item. Transkoder will open up to a maximum limit of five new external file browser windows showing the location of the selected clips in the file system. If the media files are located in the cloud, it will show the remote path on the screen.

On Windows, the external file browser is Explorer while on OS X it is Finder.

3.3. Change Bin Window Layout

3.3.1. Selection Pane

To hide or show the selection pane, click on the leftmost icon in the upper right corner of the Bin Window. Click on it again to return to the previous state.

Transkoder Bin Window - Hidden Selection Pane
Transkoder - Bin Window - Hidden Selection Pane

3.3.2. Thumbnail View

To switch the asset list to thumbnail view, click on the middle icon in the upper right corner of the Bin Window.

This is the default view and displays a thumbnail of the assets with the asset Name underneath.
Transkoder Bin Window - Asset List Thumbnail View
Transkoder - Bin Window - Asset List Thumbnail View

3.3.3. Details View

To switch the asset list to details view, click on the rightmost icon in the upper right corner of the Bin Window.

This view displays a thumbnail of the assets with the assets' Name, Clip Name, LabRoll, MediaPath, TCin, Duration, Length, Framerate, and Channels to the right of the thumbnail for video, and also MediaPath, AudioFile TCin, Duration, SoundRoll, Samplerate, Name, Channels, and LabRoll for audio.
Transkoder Bin Window - Asset List Details View
Transkoder - Bin Window - Asset List Details View

3.3.4. Filter Based on Asset Types

To hide or show video assets in the Bin Window asset list, click on the Video button on the top right-hand side. Click on it again to return to the previous state.

Transkoder Bin Window - Hidden Video Assets
Transkoder - Bin Window - Hidden Video Assets

To hide or show audio assets in the Bin Window asset list, click on the Audio button on the top right-hand side. Click on it again to return to the previous state.

Transkoder Bin Window - Hidden Audio Assets
Transkoder - Bin Window - Hidden Audio Assets

To hide or show subtitle and caption assets in the Bin Window asset list, click on the Text button on the top right-hand side. Click on it again to return to the previous state.

3.3.5. Selection Pane List

To hide a list item in the selection pane, position the mouse cursor over the entry to be hidden. A label titled Hide will appear to the right of the entry. Click on it to hide the list item.

Transkoder Bin Window - Hide Selection Pane List
Transkoder - Bin Window - Hide Selection Pane List
Transkoder Bin Window - Hidden Selection Pane List
Transkoder - Bin Window - Hidden Selection Pane List

To show a list item in the selection pane, position the mouse cursor over the entry to be shown or unhidden. A label titled Show will appear to the right of the entry to be shown. Click on it to show the list item again.

Transkoder Bin Window - Show Selection Pane List
Transkoder - Bin Window - Show Selection Pane List

4. Timelines

When pressing Alt+I, the Timeline Window shows up. This is a standard graphical representation of a media sequence or cut, which can include all types of clips (video, audio, subtitles, etc.). The Timeline Window is not a floating window, it is always docked on the bottom of the UI. On the top of the timeline, there is a configurable metadata bar. To change the displayed metadata entries, adjust the Metadata Overlay setting in the Metadata Settings section of the Settings Page.

Transkoder Single Track Timeline with Condensed Audio Tracks
Transkoder Single Track Timeline with Collapsed Audio Tracks

4.1. Multitrack Timeline

In Transkoder, the user can add any number of audio or video tracks to the timeline. Subtitle clips and reel markers can also be added to their dedicated tracks, where all clips and events can be manipulated and edited the same way. To add a new video track, use the shortcut Alt+V. To add a new audio track, use Alt+Shift+V.

Transkoder Multitrack Track Timeline
Transkoder Multitrack Timeline

When multiple video track files are present, the following view modes or Timeline Modes can be activated from the Timeline menu:

  • Enable Vertical Editing - This is the default editing mode for most editing applications, where the user sees the image from the highest video track.

  • Enable All Tracks as Source - Enabling this one will see all of the video tracks as input sources in the node page. This allows the combination of multiple source files from the video tracks, e.g. to calculate PSNR to measure similarity.

  • Enable DualTrack - In this mode Transkoder will do vertical editing with the V1 and the selected track. So, basically the user will see the clips from the selected track. But if there is a gap on the selected video track, clips from the V1 track will be shown.

When the operator needs to collapse the video tracks, the Merge Down Track option can be used from the Timeline menu. This operation will merge down the selected track to the lower track, potentially overwriting clips.

4.1.1. Compositing with Alpha Channel

One can composite a clip with alpha channel over the V1 track. Please load a QT ProRes 4444 clip with alpha into the bin, and add to the V2 track. Make sure the Enable Vertical Editing is on in the Timeline menu. Users can disable this feature by selecting the V2 clips and clicking on the "Timeline/Toggle Transparency of Selected Clips" menu item.

Title composited over the V1 video media
Title composited over the V1 video media

4.2. Timeline Auto-Edit Function

Some deliverable formats require specific slate/bars/black configuration at the head of the timeline, and a predefined timeline timecode. The Auto-Edit window can be used to make it easy to create such timelines and can be accessed by selecting the Edit  Auto-Edit menu option.

The Auto-Edit Window
The Auto-Edit Window

The Auto-Edit window has a pop-up selector where one can choose any of the presets. The timeline configuration is defined by a string of specific, easy to use keywords:

  • Black: Black frames without audio.

  • Bars: Color bar with 1Khz audio.

  • Slate: This should refer to an existing Burn-in preset, like "5s SLATE MYSLATE" where MYSLATE is an existing burn-in node preset already saved in the database. The event will be a black image with the corresponding burn-ins visible. One can use fixed strings or keywords like $TRT$ or $maxfall$ and $maxcll$.

  • Program: Defines the start timecode of the main program.

  • Freeze: Freezes the last frame of the program for the given time.

One can edit this string to create any custom preset and simply save it for later use.

For Example
2s black,3s bars,program 01:00:00:00,5s black

2 seconds of black followed by 3 seconds bars and tone. Program starts at 1 hour, followed by 5 seconds black

5s10 black,1 bars,1s23 black,program 02:00:00:00

5 seconds and 10 frames of black followed by a single frame bars and tone, effectively a 2-pop. Program starts at 2 hours

5s SLATE MYSLATE, 5s black,program 01:00:00:00

5 seconds slate, where the "design" of the slate is coming from a burn-in preset called "MYSLATE". Then 5 seconds black before the program.

Timeline corresponding to the string *2s black,3s bars,program 00:01:00:00,5s black*
Timeline corresponding to the string 2s black,3s bars,program 00:01:00:00,5s black
Any new project by default has a Black and a Color Bar event in the master bin, that can be stretched to the desired length manually as well on the timeline.

4.3. Audio Channels

To collapse or expand the audio channels in the Timeline window please double-click on the header area of the audio track.

Transkoder Single Track Timeline with Video and Audio Tracks
Transkoder Single Track Timeline with Video and Audio Tracks

4.4. Add and Manipulate Clips

The drop-to-position menu item or the V shortcut will place the currently selected clips in the Bin window to the playhead position. Depending on Ripple is On or Off clips on the timeline will be shifted, or overwritten. Drop-to-TC mode will force clips to be dropped at the - record - position defined by the timecode of the clip. This mode can be adjusted from the Menu, or using a small icon on the top-left corner of the timeline window.

When working with previous projects, it may be helpful to easily find where project clips are located in the file system. The File  Reveal Selected Clip in Browser menu item will do just that. This function opens a new external file browser window showing the location of the selected clips in the file system up to a maximum limit of 5 external file browser windows. It works with clips selected in either the Bin window or the timeline.

On Windows, the external file browser is Explorer and on OS X the external file browser is Finder.

Navigation on the timeline is possible using the mouse, keyboard shortcuts or control panel. There are two different types of timeline views to help navigation and operations such as finding shots, copying grades or editing events. On the right side of this bar is the Edit/Grade switch which controls how mouse interaction should work:

  • Edit mode: basic editing operations are possible with the mouse, such as reorganizing clips, moving or trimming

  • Grade mode: color correction settings can be dragged from a clip to others

4.4.1. Edit Mode

  • Change timeline window size - click the upper edge of the timeline area (A), hold down the mouse button while moving the mouse up and down over the timeline area.

  • Change track size - click the upper edge of the track (B), hold down the mouse button while moving the mouse up and down over the timeline area

  • Navigate by left dragging just above the clips where the timecode track is showing

  • Move tracks up and down in the timeline window - right-click drag up and down the track headers (V1,A1,A2…​)

  • Horizontal scroll - press and hold the right mouse button, while moving the mouse to the left/right over the timeline area

  • Zoom in/out - press and hold the Ctrl key and the right mouse button, while moving the mouse up and down over the timeline area

  • Fit the cut into the timeline - press C key while the mouse pointer is over the timeline window

  • Select multiple clips - press and hold the Ctrl key and drag over the clips to be (de)selected

  • Trim the clip - click the left/right edge of the clip ( D ), hold down the mouse button while moving the mouse to the left/right

  • Move the clip - click the clip itself, hold down the mouse button while moving the mouse to the left/right, or drag&dropping to another timeline track

Flexible Speed Assignment

Transkoder has functions to deal with clips shot at frame rates different than the project frame rate. The operator could either incrementally increase or decrease the speed by a factor of two, or use the Edit  Compensate Off-Speed menu option. The latter adjusts the speed of the clip, so it plays real-time in the project. However, it does not work if there was no proper metadata indicating the recording frame rate.

edit change speed01
Menu items to manage clip speed

The operator can also adjust the speed of the clip manually by choosing the Edit  Set Custom Speed menu option. This opens up a new window where any speed factor can be entered or a preset can be chosen.

edit change speed02
Flexible speed assignment
A value of 100 means no speed change and can be used to reset the clip to the original speed.

4.4.2. Grade Mode

Please keep in mind that grading in Transkoder is done either by running it as On-Set Dailies, or activating the grade mode on the timeline.

Transkoder Grade Mode on the Timeline
Transkoder Grade Mode on the Timeline

In grade mode the user can drag-and-drop color correction settings from a single clip on the timeline to the still store/look store or to one or multiple clips on the timeline. Please see the section titled Applying / Reapplying Grades for more information.

4.5. Thumbnail Timeline

Transkoder Thumbnail Timeline
Transkoder Thumbnail Timeline

For color correction, the thumbnail timeline may be a better choice. Press numpad 1 to enable/disable the thumbnail timeline. In this mode, each clip is represented by a single thumbnail, regardless of its length. The timeline can be moved by right dragging it. The graphical indicators over the thumbnails show if a clip is synced with audio (green/red bar), or if the clip has been graded (rainbow bar). Clips can be selected by Ctrl+click on a single thumbnail, or pressing Ctrl while dragging over a number of clips. Drag the upper edge of the rows of thumbnails to resize them. Click any thumbnails to jump to the middle of the corresponding clip. The current clip has an outline around its thumbnail on the thumbnail timeline.

A full screen version of this timeline is also available by pressing numpad 3. Pressing the Ctrl++ (plus) or Ctrl+- (minus) keys will resize the thumbnails larger or smaller. Click the mouse or press the ▲ (up arrow) or ▼ (down arrow) keys to move between clips. The selected clip can be played on the broadcast monitor. Click on any of the thumbnails and press the numpad 3 key again to return to the selected clip in the main display.

4.6. Timeline Navigation

4.6.1. Keyboard and Mouse Operations

Space

Start or Stop playback

or

Step to the previous or next frame

Shift+ or Shift+

Advance to the previous or next frame with audio playback

or

Step to the previous or next clip

J

Play backward; press repeatedly to speed up playback

K

Pause playback

L

Play forward; press repeatedly to speed up playback

Hold J + K

Play backward in slow motion

Hold K + L

Play forward in slow motion

Hold K and press J or L

Step by single frames backward or forward

Shift+G

Group timeline tracks (First, select tracks while holding Ctrl)

Alt+Shift+G

Ungroup timeline tracks (First, select tracks while holding Ctrl)

Hold left mouse button and drag

While holding the middle mouse button, drag to scrub picture.

Hold middle mouse button and drag

While holding the middle mouse button, drag to scrub picture rapidly.

Middle click

Click near the left side of the screen to jump to the beginning of the timeline, and click near the right side to jump to the end.

Hold Ctrl + right mouse button and drag

Drag the mouse to any direction to zoom in or out of the image around the cursor position.

Hold right mouse button and drag

Moves the image in the main viewer.

Double right click

Fills the screen and center the image.

Hold C key (while cursor is over the image)

Toggles between center/fit modes.

You can also group and ungroup clips if you go to Timeline  Group in the menu.

4.6.2. Jump to Timecode

To directly navigate to a specific timecode, simply press the decimal point on the numpad to turn on the timecode entry mode. Type in the timecode, and press Enter. To jump forward, type in the number with a + sign. For example, when typing in +1202, the video jumps forward by 1202 frames (e.g. jumping from zero to 00:00:12:02). To jump backwards, use a minus instead of a plus sign. To jump to a specific timecode, simply enter 12512 (corresponding to 00:01:25:12).

In the GUI tab of the Settings Page, you can change the Frame Entry Mode setting from Always TC to TC or Frames.

In the latter mode, you can enter either frames by adding a number, such as 123, or a timecode by adding 12.02 (that is 12 + numpad decimal point + 02, corresponding to 00:00:12:02).

4.6.3. Scaling Modes when Zoomed In

When zooming into an image, it is possible to control how to see the magnified area in the main viewer by selecting the different scaling/interpolation modes. There are two pixel interpolation modes:

  • nearest neighbor interpolation to see individual pixels

  • linear interpolation to see smooth images

You can zoom in and out by holding the Ctrl key and the right mouse button, and dragging the mouse to any direction over the image. The shortcut to toggle the linear interpolation mode is Alt+W.

Different scaling/interpolation modes when zoomed in

Nearest Neighbor Zoom, 250 Linear Zoom, 250

Linear interpolation can also be activated through the View  Zoom  Using Linear Interpolation menu option.

Transkoder Using linear interpolation
Transkoder - Using linear interpolation

The linear interpolation option is also available in the Settings Page:

  • Go to the Settings Page by pressing Tab and select the GUI section.

  • Open the Advanced settings at the bottom of the left menu, and roll down while holding the right mouse button to find the Image Linear Scale option.

  • Turn it on or off to toggle between the interpolation modes.

4.7. Edit Decision List Handling

4.7.1. Loading EDLs

Transkoder is capable of conforming basic EDL timelines, ones without dissolves and fades. First the EDL files need to be imported to the database, either by auto-upload from the "edl" folder under the date folder, or via the project web page. As soon as the EDL is in the database it will show up in the load page under the Cuts tab. EDLs that are uploaded but not yet loaded (e.g. Conformed) will be indicated as NEW. To load an EDL with handles, first it needs to be loaded normally, as NEW type of EDLs are not listed in the "load Cut w/ Handle" list.

When Transkoder conforms the timeline and some shots fail to match, warning message is displayed and printed into the console and the log file. One can investigate the missing shots based on the tapename and timecode information. Most often the EDLMatchTapeName / EDLMatchTimecode settings are not set properly, or the EDL contains shots that are not in the project database of the current project.

4.7.2. Export EDL

It is possible to export a basic EDL directly by using the File  Export Cut  EDL menu option. If there are grading nodes used on the timeline compatible with CDL, these will be exported as well. The tape name in the EDL will be generated based on the metadata specified in the Rendered Media Tape Name \ AleTapeName field specified on the Encode Settings section of the Settings Page. Please make sure to adjust the EDLTapeNameLength setting on the Metadata Settings section of the Settings Page if you want to truncate the tape name to be compliant with the old CMX 3600 format.

4.7.3. Conform Settings

To control the conform process please adjust the settings on the Conform section of the Settings Page:

  • EDLHandle: if an EDL is loaded with handle, Transkoder attempts to add extra frames to the head and tail of each shot. If the source does not contain enough frames (e.g. EDLHandle is set to 20 but there are only 6 extra frames) the handle will be shorter.

  • EDLMatchTapeName: This metadata will be matched to EDL TapeName. If set to Custom, then the EDLMatchCustomTapeName is used.

  • EDLMatchTimecode: EDL Timecode will be matched to this TC in the database

  • EDLMatchRelativeTC: to be used only if the TapeName uniquely identifies the shot, and the source TC in the EDL is a timecode relative to the first frame of the given clip. This often happens in some file based workflows where there is no time-of-day timecode available (e.g. DSLR or consumer cameras).

4.7.4. Removing EDLs from the Database

By deleting a conformed cut timeline on the load page, the EDL becomes new. This effectively undoes the conform step. To completely remove it from the database it needs to be deleted again. 'NEW' type of cuts will be completely removed from the database upon deleting them.

4.7.5. Cut Timeline with EDL

The Timeline  Cut Timeline with EDL menu option parses the EDL and executes a cut operation at the editing points defined by the record start/end timecodes in the EDL.

5. Image Processing Pipeline

5.1. Node Page

The node page contains the flexible node pipeline that allows to easily customize the image processing chain.

The left pane of the node page represents the source image buffers, while the right pane contains the result image buffers that can be displayed or encoded into any format.

Transkoder Node Page
Transkoder - Node Page

5.1.1. Processing Template Wizard

The flexible node pipeline allows you to fully customize your color/image processing pipeline for every camera format in your project. When loading any type of media for the first time, the system auto-creates a pipeline based on the project’s color correction settings.

By pressing Shift+O, the pipeline wizard will open, where users can create pipeline presets by specifying the source color space and the grading color space. With this wizard one can create a color pipeline that is guaranteed to have the proper color science.

The desired input color space, such as Log, Log C or RED defines the type of the source footage. Upon pressing Shift+O, this pop-up will auto adjust to match the current clip.

The grading color space selection, such as Colorfront Engine, Camera HDR, Mastering HDR, ACES, Log, Lin (or ARRI Look in case of using ARRI footage) defines how the colors are manipulated by the color correction node. See the Color Grading chapter for more information.

Transkoder Wizard to create a proper pipeline that can be later modified
Transkoder Wizard to create a proper pipeline that can be later modified

The node pipeline may be adjusted and new templates may be defined at any time, even after using the Processing Template Wizard.

5.1.2. Node Pipeline

By pressing Enter, the node pipeline will open. Every node represents a particular image processing operation. Each clip has its own node pipeline. Nodes left to the gray separator line are specific to the given clip, while nodes right to the separator line - including the results - are coming from a global template (often referred as the viewing template).

When the user loads a clip for the very first time (the clip is new on the load page), there are no clip-specific nodes saved to load. In this case, Transkoder loads the clip-specific nodes from the grade template. Each camera format may have its own grade template, which allows to deal with different color spaces and geometric properties. The current camera format can be inspected on the node page.

Transkoder - Node pipeline of Log C in a
Transkoder - Node pipeline of Log C in a "Lin" project, where the color correction is done in Rec709/Video gamma

To successfully use this valuable feature, one needs to understand the role of the Grade template and the Viewing template.

Looking at the node page, you can see the media source on the far left, and processed image format(s) on the right side. The node pipeline in the middle consists of two types of nodes:

  • nodes left to the separator line are saved per shot

  • nodes right to the separator are global nodes; they are saved per project

5.1.3. Grade Template

The Grade template is an XML file containing the default set of grading nodes for each camera format. Thus, if you load a new timeline, the grade template will be applied to the clips. However, from that point on, those nodes will be saved in the database and can be adjusted per clip. This also means that changing the grade template has no effect on shots already saved in the database.

Please note the red rectangle on the node page. Nodes inside this box are the creative grading nodes, which are saved per clip, as this red box is left to the separator line. Creative grading nodes are updated when the user drags and drops grades from one clip to the other on the timeline, or from the still store (look store) to the timeline. Typically, raw nodes (containing camera metadata needed to interpret the image in a technically correct way) are saved per clip, but should not be altered by applying different looks, so they are outside of the red box. To move a node in or out of the red box, press the V key while the node is highlighted, and then save the grade template (for details refer to section Altering Templates below).

5.1.4. Viewing Template

Global nodes are right to the separator line. These are typically image processing nodes after the color correction adding global burn-ins, applying color space transformations for viewing and rendering and resizing. The results on the right pane of the node page are also part of the viewing template. Nodes in the viewing template are global, which means that these nodes are always loaded regardless of the timeline’s type.

5.1.5. Alter Templates

In Transkoder, when loading any camera format for the first time, the Processing Pipeline Wizard window will pop up automatically. Creating the node pipeline with this window will generate proper node templates. Please note that for the second, third, etc. time, the pipeline window will be used to define the node structure. The global nodes will not be overwritten to prevent destroying used defined burn-in or LUT setups. However, if the user changes the color correction mode (e.g. from Log to ACES) in the pipeline wizard, the global nodes need to be overwritten. It is not recommended to do this during a running project.

Alter Grade Template Manually

Sometimes it is needed to change the node structure manually. However, it is strongly recommended to finalize the nodes in the grade template at the beginning of the project, and avoid changing it during a project. To add/remove a clip specific node from the left side, or move nodes in or out of the red box, and save the change in the grade template (so that any new footage of the corresponding camera format will load this setup in the future), please follow these steps:

  • Load a new timeline (or a timeline without any creative grade on it) of the camera format you need to adjust.

  • Execute the changes to the grading nodes.

  • Make sure all nodes are at their default setting (e.g. there is no color correction).

  • Press Shift+T to save the grade and the viewing template.

This will overwrite all the grading nodes on the timeline destroying color correction setups and setting them to their default values.
Alter Viewing Template Manually

If global nodes need to be adjusted (such as burn-in text or position, resize parameters), or new global nodes are needed (such as a new result or a new LUT), use the Shift+Y key combination to save the Viewing template. This only saves the global nodes without destroying any of the per-shot node settings. If you simply adjust any of the global nodes, but do not save it explicitly with Shift+Y, your changes will be lost upon next reload.

5.2. Edit Nodes

You can add new nodes to the node page if you move your mouse to the node area (dark gray area in the middle of the screen), and press the A key, or click with the right mouse button. The Tools list window will appear, and you can select the desired node from the list.

Tools list on the node page
Tools list on the node page

To attach a node to the pipeline, click and drag it onto the pipeline and press Ctrl. To remove a node from the pipeline, click and drag it away from the pipeline, and press the Ctrl key.

To select one or more nodes, hold down the Ctrl key and click on the nodes that may be selected. Nodes can be deleted by holding down the Ctrl key, and drawing a red line through the node with the left mouse button.

The Node Operations window for the selected node will appear by clicking on it with the right mouse. You can perform the following operations on the selected nodes:

  • Duplicate selected tools

  • Delete tools

  • Delete Tools On All Shots

  • Copy to Selected Shots

  • Copy to All Shots

  • Make Nodes Global

  • Make Nodes Clip Specific

  • Save Pipeline as XML

  • Load Pipeline from XML

Node Operations window on the node page
Node Operations window on the node page

By pressing the H key, the HUD (heads-up display) of the selected node will appear, where you can adjust the settings for the selected node. Pressing the G key repeatedly will make the HUD cycle through the controls for all the nodes on the pipeline. It is possible to view the HUD on the main screen as well as the node page. Double-clicking a node will navigate to the main viewer and the HUD of the selected node will be displayed.

5.2.1. Node Selector Floating Window

It is not necessary to enter the node page to select a node, or keep pressing the G button to cycle through nodes. By pressing Alt+G in the main viewer, a small Nodes selector floating window will open containing a list of nodes in the pipeline. You can select any node by simply clicking on it.

Nodes selector floating window to quickly access nodes
Nodes selector floating window to quickly access nodes

5.3. Edit Output Results

Output results of the pipeline can be created, selected and edited in the right-hand light gray area of the node page. If there are multiple outputs for the pipeline, you can scroll down to see them by right-clicking and dragging down in the result area.

Transkoder Node page with multiple results
Transkoder - Node page with multiple results

5.3.1. Add New Result

You can add a new output result to the node pipeline, if you move your mouse to the result area of the node page (light gray area on the right) and press the A key. Right-clicking in the result area will also add a new output to the pipeline.

5.3.2. Compare Results

Two output results can be compared with each other in the main viewer.

First, select the primary result by clicking on it with the left mouse button. The name of this primary result will be highlighted in white. Then, hold down the Ctrl key and select another - secondary - result by clicking on it for comparison. The name of this secondary output will be highlighted in yellow. (To remove the secondary result from the comparison mode, click on the primary - white - result again.)

After selecting two results for comparison, press Enter to navigate back to the main viewer. By pressing Shift+F4, you can toggle between the two outputs to compare them.

The Shift+F2 key combination will display the two output results in a blended view. To turn off the blended view, press Shift+F1.

Pressing Shift+F3 repeatedly will show the primary and secondary results in wiped, butterfly and side-by-side views alternately.

5.3.3. Delete Results

Results can be deleted by holding down the Ctrl key, and drawing a red line with the left mouse button, starting from the middle dark gray area through the result name.

5.3.4. Enable Result to Video Output

To enable any of the results to the video output, select the desired result by clicking on it while holding down the Shift key. The result’s name will be highlighted in cyan, while the small gray rectangle next the result’s name will be highlighted.

You can also enable the video output for any of the results by simply clicking on the small rectangle.

6. Color Grading

Transkoder has a flexible image processing pipeline, thus allowing different styles of color correction. When setting up a project, it is important to decide upon the desired grading color space, and CC (color corrector) tool(s).

6.1. Grading Color Space

When creating a project, it is important to choose the appropriate grading preset with the required color space. Transkoder auto-creates a corresponding color pipeline based on the color space of the media loaded after project creation. The image processing in Transkoder is calculated in 32-bit floating point precision per-channel and all pipelines are designed to avoid any unwanted clipping or quantizing.

Specifying the grading color space upon creating a new project
Specifying the grading color space upon creating a new project

Whichever setting is used to create a project, Transkoder allows the user to adjust or redefine the color pipeline for every media format independently. The options for color grading are the following:

Colorfront Engine

This is a custom Colorfront designed pipepine. Color grading is executed in a log color space (CF Log) that is ideal for this purpose. The tone mapping from this log color space to Rec.709 video (or any other display color space) is executed through a LookBlend node, where the operator/DIT/DOP may blend different tone mapping curves.

The Colorfront Engine color pipeline auto-created for Log C media
The Colorfront Engine color pipeline auto-created for Log C media
Camera HDR

This pipeline is identical to the Colorfront Engine pipeline, but the default output is set to Rec.2020 PQ HDR.

Mastering HDR

This pipeline adds only a single HDRSDRRemap node to use for any brightness remapping.

ACES

This is the color pipeline recommended by the Academy ("Academy Color Encoding System"). Media is graded in a log color space (ACES Proxy) and then - using a synthetic film-like tone mapping curve - converted to video Rec.709.

ARRI Look

This pipeline is fully compatible with the color processing of the Alexa. If round-tripping of look files is a requirement this pipeline is recommended.

ARRI Amira

…​

ARRI ALF-2

ARRI Look File v2, or ALF-2 is supported both with ProRes and ARRI RAW .ari files. This color pipeline contains a CDL color correction in the camera native Log C color space, and in the QuickTime ProRes a custom Lookup Table to map to Rec.709 video color space and optionally a wide gamut color space conversion as well, and in ARRI RAW a single Lookup Table.

Creating a new project with ALF-2 color pipeline
Creating a new project with ALF-2 color pipeline

Embedded look information stored in the file header is automatically applied to the clips. When using an ALF-2 project, the node pipeline contains a CDL node in the camera native Log C color space, and an ALF-2 node to apply the Log C to Rec.709 / Wide gamut output LUT. If there is a wide gamut LUT present, it can be activated on a per-clip basis using the Active LUT selector on the ALF-2 node.

ALF-2 color pipeline for ProRes source.
ALF-2 color pipeline for ProRes source.
Lin

In this mode the color correction is done in "video" (Rec.709/Video Gamma) color space. The auto-created pipeline converts all media to Rec.709 and the grading node is at the end of the chain. It is recommended to use the Color Decision List or the RGBLin grading nodes.

Log

Color correction is applied to the Log images, even if the input media is not log (e.g. GoPro). This guarantees a unified workflow, so color corrections copied from one shot to the other have the same effect. All media is converted to Rec.709 at the end of the chain.

Native

In this mode, Transkoder does not apply any color space conversion by default. All media is treated in its native color space. LUT nodes may be used to apply any custom lookup table.

The auto-created pipelines include a color correction tool, but it is possible to change them or use a combination of color correction nodes. For details, see the Tools on the Node Page and the Grading Tools chapter.

Customizing the color grading toolset
Customizing the color grading toolset

6.2. Using the Color Grading Tools

Most often color correction is executed on some control panel, such as the Tangent Element panel set.

The mouse and keyboard may also be used for grading, in connection with the Heads Up Display for the tool: press the h key, or swipe left of the screen to open the HUD.

The g key cycles through the HUD tabs.

Use the number keys 1 to 9 to select the specific parameter or parameter set to change, and drag the mouse across the screen. Press the 0 (or the corresponding number again) to return to navigation mode. This allows fine master and balance adjustments while focusing on the image. To visualize the current color correction using the Color Circle tool, press numpad 6.

6.2.1. Color Grading with the HUD

One of the common grading tools in Transkoder is the CDL (Color Decision List) node. Please refer to the section titled Grading Tools for details on a variety of other grading tools supported by Transkoder. Open the HUD by pressing the h key.

HUD Color Tab - Color Grading with the CDL Node Tool
HUD Color Tab - Color Grading with the CDL Node Tool

There are individual controls for each RGB color channel and a master control which moves all three color channels simultaneously.

The Power sliders work in reverse. Gain increases towards the left.

An overall saturation control is at the bottom.

Drag a slider or double click on the numeric value to the right of the slider to manually enter a value.

Ctrl-click a slider to reset it.

At the bottom are the following buttons:

Button Keyboard Shortcut

Undo

Page Down

Redo

Page Up

Bypass

B

Reset

R

Close the HUD by pressing the H key again.

Using the Color Circle Tool

To open the Color Circle tool, press the numpad 6 key.

Color Circles Tool
Color Circle Tool
  • To adjust the tint, use the mouse to move the dots in the center of the color circles.

    • Move the dot towards the tint you wish to add, or away from the tint you wish to remove.

    • The further from center the dot is, the more of the color tint is added.

  • To adjust the gain, move the sliders on the right edge.

  • Click and drag any of the numeric fields to adjust individual values.

Color Grading with the Keyboard and Mouse

The easiest way to grade without using a dedicated grading panel is to use the HUD with the number keys on the main keyboard.

  • To select the master Slope slider, press the 1 key.

  • To select the three Slope RGB sliders, press the 2 key.

  • To select the master Power slider, press the 3 key.

  • To select the three Power RGB sliders, press the 4 key.

  • To select the master Offset slider, press the 5 key.

  • To select the three Offset RGB sliders, press the 6 key.

  • To select the Saturation slider, press the 7 key.

  • To adjust the selected sliders, click and drag over the entire area of the main display.

  • To adjust the gain, make sure that the master slider is selected, drag towards the left or right.

  • To adjust the tint, make sure that one of the RGB sliders is selected and drag in the same direction as the Color Circles.

  • To return to normal, press the 0 key or the previously selected number again.

Using the Sector Grading tool

The Sector Grading tool allows for brightness and hue control over very narrow color bands.

HUD Color2 Tab - Color Sector
HUD Color2 Tab - Color Sector

When the Sector Grading tool is active, the numpad 6 Color Circle tool becomes the Color Sector Wheel.

Color Sector Wheel
Color Sector Wheel
  • To open the Sector grading tool, click on the HUD Color tab a second time..

  • To change the brightness of a color band, adjust the sliders on the right.

  • To change the hue and saturation, click and drag the individual color band pucks.

    • To change the hue, drag a puck towards the left or right.

    • To reduce the saturation, drag a puck towards the center.

The sector color corrections are not represented in the ASC CDL values that are passed into an AVID ALE.

6.2.2. Look Stores (Still Stores)

Color grading may be saved in the Look Stores (numpad 2), that stores both the color parameters and the still image of what the currently graded frame is. There are direct shortcuts on the control panels as well for saving and recalling to/from the Look Store. The mouse may be also used to drag a color grade from the timeline to the Look Stores, or back. There are currently three different Look Stores, one may toggle between them by clicking on the colored rectangles on the left edge of the Look Store:

  • Green - Global Look Store: Whatever is saved in this look store is available from all timelines.

  • Blue - Daily Look Store: This option is to be used to store looks of a specific day in a dailies process.

  • Red - Import Look Store: Look files - such as CCCs, CCs, CDLs, CDL-EDLs, reference still images, Colorfront EMDs or LUTs - from the looks folder are imported into this stillstore.

The Bin window ALT+j may also be used as a look store. If the timeline is in grading mode, simply drag a clip from the bin to the timeline to apply color correction.

Please refer to the section titled Look Store Shortcuts for helpful keyboard Shortcuts when dealing with a large number of looks (stills).

6.2.3. Look Presets in the LookEditorWindow

Transkoder supports a look-based workflow where named looks may be saved and recalled by the function keys F1 to F12. To open the Look Preset window, press SHIFT+i.

Look Presets in the Look Editor Window
Look Presets in the Look Editor Window

The current grade can be stored in a Look slot by selecting the Look number button and clicking the Save button. Saved looks can be named by clicking in the label field, entering the desired name, and pressing the ENTER key. The name will be saved in the database. If you press the e key, you may also see the look name in the upper left corner of the screen. To rename a look, click on the look name and enter a new name. Press the ENTER key when finished.

Grades can be applied to the current clip by selecting the Look number and clicking the Load button or pressing the appropriate F1 to F12 function key. This will to apply Look1 - F1, Look2 - F2 and so on.

To use even more predefined looks, save them by pressing CTRL+1 to CTRL+12, and recall them by pressing SHIFT+1 to SHIFT+12.

Automatic Application of Looks Based on Metadata Name

It is possible to auto-apply looks saved from the Look Editor Window to clips based on metadata names. To set this up:

  • make sure the grade / LookReferenceMetadataName setting is set to the metadata name you wish to drive the automatic application (this could be any metadata name available for the shot, like Scene, Take, Slate, ClipName, etc.)

  • open the Look Presets window and save the looks such that the look names match the corresponding metadata driving the automatic application

After this has been set up the operator can load any timeline and choose the Grade / Apply Looks to All Shots menu item to apply the looks based on the metadata field set up in the tab page.

6.2.4. Global Looks in the HUD Presets Panel

Storing Global Looks in the HUD Presets Panel
  1. Open the grading HUD

  2. Click the PRESET button at the bottom of the HUD to open the Presets panel

  3. Click the PLUS key or the PLUS SIGN to store the current look in to the presets.

    Hold the OPTION key and click on the preset name to edit the preset name. Hold the CONTROL key and click on a preset’s name to delete it. Simply click on a preset name to apply it to the current clip.

Since HUD presets are Global, they will appear in all projects.

6.2.5. Using Dual Playheads for Comparison

Transkoder has dual playheads that allow for two different locations on the timeline to easily be compared.

  • To view the first playhead, press the LEFT SQUARE BRACKET ([) key.

    • The corresponding thumbnail will show a yellow outline.

  • To view the second playhead, press the RIGHT SQUARE BRACKET (]) key.

    • The corresponding thumbnail will show a blue outline.

  • To toggle between the two playheads, press the BACKSLASH (\) key.

  • To split the screen between the two playheads, press SHIFT+F6.

  • To rotate the split line, hold the SHIFT key and drag the mouse towards the left or right.

  • To change the position of the split, drag the blue dot.

  • To navigate that side of the split, select either playhead.

Dual Playheads Split
Dual Playheads Split

Transkoder may display a single color channel.

  • To show the red channel only, press SHIFT+F9.

  • To show the green channel only, press SHIFT+F10.

  • To show the blue channel only, press SHIFT+F11.

  • To show the luminance only, press SHIFT+F12.

  • To show the normal color display again, press SHIFT+F1.

Save the current clip still as a 1920x1080 JPEG to…​ /Library/Application Support/TKD/stills/, by pressing CONTROL+SHIFT+s

6.2.6. Using the Analyzer

  • To open or close the built in floating scopes, collectively known as the Analyzer, press CONTROL+h.

  • To select between waveform, parade, vectorscope or histogram displays, with or without colors, use the Window  Analyzer Mode menu entry.

Analyzer and Vectorscope
Analyzer and Vectorscope
  • To resize the Analyzer, drag the lower right corner.

The Scopes display the full range internal RGB color space.

Analyzer in RGB mode
Analyzer in RGB mode

6.3. Animation of Keyframes

The animation or automation system of Transkoder is based on segments separated by transition areas. Parking on any frame within such segment and changing the node parameters will affect all frames within that range.

The beginning of the animation transition range is marked by moving the playhead to the desired position and pressing a. The end of the animation transition range is marked by moving the playhead to desired position and pressing a again. This will divide the shot into three parts: before the animation transition range, the animation transition range, and after the animation transition range. The animation itself is configured by specifying the desired color science settings for both before the animation transition range, and after the end of the animation transition range. The animation transition range will interpolate between the two settings. It is possible to create multiple animation transition ranges for each shot, by repeating the procedure. Each animation transition range will be animated as specified.

Animation of Keyframes Timeline View
Animation of Keyframes Timeline View
Reverting to the original grade.
  1. Press the a key to set a keyrame at that point before changing the grade.

  2. Jump the playhead to the previous keyframe by pressing the COMMA key or next keyframe by pressing the PERIOD key.

  3. Remove all keyframes within the current clip, and keep the current grade at the playhead position by pressing SHIFT+a.

Animation in Transkoder is done on a per node or node group level and not in a channel specific way. Key frames are chosen from the currently active node. The Animation Keyframes field in the Advanced Settings section of the Grade Settings section of the Settings Page, can be used to specify how the keyframes are created. The available options are Individual, Group, and All. For the Individual option, only the currently selected node settings will be animated using the animation transition range. For the Group option, the settings of nodes of the similar type as the currently selected node will be animated, such as color correction. For the All option, the settings, of all the nodes will be animated.

The settings can not be adjusted inside the animation transition range.

6.4. Grading Navigation and Shortcuts

There are several keyboard shortcuts which help navigation and grading.

6.4.1. Grading Timelines

Grading Timelines Shortcuts
Shortcut Function Note

numpad 1

Thumbnail timeline

Each shot is represented by a small thumbnail image, regardless of the shot length.

numpad 2

Look Store timeline

Please refer to the Look Stores section.

numpad 3

Big thumbnail timeline

You may scroll the timeline with a right-click drag, or "shuttle" by holding the SHIFT key and right-click dragging to left or right to move the timeline backward or forward.

6.4.2. Applying / Reapply Grades

The user interface of Transkoder has been designed to allow the operator to import, apply or copy grades very easily. Color grades may be dragged from one clip to another, by using the mouse. They may be dragged from the Lookstore to the Timeline or from between two timeline clips. Also multiple clips may be selected, by holding the CTRL key and clicking on individual thumbnails or swiping over them. This way, the selected clips will be highlighted, and when a color grade is dragged to one of the clips, it will be applied to all of the selected clips. Some of these operations are also mapped to the MF Element panel. Clips on the timeline with grades will display a rainbow stripe at the top of the thumbnail.

Grade Management - Thumbnail Drag
Grade Management - Thumbnail Drag
  • To copy a grade from one clip to another, drag and drop one thumbnail on top of another.

  • To select or deselect a thumbnail, CONTROL-click on it.

  • To select or deselect the current clip, press the n key.

  • To select or deselect all the thumbnails in the current timeline, press CONTROL+a.

  • To select or deselect the current clip and all clips to the left, press SHIFT+LEFT SQUARE BRACKET.

  • To select or deselect the current clip and all clips to the right, press SHIFT+RIGHT SQUARE BRACKET.

  • To clear all selections, press CONTROL+a two times.

Selecting multiple thumbnails
Selecting multiple thumbnails

Any thumbnail or look dragged onto a group of selected clips will update them all.

Transkoder will display yellow dividing lines between groups of thumbnails that share the same Scene number. This makes it easy to grade the first clip for a given Scene, select the remaining members and copy the grade.

The operator may also build a series of grade versions, or a "stack" of grades which is very easy to recall at any time. To update the thumbnail and add the current grade as a new version, press the END key, s key, or FUNCTION+RIGHT ARROW. To open the grade version stack, click the current thumbnail. When exiting the timeline the latest current version of the grade is always added to this stack.

Grade Version Stack
Grade Version Stack

Further options when working with the grade stack:

  • To apply the grade to the main image, click on any thumbnail grade in the stack.

  • To close the thumbnail grade stack, click the thumbnail again.

  • To select the previous grade in the stack, press the DELETE key.

  • To select the most recent grade, press the HOME key or the x key.

A by-product of this feature is the playhead position for this clip is remembered only so long as the playhead is not moved again within this clip. This allows you switch to the Big Timeline and click on the thumbnail to jump to that frame. In the Big Timeline you may play anywhere within the clip and still click to return to the poster frame. This feature may be used to mark slates and sync audio as previously mentioned.

6.4.3. Grading Quick Memory Slots

There are also 3 quick memory slots. Save grades by pressing numpad 0+numpad 7, numpad 0+numpad 8, or numpad 0+numpad 9 into slots 1, 2, or 3 respectively. Recall grades by simply pressing numpad 7, numpad 8, or numpad 9.

6.4.4. Look Store Shortcuts

There are a number of useful shortcuts to work with the look stores (still stores). These are essential when working with hundreds of frames of looks:

Look Stores Shortcuts
Shortcut Function

numpad 2

Open/Close the Look Store Timeline

numpad 4

Full screen look store

Hold numpad 5

Recall the look store look for visual comparison

CTRL+BACKSPACE

Delete look (mouse pointer should be over the look store)

ALT+UP ARROW

Jump to next look

ALT+DOWN ARROW

Jump to previous look

ALT+LEFT ARROW

Move look store left by one frame

ALT+RIGHT ARROW

Move look store right by one frame

ALT and Middle-Click drag

Direct (fast) move of the look store

6.5. Importing Looks, LUTs, Grades, and EMDs

External looks, Look Up Tables, grades, and EMDs, created with either third party or Colorfront products, may be easily imported into Transkoder by browsing to the looks folder using the Grade  Import Looks Folder menu option for looks and LUTs, and browsing to the emd folder using the File  Import Looks  From EMD Folder menu option for EMDs.

The list of supported look files are the following:

  • CDL’s (individual ASC CDL grades in XML format)

  • CCC’s and CC’s (Color Correction Collection - multiple ASC CDL grades in one file)

  • CDL EDL (EDL with ASC CDL grades included and will automatically apply to clips with correct timecode)

  • ARRI Looks (in XML format)

  • DPX, JPG and TIFF stills.

In addition to the above, varying LUT formats and Colorfront EMD files are also supported by Transkoder for grading.

6.5.1. Loading 3rd Party Look-Up Tables

All of the most popular LUT formats are supported for import, with flexible LUT sizes. The only limitation of LUT data size is that they should fit into gpu memory beside all the other processing data, so multiple gigabytes of different LUTs will not work, but 100 MB will.

Currently, the following Look-Up Table formats are supported by Transkoder:

  • Luther ASCII 17x17x17 (.txt)

  • Autodesk Lustre 17x17x17 LUT format (.3dl) 8 Colorfront’s own 3dmesh LUT format (.3dmesh)

  • Cube 3D LUT format (.cube) 8 1dlut (limited to 1024 levels) and 1dcurve formats

LUT files must be placed in lut directory under the install home. In Transkoder, LUT node must be added to your project pipeline, press h for displaying the HUD, then select the desired LUT. By adding LUT node left to the grey line, it may vary on a shot-by-shot basis.

Please do not copy CDL, CCC, CC, CDL-EDL, reference stills or Arri Looks files into the luts folder, it is only for Lookup Tables. To use these files please refer to the next sections.

6.5.2. Importing ASC Color Decision List

The American Society of Cinematographers Color Decision List (ASC CDL) is a format for the exchange of basic primary color grading information between equipment and software from different manufacturers. Transkoder may import Color Decision Lists coming from other grading systems. The Slope, Offset and Power parameters, as well as Saturation are all supported, as long as there is a CDL or an RGBLin node in the pipeline.

Loading these looks, the CDL/CCC/CC files, or CDL-EDL (EDL files with embedded CDL information) in Transkoder is done by browsing to the looks folder using the Grade  Import Looks Folder menu option.

If a reference still of matching name is also there, the system will show the looks with the correspinding still. If there is no still, Transkoder will use whatever reference image it finds.

An CDL-EDL will auto apply the grades if the clip timecode matches the EDL and value of the Apply Looks field is set to enabled in the Grade Settings section of the Settings Page.

CMX EDL with inline CDL parameters sample:
TITLE:   0214  VFX TEST 01 EDL
FCM: NON-DROP FRAME
001	LR229	V	C	09:00:00:00 09:16:53:05 01:00:00:00 01:16:53:05
*ASC_SOP ( 1.19 1.20 1.38 ) (2.19 2.22 2.38 ) (3.19 3.22 3.38 )
*ASC_SAT 4.41
002	LR230	V	C	10:00:00:00 10:21:30:06 01:16:53:05 01:38:23:11
*ASC_SOP ( 5.16 5.22 5.38 ) (6.19 6.22 6.38 ) (7.16 7.22 7.38 )
*ASC_SAT 8.41
The ASC-CCC format contains multiple CDL XML chunks that Transkoder imports.
<ColorDecisionList xmlns='urn:ASC:CDL:v1.01'>
    <ColorDecision>
        <ColorCorrection id="test_cc.101">
            <SOPNode>
                <Slope>1.04 1.21 1.2</Slope>
                <Offset>0.017 -0.0456 0.045</Offset>
                <Power>0.987 1.03 0.974</Power>
            </SOPNode>
            <SatNode>
                <Saturation>1</Saturation>
            </SatNode>
        </ColorCorrection>
        <ColorCorrection id="test_cc.102">
            <SOPNode>
                <Slope>0.987 1.03 0.974</Slope>
                <Offset>1.04 1.21 1.2</Offset>
                <Power>0.017 -0.0456 0.045</Power>
            </SOPNode>
            <SatNode>
                <Saturation>1</Saturation>
            </SatNode>
        </ColorCorrection>
    </ColorDecision>
</ColorDecisionList>

6.5.3. Importing Arri Look Files

From ARRI ALEXA SUP 4.0, Look XML File support is available on ARRI ALEXA, which means that look XML files may be uploaded to the camera. These looks may be applied in the camera either to the monitoring output or the recorded media. If looks are not baked into the recorded files but only used for monitoring, Transkoder may apply the very same color transforms to the recorded media based on metadata travelling in the files or based on look XML files.

Loading these xml look files together with an optional matching still (DPX, TIFF, JPG, or PNG with matching file name), in Transkoder is done by browsing to the looks folder using the Grade  Import Looks Folder menu option. If the project is set up in the ARRI Look Color Space mode, and if the media contains Arri look metadata, it will be automatically applied to the HUD CDL values and the Printer Lights and Saturaton settings.

ARRI Looks - Metadata Label
ARRI Looks - Metadata Label

To auto-apply Arri Look to ProRes or raw media from the header, no extra look files are necessary. The complete look is stored in the file headers as well.

6.5.4. Importing Reference Stills

Loading these 10 bit uncompressed DPX image files, TIFF reference image files, or JPG/PNG look files in Transkoder is done by browsing to the looks folder using the Grade  Import Looks Folder menu option.

When Transkoder loads any imported still image or look file, it applies the color pipeline that is defined in the project. Because different camera formats may have different pipelines, Transkoder uses the one created first. This means that any color or geometric correction used for camera original processing saved as template, will apply to the stills as well. If this pipeline is not suitable for the stills, it is possible to change this behaviour by following the below steps:

Creating a new color pipeline for stills.
  1. Create a new temporary card (folder) in the <ProjectBasePath>/<TransferDate>/<Format>/my_stills/ format, such as projects/my_project/20171002/my_still_format/my_stills/ in any date folder in the project.

  2. Copy some ProRes media, or any media of your choice that has the same color and framing as the reference stills to this card.

  3. Launch Transkoder and load this card so that Transkoder may detect a it as a new camera format (my_still_format) and create a new pipeline.

  4. Set up the parameters of the pipeline, making sure to also save the template with SHIFT+t.

  5. Exit to the load page.

  6. Set the value of the Pipeline for Imported Looks field under the Advanced Settings section of the Base Settings section of the Settings Page to my_still_format.

After this all stills should load with the pipeline you have set up.

6.6. Exporting Looks and Grades

  • Exporting Look-Up Tables: Look-Up Tables are easily exportable by adding a LUT renderer in the encode (p) page. Just like any other encoder, the name of the generated LUT files may be customized from metadata.

  • Exporting CDLs: Transkoder may export individual CDLs for every clip by adding the corresponding metadata deliverable.

  • Exporting CDL-EDLs: The ASC-CDL parameters are passed to AVID MediaComposer and are available there as metadata, and may be later exported from MediaComposer in a CMX-EDL for finishing, allowing for a full CDL pipeline using Transkoder through Dailies and Editorial to final grading. To embed the CDL information in the ALE file enable "AleCDL" parameter on the Settings/Encode tab. ALE files are auto-generated when rendering Avid DNxHD deliverables, and may also be rendered separately by adding the corresponding metadata encoder.

  • Exporting Arri Look files: Transkoder may export individual look xmls for every clip by adding the corresponding metadata deliverable. The Arri Alexa round tripping works correctly if Arri Look pipeline is used.

  • Exporting Reference Stills: If you save your look to the Blue - Daily Look Store, then it is automatically exported as a JPEG image with both grade and viewing LUT applied as well as a DPX image with color grading applied only under <ProjectBasePath>\thumbcache folder.

6.7. EMD (Express Metadata)

Colorfront’s EMD file is a proprietary metadata file that describes everything that Transkoder knows about a clip and all the work that has been done to it.

6.7.1. EMD Workflows

EMD (Express MetaData) files are clip specific sidecar files using XML encoding to track media metadata as well as grading information. All metadata associated with the clip (e.g. scene,take,card name, timecode), original file metadata (as visible in the SHIFT+u window) as well as color grading and node setup data. The EMD file does not contain global nodes, such as burn-in node setups and grade version history.

There are a number of ways EMDs may be used, such as:

  • Exporting looks from On-Set Live to Transkoder.

  • Exchanging Color and metadata in a Colorfront product environment.

  • Using EMDs through the DMS XML format.

The EMD files are the same, the workflow is slightly different.

Exporting looks from On-Set Live to Transkoder

As On-Set Live operates on a live video feed which may or may not have rich metadata, it is not possible to generate EMDs for recorded clips with matching file names. Instead EMDs exported from Live contain a timecode, which may be used to auto-apply these to the recorded clips in Transkoder or other Colorfront products. Colorfront products which support importing EMD based looks from On-Set Live, are Express Dailies, On-Set Dailies, and Transkoder.

For a dailies workflow in Transkoder, the EMD files exported from On-set Live need to be placed in the looks folder under the current transfer date folder. These are automatically parsed upon loading any card from that specific date. If Auto Apply Looks option is enabled the looks from the EMDs are auto-applied to the clips based on timecode matching. The EMD timecode needs to be earlier than the (source) end timecode of the clip. If Auto Apply Looks is not enabled or timecode based matching is not successful these looks are also listed in the Red - Import Look store (numpad 2) from where the user may manually apply them by dragging to the clips.

By selecting the Grade  Import Looks Folder menu option, any folder may be used to populate the Red - Import Look store.

EMDs may be manually applied to the shots by dragging and dropping them.

Color and Metadata Exchange in a Colorfront Product Environment

A robust workflow exists for taking metadata (audio sync, scene, take, QC comments) and grading info from one Colorfront product based system to an other. By selecting the File:  Export EMD menu option, an emd folder is created, (..\projects\<project name>\<TransferDate>\emd) and the files for all the clips are exported there. The file name of the EMD will match the file name of the source video clips (corresponding to the ClipName metadata in the database). By copying these emd files to the emd folder of another station and loading a New card into the project, the data from the EMD will auto apply. The matching is based on file name, so make sure not to rename the EMDs.

Using EMDs through the DMS XML format

DMS files are simple XML files defining a timeline with multiple video events and their corresponding color grade by means of EMD files. Below is a sample DMS XML illustrating how clips may be loaded into Transkoder with grades defined in an EMD file:

DMS XML sample:
<project name="mytest">
  <timebase tb="24"/>
  <timecode tc="NDF"/>
  <width w="1920"/>
  <height h="1080"/>
  <pipeline type="ACES"/>
  <clip>
    <name LabRoll="testroll" VTRoll="vt test"/>
    <emd file="c:/emd/sony.emd"/>
    <clipname file="C:/projects/multicam/20120303/sonyF55_raw/A001C004_120104DZ.mxf"/>
    <timecode in="00:00:00:00" out="00:00:00:00"/>
  </clip>
  <clip>
    <emd file="c:/emd/red.emd"/>
    <clipname file="C:/projects/multicam//20120303/EPIC_SAMPLE/A001_L010_0909HR.RDC/A001_L010_0909HR_001.R3D"/>
  </clip>
  <clip>
    <emd file="c:/emd/prores.emd"/>
    <clipname file="C:/projects/mixed/20131313/mixed/B067/a382c001_110727_r24j_002.mov"/>
    <timecode in="11:19:31:17" out="11:19:31:17" dur="00:00:20:00"/>
  </clip>
  <clip>
    <clipname file="C:/projects/multicam/20120305/arriraw/ArriRaw/s001/1-1_A001.%07d.ari" start="5" end="24"/>
    <timecode dur="00:00:20:00"/>
  </clip>
</project>

6.7.2. Exporting EMD

Exporting EMD files
  1. Select the File  Export Looks  From Timeline as EMD menu option to export individual EMD files for every clip in the current timeline.

    EMD files are exported to the emd sub-folder inside of the TransferDate folder.

Archiving EMD files
  1. Archive or rename the emd sub-folder inside of the TransferDate folder to backup your work at any time.

    This is easier than backing up the database, and much easier to restore.

EMD files contain the grade, clip metadata, zoom settings, and audio sync. The EMD files do not contain global settings like burn-ins and preferences.

6.7.3. Importing EMD

Importing EMD files
  1. Copy the EMD files into the emd sub-folder within the TransferDate folder.

  2. Open the All Takes timeline.

  3. Select the File  Import Looks  From EMD Folder menu option.

EMD files are an excellent way to move work between two workstations. Two workstations could share the same storage. One workstation could sync and grade, and then export the EMD files. The other workstation could load the same media and import the EMD files, and then add burn-ins and render.

EMD files and still images are loaded into the Red - Import Look store when importing looks from On-Set Live, Colorfront’s live grading application. Make sure to place EMD files and .sdi stills into the emd sub-folder of the TransferDate folder.

7. Tools on the Node Page

Transkoder has many tools which can be connected in a flexible way on the Node pipeline. For an indepth description of the purpose and functionality of specific nodes, please refer to the List of Tools section of this chapter.

7.1. Introduction to Tools

The Node Page is accessed by pressing the Enter key. The left-hand side will show the Source image, the right-hand side will show one or more Result renders. The middle section will show a flexibly configurable node pipeline that graphically represents the image processing workflow. Each node is an image processing tool that can be configured to use predetermined defaults or per-shot settings.

7.1.1. Controls on the HUD

The Heads-Up Display, i.e. HUD, is used to adjust settings for the various tools/nodes used by Transkoder. In the node page, selecting a node by a single click and pressing the H key will bring up the corresponding HUD. Press H again to make the HUD disappear. Pressing the G key several times will make the HUD cycle through the controls for all the nodes on the pipeline. It is possible to view the HUD on the main screen as well as the node page. Double-clicking a node will navigate to the main viewer, and the HUD of the selected node will be displayed.

Transkoder Bin Window - HUD on the Node Page
Transkoder - HUD on the Node Page

By dragging the sliders, the user can adjust the controls individually. By clicking on any slider, while holding the Ctrl key, resets it to the default values. To set the value to a specific number, please double-click anywhere on the slider. A text input field will come up, where the operator can enter any specific number.

For more precise control of the sliders, you can attach the mouse pointer to them and adjust the values by dragging the mouse over the screen. For this, press the 1, 2, 3, etc. number buttons, and the connected sliders will be highlighted. To adjust the selected sliders, click and drag the mouse anywhere on the screen. If two or three sliders are highlighted, it is possible to "balance" the components without changing the overall gain.

7.1.2. Tool Presets

On the bottom of the HUD window, the tool specfic presets can be accessed by clicking on the PRESET title. This function is global to all projects of the current database server. Tool presets are named set of parameters that the user can easily save and recall. Typical use cases include frequently used burn-in presets, format specific framing settings, or even color grades.

To save current parameters as a preset, press the small ⊕ (plus sign) button. To rename the preset, click on its title while holding the Alt button.

To recall presets, simply hold the left mouse button, and hover over the presets to apply them in real time. This way, it is very easy to rapidly find the right preset.

7.1.3. Undo / Redo / Reset / Bypass Operations

The Undo and Redo operations accessed from the main HUD are not specific to the node, but reverse or start over an action in the grading history in general. This means, pressing Undo may also reverse the last change made in some other node.

The Reset button, however, is specific to the current node. Click the Reset button to revert all parameters to their default state, with a few exceptions:

  • Camera raw nodes (such as Red, SonyRaw, ArriRaw) will default to the camera metadata

  • HDRSDRRemap node will reset the sliders, but will not change the color space settings

  • DolbyVision node will reset the creative trimming sliders, but not reset or remove the L1 level statistics

The Bypass operation disables the effect of the node. The user can validate if the functionality is working correctly by clicking on this option.

7.2. List of Tools

To provide a rough picture about the tools available in Transkoder, this chapter contains a list of tools with a short description about their functionality and purpose. For detailed information, see the next chapter, where you can find the full description of each tool and their controls.

7.2.1. List of Color Space Tools

For details about the functionality and controls of the following tools, see Color Space Tools chapter, or click the name of the certain tool to see its detailed description.

Transkoder has been developed so that all these tools use the full RGB range internally.
Tool Name Tool Description

ARRILook

Part of the ARRILook pipeline and maps Log C color to Rec.709.

ARRIRAW

Required for ARRIRAW debayer pipelines and it controls the debayer metadata.

Amira

Required for the pre-ALF2 color pipeline and maps Log C color to Rec 709. using the 3dmesh LUT metdata contained in the Amira source assets.

ColorConv

A "Swiss Army knife" color conversion tool used to map from any color space representation to any other color space representation.

DisplayLookHDR

The final node of the Colorfront Engine pipeline which maps from the Colorfront internal color space to the viewing color space. This tool supports both PQ (Perceptual Quantizier) and HLG (Hybrid Log-Gamma) encoding. HLG mode implements the variable HLG transfer function as per the recent standard with advances handling of out-of-gamut colors.

SonyRAW

…​

HDR IDT

This is an Input Device Transform, at the begining of the ACES workflow pipeline, which maps source HDR assets into the ACES color space and encoding specifications.

HDRSDRRemap

This a proprietary Colorfront tool, which maps from HDR to SDR, and the other way around. It uses custom algorithms based on human perception to flexibly do the mapping without harming the color, tone or saturation of the input while preserving the original creative intent. This tool supports both PQ (Perceptual Quantizier) and HLG (Hybrid Log-Gamma) encoding. HLG mode implements the variable HLG transfer function as per the recent standard with advances handling of out-of-gamut colors.

HDR10plus

HDR10plus is an open standard for dynamic HDR metadata. In contrast to static HDR10, this standard provides a per-scene dynamic tone mapping curve, so TV sets are able to apply a proper remapping of bright pixels that would clip without adjustment. Transkoder on the Windows platform can generate this metadata and also apply the remap itself to the image.

LUT

This is needed for applying LUTs. They must first be made available to Transkoder according to the procedure in the section titled Loading 3rd Party Look-Up Tables in the Grading Chapter.

Matrix

This is a 3x3 matrix multiplication of the RGB values method used to specify a custom color mapping.

LUTACES

This tool is used to apply ACES compliant LUTs. They must first be made available to Transkoder according to the procedure in the section titled Loading 3rd Party Look-Up Tables in the Grading Chapter.

ODT

This is an Output Device Transform at the end of the ACES workflow pipeline, which maps from the Reference Rendering Transform display-referred result to the viewing color space used on a given output device.

ODTInv

If the viewing color space of a given external device was mapped to using ODT, it can be inverted back to the Reference Rendering Transform display-referred color space, used in the ACES workflow pipeline, by using this tool.

RED

Required for RED debayer pipelines and it controls the debayer metadata.

RRT

This the part of the ACES workflow pipeline used to map from the scene-referred ACES color space to a display-referred color space.

RRTInv

This tool is used to invert the display-referred color space back into the scene-referred ACES color space.

VideoRange

This is a flexible range conversion tool allowing the conversion between legal range (aka video range) and full range, or to any custom range. If loading legal range source assets, which are not detectable by Transkoder, then the VideoRange tool should be used as the first node after the source to convert to the full range used by all the tools internally. If, however, a legal range resulting render is desired, then the VideoRange tool should be used as the last node before the result render to convert from the full range to the legal range.

7.2.2. List of Grading Tools

For details about the functionality and controls of the following tools, see Grading Tools chapter, or click the name of the certain tool to see its detailed description.

Tool Name Tool Description

CDL

This is the American Society of Cinematographers Color Decision List which is used to exchange color grading information.

CFPrimary

This is a Colorfront proprietary shadow highlight LOG color correction tool.

CameraColor

This simulates physical camera based exposure and temperature on any given image asset.

HDR

This uses a simple highlight boost to convert from SDR to HDR.

CFSector

The legacy Colorfront Sector tool, which is part of the Colorfront Engine, allows for the selective modification of color sectors, including saturation, color, hue or brightness.

CFSector2

The next generation Colorfront Sector tool, which is part of the Colorfront Engine, allows for the selective modification of color sectors, including saturation, color, hue or brightness.

SkinTone

This skin tone enhancement tool, which is part of the Colorfront Engine, allows for the simple correction of problematic skin tones without the use of color keys or shapes. It works by collapsing the tones that are near skin tone toward the actual skin tones, with the ability to specify either warmer or cooler tones.

Curve

This is a flexible curve tool which uses four point graph to remap brightness.

LookBlend

This is a creative mixer and creater tool, and is one of the main components of the Colorfront Engine. It is similar to the RSD tool in the ACES workflow.

PrinterLight

This is a traditional PrinterLight offset tool.

RGBlin

This is a Lift, Gamma, Gain type of color correction tool.

DolbyVision

This is a tool for Dolby Vision metadata adjustment and creation tool including creative adjustments and statistics.

7.2.3. List of Enhance Tools

For details about the functionality and controls of the following tools, see Enhance Tools chapter, or click the name of the certain tool to see its detailed description.

Tool Name Tool Description

AddGrain

This tool adds film grain.

BlurMix

This is a blur tool that also allows for all sorts of visual effects.

Detail

This is a detail enchancement tool.

Gradient

This is a color gradient overlay creative tool.

LensDistort

This tool allows for making certain geometrical adjustments, i.e. correction or distortion to the image.

SoftFilter

This tool allows for the control of the amount, blur, range and saturation of a soft filter.

Sharpen

This tool adds a sharpen filter.

Vignette

This tool is used to darken or color the sides and corners of the image by an elliptical transparent shape.

7.2.4. List of Mask Tools

For details about the functionality and controls of the following tools, see Mask Tools chapter, or click the name of the certain tool to see its detailed description.

Tool Name Tool Description

GradientMask

This tool allows for the creation of a linear gradient type mask on top of images.

HueKey

This tool allows you to adjust hue-based color key for color separation.

MaskOps

This tool allows for the application of logical operation on two input masks, such as GradientMask, HueKey, etc.

Shape

This tool allows you to draw and use customized shapes as masks.

VignetteMask

This tool is used to create an elliptical shaped mask.

7.2.5. List of Transform Tools

For details about the functionality and controls of the following tools, see Transform Tools chapter, or click the name of the certain tool to see its detailed description.

Tool Name Tool Description

Crop

A basic image cropping tool.

DVE2D

A basic 2D image transformation tool with optional auto-zoom, allowing for zoom, pan X, pan Y, flip X and flip Y. Yet, we recommend you to use the_Reframe_ tool instead, because that is the most complete tool for 2D transformations.

Reframe

The preferred tool for image geometry correction including zoom, letterbox, pan, tilt, etc.

Resize

A tool for resizing the image to any required size. Presets are available for industry standards and specific devices.

UltraScaler

A special Colorfront proprietary scaler tool with advanced detail and edge preservation. It is the best quality upscale tool, which allows for creating UHD from HD, 4K from 2K, etc.

Zoom

A legacy simple Zoom tool with pan X, and pan Y support.

7.2.6. List of Other Tools

For details about the functionality and controls of the following tools, see Other Tools chapter, or click the name of the certain tool to see its detailed description.

Tool Name Tool Description

Analyze

This is a tool for using the Analyzer on any point of the node pipeline different from the result render.

Calibration

…​

FrameBlend

This is a Colorfront proprietary tool, which is used to flexibly blend consecutive frames, on ARRIRAW and Panasonic VRAW.

Burn-In

This is a text burn-in tool.

Subtitle

This is a required tool for subtitle rasterization from the subtitle track.

SubtitleHDR

This is a required tool for subtitle rasterization from the subtitle track for HDR content.

PSNR

This is a required tool for manually calculating PSNR statistics.

ICtCp

…​

Bitrate

…​

Blend

…​

AlphaAdjust

This tool allows for the adjustment of the Alpha channel.

Split

The SplitView tool combines the two input images in different split modes, making it an ideal tool to compare different image processing pipes on the same input image.

7.3. Color Space Tools

7.3.1. ARRILook (Legacy workflow)

This is a node in the ARRI ALEXA camera ARRILook (legacy) workflow pipeline. It is used for mapping from the source media Log C image format to the Rec.709 image format. The ARRILook node is automatically created as the second of the three grading nodes of the ARRILook (legacy) workflow pipeline, with the first being the PrinterLight node and the third being the CDL node.

Transkoder ARRILook Pipeline creation
Transkoder ARRILook Pipeline creation

The Knee and Roll slider controls are used to refine the highlight and shadow handling.

The LutFromMedia field is used to determine if any potentially present LUT metadata from the source footage is to be used in the mapping.

The Matrix field value is On by default which means that the Log C color primaries are mapped to Rec.709 color primaries.

In some rare cases, it may be desirable to preserve the native Log C camera color primaries. This may be done by setting the Matrix field value to Off. For expert use only.

The Saturation slider control is used to control the intensity of colors, and the value is automatically adjusted if LUT metadata is present in the source image, and enabled via the LutFromMedia field.

Transkoder - ARRILook node controls
Transkoder - ARRILook node controls

7.3.2. ARRIRAW

The ARRIRAW node is used to configure the color transformation process of the ARRIRAW image format, debayering into a working RGB image format with the help of source image metadata, and the application of color science based tools. The controls in the node itself are automatically set to the values contained in the source image metadata, and they may be overridden using these controls.

The ASAPreset field is used to retroactively adjust the camera sensor sensitivity with either predefined values, or a Custom value that can be configured using the ASA slider control directly below it.

The white balance may be adjusted with the ColorTemp field using either predefined values, or a Custom value that can be configured using the CustomTemp slider control directly below it.

The Tint slider control is used to increase or decrease the lightness of the colors.

The Sharpen slider control is used to increase or decrease the perceived sharpness of the image through the use of edge sharpening.

The Apply field is used to apply the adjusted settings to selected, or to all shots.

Transkoder - ARRIRAW node controls
Transkoder - ARRIRAW node controls

7.3.3. Amira

The Amira node is a node in the pre-ALF2 Amira camera Amira (legacy) workflow pipeline. It is used for mapping from the source media Log C image format to the Rec.709 image format using a 3dmesh LUT, which is contained as metadata in the source image. The Amira node is automatically created after the CDL grading node of the Amira (legacy) workflow pipeline.

The Target field value is automatically set to the Default Rec709 value for mapping from the Log C color space of the source image.

Transkoder - Amira node controls
Transkoder - Amira node controls

7.3.4. ColorConv

The ColorConv (CSC) node is a "Swiss army knife" color conversion tool used to map from any color space representation to any other color space representation.

Transkoder - ColorConv node controls
Transkoder - ColorConv node controls

7.3.5. DisplayLookHDR

The DisplayLook HDR (DLook HDR) node is the final node of the Colorfront Engine pipeline, which maps from the Colorfront internal color space to the viewing color space. This tool supports both PQ (Perceptual Quantizier) and HLG (Hybrid Log-Gamma) encoding. HLG mode implements the variable HLG transfer function as per the recent standard with advances handling of out-of-gamut colors.

Transkoder - DisplayLookHDR node controls
Transkoder - DisplayLookHDR node controls

7.3.6. SonyRAW

7.3.7. HDR IDT

The HDR IDT node is a tool for an Input Device Transform (IDT), at the begining of the ACES workflow pipeline, which maps source HDR assets into the ACES color space and encoding specifications.

Transkoder - HDR IDT node controls
Transkoder - HDR IDT node controls

7.3.8. HDRSDRRemap

The HDR-SDR Remap node is based on the advanced color volume transformation engine used to map between HDR and SDR color spaces. It can also be used for converting between HDR and HDR, such as from 1,000 nits to 100 nits Rec.709, or between different flavours of HDR, including 4,000 or 1,000 nits PQ, or Hybrid Log-Gamma (HLG). Simulating an SDR image on an HDR monitor is also possible using the HDRMode field options. Please refer to the High Dynamic Range (HDR) section for more information on HDR in general, and specifically SDR ←→ HDR and HDR ←→ HDR conversions.

Transkoder - HDRSDRRemap node controls
Transkoder - HDRSDRRemap node controls

7.3.9. HDR10plus

HDR10plus is an open standard for dynamic HDR metadata. In contrast to static HDR10, HDR10+ provides a per-scene dynamic tone mapping curve, so TV sets are able to apply a proper remapping of bright pixels that would clip without adjustment. Transkoder on the Windows platform can generate this metadata, and also apply the remap itself to the image. The HDR10 Plus node is used to simulate the HDR10+ processing: when initialized, the output is the remapped PQ image. To initialize the HDR10+ node, please load the dynamic metadata from a JSON file, using the corresponding option from the Timeline menu.

The "Percentile" Order field is used to determine the list of percentage values, which are used to calculate percentiles for HDR10+. For more details please refer to the HDR10+ Alliance.

Transkoder - HDR10plus node controls
Transkoder - HDR10plus node controls

7.3.10. LUT

This is a general LUT node, where the operator can select a lookup table from the list of preinstalled LUTs, the ones stored in the S3 cloud storage, or from any custom one that have been copied before into the LUT folder within the install home. Supported formats at varying sizes include:

  • 3DMESH

  • LUT

  • 3DL

  • CUBE

  • ACES CLF

Transkoder - LUT node controls
Transkoder - LUT node controls

7.3.11. Tetrahedral LUT

7.3.12. Matrix

The Matrix node is a technical tool to apply a 3x3 linear matrix transformation to the RGB color vectors. The matrix entries are driven by the sliders with corresponding names. By default, this is the identity matrix.

Transkoder - Matrix node controls
Transkoder - Matrix node controls

7.3.13. LUTACES

The LUTACES (ACES LUT) node works with ACES CLF LUT files that need to be copied into the aceslut folder under the install home.

Transkoder - LUTACES node controls
Transkoder - LUTACES node controls

7.3.14. ODT

The last node in the ACES pipeline is the ODT (Output Device Transform) node. Its input must be the RRT, while its output color space corresponds to some physically realized output device with limited gamut and dynamic range.

Transkoder - ODT node controls
Transkoder - ODT node controls

7.3.15. ODTInv

This is the inverse of the ODT above. It is used only for advanced color pipeline design.

Transkoder - ODTInv node controls
Transkoder - ODTInv node controls

7.3.16. PhantomRAW

7.3.17. RED

The RED node serves as the interface to the RED debayer settings. Configurations set up here apply to how the Red images are extracted from the source files. The Color and Gamma options are global to the project. These are the same controls as in the Red section of the Camera settings in the Settings Page.

By using the RMD button, the operator can save back the metadata to the RMD sidecar files, so the customized settings can travel further with the original media.

The RED node does not do actual pixel processing and even works when not connected to the processing chain.
Transkoder - RED node controls
Transkoder - RED node controls

7.3.18. RRT

The RRT (Reference Rendering Transform) node is part of the ACES pipeline. It converts the scene-referred colorimetry to display-referred, and resembles traditional film image rendering with an S-shaped curve. It has a larger gamut and dynamic range available to allow for rendering to any output device (even ones not yet in existence). RRT node uses the 1.0 definition, which is in the ACES 1.0.3 specification.

Transkoder - RRT node controls
Transkoder - RRT node controls

7.3.19. RRTInv

The RRTInv node is simply the inverse of the RRT node. This tool is used only in custom color pipeline designs.

7.3.20. VideoRange

The Video Range node is a tool to apply range mappings: legal/head range to fill and vica versa. By using the preset drop-down selector, one can pick one of the standard remappings, but it is also possible to adjust the sliders to control exactly what is happening.

The Clip options allow for clipping the super-white and super-black color values.

Transkoder - VideoRange node controls
Transkoder - VideoRange node controls

7.4. Grading Tools

7.4.1. CDL

The CDL node is the standard Color Decision List (CDL) compatible grading node. It is used to adjust colors in a variety of color spaces such as Log C and Rec.709. It is used in a variety of predefined workflow pipelines. Individual and master slope/offset/power parameters and the saturation may be adjusted. Files with the .CC file extension are also supported.

Transkoder - CDL node controls
Transkoder - CDL node controls
CDL in ARRILook (Legacy) Workflow Pipeline

This is a node in the ARRI ALEXA camera ARRILook (legacy) workflow pipeline. It is used to adjust the colors in the Rec.709 color space. The CDL node is automatically created as the third of the three grading nodes of the ARRILook pipeline, with the first being the PrinterLight node and the second being the ARRILook node.

CDL in Amira (Legacy) Workflow Pipeline

The is a node in the pre-ALF2 Amira camera Amira (legacy) workflow pipeline. It is used to adjust the colors in the Log C color space. The CDL grading node is automatically created before the Amira node of the Amira (legacy) workflow pipeline.

7.4.2. CFPrimary

The CF Primary (CFRGB) node is a Colorfront designed shadow/midtone/highlight specific advanced color correction tool, where the cut-off points and transition softness of these ranges are adjustable. Ideal for LOG color correction.

Transkoder - CFPrimary node controls
Transkoder - CFPrimary node controls

7.4.3. CameraColor

The Camera Color node is part of the Colorfront Engine color pipeline and is supposed to operate on CF Log images.

Full scene referred linear light adjustments are used to compensate for scene lighting intensity and color temperature. This is the equivalent of adjusting the photons landing on the camera sensor. It uses familiar terms such as F-stops and color temperature in Kelvin. The tool works the same way with any camera as the input providing a unified approach to grading input images from various sources. For example, a one-stop adjustment applied to an ARRI Alexa will have the same effect when applied to a SONY or RED camera. Rather than having to use standard creative color correction tools - like other systems do - this tool is designed to deal with a specific need quickly and efficiently.

Allows adjustment of camera "RAW" settings - even if the media is not shot as raw - such as: exposure, color temperature, tint.

Transkoder - CameraColor node controls
Transkoder - CameraColor node controls

Automatic camera matching is available to quickly and accurately compensate for different lenses and filters. To match one image to another first park on the reference frame, press the Pick button a click on a color sample. Ideally on a grey patch of a Macbeth chart. Then part on the clip to be adjusted, press Balance and sample the same color with a mouse click. To balance an image without a reference simply skip the first step, enable Balance and click on a 18% grey patch.

7.4.4. HDR

The HDR node is used to convert from SDR to HDR via a simple highlight boost. The Brightness, Higlight, and Boost slider controls are used to control this remap and the On/Off button of the V2 field controls the algorithm used for this conversion. The default value is On indicating that V2 of the algorithm is being used instead of the legacy V1 algorithm.

Transkoder - HDR node controls
Transkoder - HDR node controls

7.4.5. CFSector

The CF Sector (CFSector) node is the legacy Colorfront designed grading tool to control individual sectors of red, yellow, green, cyan, blue, and magenta.

Transkoder - CFSector node controls
Transkoder - CFSector node controls

7.4.6. CFSector2

The CF Sector 2 (CFSEC) node is the current Colorfront designed grading tool to control individual sectors of red, yellow, green, cyan, blue, and magenta.

Transkoder - CFSector2 node controls
Transkoder - CFSector2 node controls

7.4.7. SkinTone

The SkinTone (Skin) node is a sophisticated toolset, only available in Colorfront Engine projects, and is used to adjust skin tones with control of tonal variation, hue, and color. As the Amount slider is increased the hue of skin tone colors are “pulled” towards a reference skin tone while not changing their brightness. To adjust the tone of this target color the Tone slider should be used. The Color slider is used to adjust the saturation of the skin tones. This adjustment is not limited by any shape or spatial mask, and is a very effective tool to fix skin tone problems.

SkinTone node controls
Amount

Controls the strength of the effect.

Tone

Controls the tone of the target skin tone and is where the colors are pulled towards a reference skin tone.

Color

Controls the saturation of the final skin tones.

Transkoder - Skin tone node controls
Transkoder - Skin tone node controls

7.4.8. Curve

The Curve node provides a simple curve tool to remap the master or individual channel intensities of the image.

Transkoder - Curve node controls
Transkoder - Curve node controls
Tonality Adjustment Using the Curve Window

In addition to the Curve node too, it is possible to adjust tonality using the Curve window which may be accessed by pressing Shift+Z or clicking on the Window  Show Curve Window menu option. The image tonality is displayed as a diagonal line. The top right corner is used to manipulate highlights and the bottom left corner is used to manipulate shadows. As you move the line or the point on the line, the image tonality is adjusted. The adjustments may be made each RGB channel individually or combined.

Tonality Adjustment Using the Curve Window
Tonality Adjustment Using the Curve Window

7.4.9. LookBlend

The LookBlend node as part of the Colorfront Engine pipeline, lets the user to mix different tone mapping operators to achieve the desired look. As an example, one can balance between the tone mapping effect of the ACES color transformations and a traditional film stock. These operators, or looks are high precesion floating point mappings.

Transkoder LookBlend node controls
Transkoder LookBlend node controls

7.4.10. PrinterLight

The Printerlight node is a classic printer light adjustment tool. The node in the ARRI ALEXA camera ARRILook (legacy) workflow pipeline. It is used to adjust the intenstity of the Log C RGB color primaries. The PrinterLight node is automatically created as the first of the three grading nodes of the ARRILook pipeline, with the second being the ARRILook node and the third being the CDL node.

Transkoder PrinterLight node controls
Transkoder PrinterLight node controls

7.4.11. RGBlin

The RGBLin node is a Lift/Gain/Gamma operator, mostly used for "video-style" color correction. Fully compatible with CDL but offers a different set of controls.

Transkoder RGBlin node controls
Transkoder RGBlin node controls

7.4.12. DolbyVision

The Dolby Vision (DoVi) node is the interface to create dynamic HDR Dolby metadata, and is also a processing node to apply the Dolby Vision remap, namely the soft-CMU. When working on a timeline with dynamic metadata from an XML source (orange), or an IMF asset (yellow), the sliders on the node are disabled and are not representative of the current metadata. Check the "D" tab of the Clip Inspector window Shift+U to inspect the current metadata in such cases.

To create Dolby Vision metadata within Transkoder, first the statistical Level 1 (L1) metadata needs to be initialized. For this, please press the Init button. As soon as the three bottom sliders are initialized, the Dolby Vision metadata in the node becomes valid. Make sure to first initialize the Dolby Vision L1 metadata on a timeline with unedited clips, without any trims and cuts. Once the statistics are processed, the clips may be edited freely.

The Dolby Vision mastering metadata comprises of global static metadata that is applied to the entire deliverable, and dynamic metadata applied on a per shot or a per frame basis. The first three levels of the metadata 2.0.5 version XML are the following:

  • Level 0 (L0) - static global metadata, including for the mastering display, aspect ratio, frame rate, target display and color encoding specifications.

  • Level 1 (L1) - dynamic per-shot statistical metadata, characterized by three floating point values: minimum, average and maximum brightness.

  • Level 2 (L2) - optional dynamic per-shot metadata, defined by creative trimming by adjusting the lift, gain, gamma, saturation, chroma and tone detail controls in order to fine-tune the result of the automatic mapping based on the L1 metadata.

DolbyVision Node Controls

The top part of the HUD is for target display selection and Level 2 (L2) metadata selection:

TargetDisplay

It selects the target display which you want to simulate the current remap for. It is effective when the Dolby Vision remapping is enabled, and works in all cases regardless of the metadata source (timeline, XML or IMF).

MetadataBank

It selects which target display you create the metadata, in other words, the current L2 metadata set, for. You can define multiple sets of L2 trim metadata for different target displays. The grading controls for the selected display can be defined with the sliders below. The metadata bank will store all Dolby Vision metadata for the selected and remapped target display.

The Dolby Vision node always stores L2 trims for 100-nit SDR. Additional metadata for other target displays is optional. The operator has to manually enable the corresponding metadata bank. It is not required to create L2 metadata for all possible target displays. The Dolby Vision algorithm will use the lower and possibly the higher nit-level metadata to remap to any other target.

The next few sliders are for adjusting the Dolby Vision remapping. This fine-tuning may be needed in case the default remap based on L1 statistics is not satisfactory for the current target display. First, go the the Dolby Vision section of the Settings Page and set the Remap field to either Preview (the adjustment will not appear in the rendered file, only in the metadata) or On (adjustment will also appear in the rendered file) setting to instantly see the result of the creative trimming.

Lift

It controls the black level of the image. You can adjust the dark tones or shadows, while the highlight area will not change significantly.

Gamma

It controls the image levels between the shadows and the highlights, commonly known as the mid-tones. By adjusting this, mid-tones will become brighter or darker, while the whites and blacks will not change significantly.

Gain

It controls the white level of the image. While it affects all brightness levels within the image, it gives more weight to the highlights or brighter areas.

ChromaOffset

It allows you to adjust the balance between brightness and saturation. You can lighten or darken the colorful parts of the image to avoid color clipping.

Saturation

It controls the overall saturation of the image.

ToneDetail

It allows you to control the level of detail in highlights that might otherwise be lost when remapping to lower dynamic ranges. Its functionality, however, is only available in higher nit target trims.

Image Aspect

With this slider, you can finetune the image aspect ratio on a per shot basis in case of special shots with custom image aspect ratio.

When the Enabled button is On, the metadata will include trim values for the selected target display. The 100-nit target display is always defined.

Remember to always click On to activate the Enabled button when switching the MetadataBank from one target display setting to another. Otherwise, your creative trimming will neither show, nor will be recorded.

The lower part of the HUD consists of the calculation operations for the L1 statistics, as follows:

Init

It initializes the DolbyVision node, and updates the L1 statistical metadata on the selected shots.

Copy

It copies the L1 analysis metadata of the shot where the playhead is positioned to, to all the shots. If one or more shots are selected (highlighted in yellow), the metadata will be copied only to those. You can reset the metadata by clicking Init.

Avg

It averages the L1 metadata among the shots. If more shots are selected, the metadata will be averaged only among the selected ones.

The sliders on the bottom show the initial L1 metadata that is generated during the statistical analysis. As their names imply, the three different parts of the tonal range are mapped on a per shot basis, or averaged over the entire footage, as follows:

Crush

The minimum of normalized PQ values between 0 and 1, referring to the darkest pixel of the clip.

Mid

The average of normalized PQ values between 0 and 1, referring to the averaged pixel brightness of the clip.

Clip

The maximum of normalized PQ values between 0 and 1, referring to the lightest pixel of the clip.

Transkoder DolbyVision node controls
Transkoder DolbyVision node controls

7.5. Enhance Tools

7.5.1. AddGrain

The AddGrain (Grain) node is used to generate film grain on top of the images and is controlled with the following settings:

Amount

Controls the strength of the effect

Blur

Controls the separation of the shadow/midtone/highlight ranges, less blur means sharper separation.

Size

Size of the grain particles

Saturation

Puts Saturation back to where the Blur effect decreased it.

Shadow

Grain strength in the shadow range

Midtone

Grain strength in the midtone range

Highlight

Grain strength in the highlight range

Transkoder AddGrain node controls
Transkoder AddGrain node controls

7.5.2. BlurMix

The BlurMix node applies a blur effect on the image by multiplication or addition.

Blur Size

Controls the range size of the Gaussian blur effect.

Mix

Controls the opacity of the effect layer. Values between 0 and 1 apply a blur effect, while values between -1 and 0 apply a sharpen effect.

Desaturation

Controls the saturation of the effect layer. Values between 0 and 1 decrease saturation, while values between -1 and 0 increase saturation.

Additive

Enables the additive mode for the blur effect. The default setting is off, which means multiplying when calculating the result of the effect.

Offset

…​ This control only has effect when the Additive mode is enabled.

Transkoder BlurMix node controls
Transkoder BlurMix node controls

7.5.3. Detail

Transkoder Detail node controls
Transkoder Detail node controls

7.5.4. Gradient

The Gradient node’s effect is similar to a vignette effect, but it applies a localized color adjustment typically on the top of the image using a smooth linear transition. This effectively simulates real optical graduated filters, a popular tool to darken the overly-bright sky area to achieve a more even exposure of the image.

Mix

Overall strength of the effect.

Softness

The size of the transition range.

Position

The position of the mask.

Rotation

Orientation control of the gradient effect.

Transkoder Gradient node controls
Transkoder Gradient node controls

7.5.5. LensDistort

The LensDistort (Lens) node allows for making certain geometrical adjustments, i.e. correction or distortion to the image.

Distortion

This parameter is to correct or add radial (i.e. barrel) or pincushion distortion. Values below 0.50 add a pincushion distortion effect to the image, which is useful to correct an image having barrel distortion in the source image. Values between 0.50 and 1.50 do the opposite and are useful to correct images having pincushion distortion originally.

Pan X

Shifts the center of the effected image between the left side (zero) and the right side of the image (1). The default setting is 0.50, which indicates the center of the image.

Pan Y

Shifts the center of the effected image between the bottom (zero) and the top of the image (1). The default setting is 0.50, which indicates the center of the image.

Horizontal Tilt

Tilts the image to the left or to the right around a vertical axis. This is useful to correct or add perspective distortion.

Vertical Tilt

Tilts the image to the front or to the back around a horizontal axis. This is useful to correct or add perspective distortion.

Horizontal Axis

Moves the axis of the Vertical Tilt from the right to the left of the image.

Vertical Axis

Moves the axis of the Vertical Tilt from the bottom to the top of the image.

Zoom

Scales the image size.

Camera Distance

Controls perspective distortion along with scale by moving the "virtual camera".

Transkoder LensDistort node controls
Transkoder LensDistort node controls

7.5.6. SoftFilter

The SoftFilter node is used to create a soft effect which reduces the image sharpness is controlled with the following settings:

Amount

Controls the brightness of the effect.

Blur

Controls the effect spread.

Range

Controls whether to apply the effect only to highlights or to midtones/shadows as well.

Saturation

Puts Saturation back to where the Blur effect decreased it.

Transkoder SoftFilter node controls
Transkoder SoftFilter node controls

7.5.7. Sharpen

Transkoder Sharpen node controls
Transkoder Sharpen node controls

7.5.8. Vignette

The Vignette node adds a well-adjustable vignette effect to the image, which resembles lens vignetting. One can darken or color the sides and corners of the image by an elliptical transparent shape. Its controls are the following:

Mix

It is the opacity of the effect. It ranges from 0 (transparent) to 1 (dark transparent).

Master

It controls the strength and the brightness of the applied effect. Negative values add a black vignette, while positive values add a white vignette to the image. The further the value is from zero in either direction, the stronger the effect.

Red

Values between 0 and 1 add a red shade to the effect, while values between -1 and 0 subtract a red shade, i.e. add a cyan shade to it.

Green

Values between 0 and 1 add a green shade to the effect, while values between -1 and 0 subtract a green shade, i.e. add a magenta shade to it.

Blue

Values between 0 and 1 add a blue shade to the effect, while values between -1 and 0 subtract a blue shade, i.e. add a yellow shade to it.

Softness

Defines the size of the transition range at the edges. Lower value means sharper, while higher value means smoother transition.

Radius

Defines the size of the ellipse from 0 to 200 percent of the image width.

Aspect

Controls the aspect ratio of the effect between round and flat. The width is constant while you set the ratio with the slider.

Size X

Controls the size of the effect horizontally from 0 to 200 percent of the image width.

Size Y

Controls the size of the effect vertically from 0 to 200 percent of the image width.

X

Controls the horizontal position of the ellipse.

Y

Controls the vertical position of the ellipse.

Rotation

Rotates the ellipse around its center point clockwise from 0 to 180 degrees.

Transkoder Vignette node controls
Transkoder Vignette node controls
Vignette node - Original image
Vignette node - Original image
Vignette node - Image with black vignette (negative master values)
Vignette node - Image with black vignette (negative master values)
Vignette node - Image with white vignette (positive master values)
Vignette node - Image with white vignette (positive master values)

7.6. Mask Tools

Mask tools are designed to limit the effects of CFPrimary, the most advanced color correction node of the color grading tools, to a certain part of the image. You can create mask tool nodes and combine them in the node pipeline workflow. These nodes have a blue video (and, in some cases, a green mask) input, and a green mask output. The exception for this is the MaskOps node that has only green mask inputs and output, and is used to combine the outputs of other mask tool nodes.

There are five mask tools, which are the following:

7.6.1. GradientMask

The GradientMask node is used to apply a gradient type mask on top of images using a linear transition. It is controlled by the following settings:

Mix

Controls the opacity of the parts masked out.

Softness

Defines the size of the transition range, thus the transition can be sharp or smooth.

Position

Controls the position of the gradient, where 0 is the top of the image and 1 is the bottom.

Rotation

Rotates the mask around the center of the image.

Log Mode

Logarithmic mode switch to be used only in case of logarithmic images.

Transkoder GradientMask node controls
Transkoder - GradientMask node controls
To create a regular gradient effect to the sky for example, you may need to use the Gradient mask together with a MaskOps node to invert the mask.

7.6.2. HueKey

The HueKey node is used to create an advanced color key mask for the image for color separation. Click on any of the Picker buttons to pick, add or remove a color by clicking the left mouse button on the image.

Picker buttons
Pick

Use this mode to select the dominant color first.

Add

Add additional colors to include them in the mask.

Remove

Remove any colors from the selection that you want to exclude from the mask.

After selecting the most important parts of the image, you can make the fine-tuning with the rest of the controls as follows. Colors used for keying can be selected by three properties: hue, luminance and saturation.

Hue

Enables color separation by hue when turned on.

Hue Center

Defines the center of the selected color range.

Hue Width

Defines the width of the selected color range.

Hue Soft

Controls the softness for the edges of the hue range. Lower value means sharper, while higher value means smoother transition.

Luminance

Enables color separation by luminance when turned on.

Luminance Min

Specifies the minimum luminance of pixels of the selected color range.

Luminance Max

Specifies the maximum luminance of pixels of the selected color range.

Luminance Soft

Controls the softness for the edges of the luminance range. Lower value means sharper, while higher value means smoother transition.

Saturation

Enables color separation by saturation when turned on.

Saturation Min

Specifies the minimum saturation of pixels of the selected color range.

Saturation Max

Specifies the maximum saturation of pixels of the selected color range.

Saturation Soft

Controls the softness for the edges of the saturation range. Lower value means sharper, while higher value means smoother transition.

Key Blur

Applies a blur filter on the image mask. Lower value means sharper, while higher value means smoother transition.

Invert

Inverts the mask.

Transkoder HueKey node controls
Transkoder - HueKey node controls

7.6.3. MaskOps

The MaskOps node combines two masks, such as Shape, HueKey, etc., by logical operators. The logical operation is defined by the Mode settings as follows:

Min

It is the intersection of two masks. The output mask uses the smaller value from the two inputs for each pixel.

Max

It is the union of two masks. The output mask uses the greater value from the two inputs for each pixel.

Add

It is the additive combination of two masks, which is like putting the two masks layered on each other.

Invert A

Inverts input A.

Invert B

Inverts input B.

Invert Result

Inverts the result (output) of the operation.

Transkoder MaskOps node controls
Transkoder - MaskOps node controls

7.6.4. Shape

The Shape node is used to draw customized shapes by spline curves to be used as masks. You can create multiple shape masks on images. The node is controlled by the following settings:

Create

Turn on this mode to add a new shape to the mask. Click on the image to create control points and draw any shape. To close a the shape, click on its first control point.

Interact

While this mode is turned on, click and drag any control point to change a shape, or select a whole shape by clicking on its edge and drag it by grabbing its center point marked by a cross.

Show Spline

Turn on this mode to see the spline curve of the shapes.

Show Points

Turn on this mode to see the control points of the shapes.

Weight

First select a certain control point or a whole shape (in Interact mode), then move the slider to change the hardness or smoothness of the spline.

Opacity

Defines the opacity of the selected shape, where 0 is fully masked out and 1 is transparent.

Blur

Defines the size of the transition range at the edges of the selected shape. Lower value means sharper, while higher value means smoother transition.

Pos X

Controls the horizontal position of the selected shape.

Pos Y

Controls the vertical position of the selected shape.

Rot

Rotates the selected shape around its center point.

Scale

Controls the size of the shape.

Transkoder Shape node controls
Transkoder - Shape node controls

7.6.5. VignetteMask

The Vignette Mask applies an elliptical shaped mask to the image. It is controlled by the following settings:

Mix

Controls the opacity of the parts masked out.

Softness

Defines the size of the transition range at the edges. Lower value means sharper, while higher value means smoother transition.

Radius

Defines the size of the ellipse from 0 to 200 percent of the image width.

Aspect

Controls the aspect ratio of the effect between round and flat. The width is constant while you set the ratio with the slider.

Size X

Controls the size of the effect horizontally from 0 to 200 percent of the image width.

Size Y

Controls the size of the effect vertically from 0 to 200 percent of the image width.

X

Controls the horizontal position of the ellipse.

Y

Controls the vertical position of the ellipse.

Rotation

Rotates the ellipse around its center point clockwise from 0 to 180 degrees.

To create a regular vignette effect, you can use the Vignette or the Vignette Mask together with the MaskOps node to invert the mask.
Transkoder VignetteMask node controls
Transkoder - VignetteMask node controls

7.7. Transform Tools

Transform nodes are designed to make basic transformations to the image.

7.7.1. Crop

The Crop node is a basic image cropping tool.

Enable

Turn off this option if you want to bypass the effect of the node.

New Height

Controls the height of the transformation output.

New Width

Controls the width of the transformation output.

Offset X

Shifts the image to the left by the number of pixels set by the slider.

Offset Y

Shifts the image to the top by the number of pixels set by the slider.

The origo of this transformation node is the top left corner of the image.
Transkoder Crop node controls
Transkoder Crop node controls

7.7.2. DVE2D

The DVE2D node stands for Digital Video Effect in 2D. This is a basic 2D image transformation tool. You can control it as follows:

Rot

Controls the rotation of the image from -20 to 20 radian.

Zoom

Scales the image size between 10 and 1000% of the original size.

Pan X

Sets a horizontal offset for the position along the X axis.

Pan Y

Sets a vertical offset for the position along the Y axis.

Flip X

Flips the image along the X axis if turned on.

Flip Y

Flips the image along the Y axis if turned on.

Auto-Zoom

Scales the image size automatically so that the image always covers the display area, thus there is no black area in the result image.

For rotation, we recommend to use the Reframe node instead.
Transkoder DVE2D node controls
Transkoder DVE2D node controls

7.7.3. Reframe

The Reframe node is the preferred tool for image geometry correction including offset, zoom, letterbox, pan, tilt, etc. It uses a high quality Lanczos filter for scaling.

Preset

Choose an output resolution preset (e.g. more variations for iPhone, NTSC, PAL, HD, 2K, 4K, 8K, etc.). You can choose to use the Project Resolution as well.

Width (X)

Sets the width of the output image in pixels.

Height (Y)

Sets the height of the output image in pixels.

Zoom

Sets the size of the input image within the zoom range of 10-300%.

Pan

Shifts the image along the X axis. With the minus values, move the input image to the left, while with plus values to the right.

Tilt

Shifts the image along the Y axis. With the minus values, move the input image to the top, while with plus values to the bottom.

Aspect Preset

Choose the aspect ratio for the input image from the given options (1.0, 1.3, 2.0) or set a customized value with the slider below.

Aspect

Changes the width/height ratio of the input image within the range of 50-200%.

Mode

Choose from the following modes for the reframing process.

Center

This mode has no effects on the size of the input image. If the resolution of the input image is bigger than the output image, it crops the parts beyond the borders. If the input image has smaller resolution than the output image, it adds black parts on the sides (pillarbox), or the top and the bottom (letterbox).

Fit

Covers the output display area with the biggest part of the input image possible. According to this, it will crop or even upscale the input image if needed.

Fit All

Scales input image size so that the whole image fits inside the output display area. According to this, some black areas will be added on the sides (pillarbox) or at the top and the bottom (letterbox) of the image. For output resolutions smaller than that of the input, the input image will be downscaled.

Stretch

Distorts the input image so that the whole image covers the output display area. Note that the aspect ratio of the input image will change.

Letterbox Preset

Choose the aspect ratio for the active image from the given options, or set a customized value with the slider below.

Letterbox

Controls the active image aspect ratio. The height of the areas covered with black at the top and the bottom of the image will change so that the aspect ratio of the active image will be the value set.

Pillarbox

Controls the width of the areas covered with black at the sides of the image from 0 to 50% of the image.

Flip X

Flips the image along the X axis if turned on.

Flip Y

Flips the image along the Y axis if turned on.

For good quality scaling, please make sure to use the Reframe node on some logarithmic or Rec.709 images. Scaling light linear images may cause severe aliasing near the edges.
Transkoder Reframe node controls
Transkoder Reframe node controls

7.7.4. Resize

The Resize node is for resizing the image to any required resolution. Presets are available for industrial standards and some specific devices. It uses a high quality Lanczos filter for scaling.

Preset

Choose an output resolution preset (e.g. more variations for iPhone, NTSC, PAL, HD, 2K, 4K, 8K, etc.).

Width (X)

Sets the width of the output image in pixels.

Height (Y)

Sets the height of the output image in pixels.

Zoom

Sets the size of the input image within the zoom range of 50-400%.

Aspect

Changes the height/width ratio of the input image within the range of 50-150%.

Mode

Choose from the following modes for the reframing process.

Center

This mode has no effects on the size of the input image. If the resolution of the input image is bigger than the output image, then it crops the parts beyond the borders. If the input image has smaller resolution than the output image then it adds black parts on the sides (pillarbox) or the top and the bottom (letterbox).

Fit

Covers the output display area with the biggest part of the input image possible. According to this, it will crop or even upscale the input image if needed.

Fit All

Scales input image size so that the whole image fits inside the output display area. According to this, some black areas will be added on the sides (pillarbox) or at the top and the bottom (letterbox) of the image. For smaller output resolutions than the input, the input image will be downscaled.

Stretch

Distorts the input image so that the whole image covers the output display area. Note, that the aspect ratio of the input image will change.

Transkoder Resize node controls
Transkoder Resize node controls

7.7.5. UltraScaler

The UltraScaler is a special Colorfront proprietary scaler tool with advanced detail and edge preservation. It is the best quality upscale tool, and scales the image to the project resolution, which can be set with the Image Width/Height parameters in the Base section of the Settings page Tab.

Detail

Add more details or blur the image with this slider. Between -1 and 0 you can add blur, between 0 and 2 you can sharpen the input image.

Sharpen

You can add further refining to the image with this slider. Between -1 and 0 you can add blur, between 0 and 2 you can sharpen the output image.

Transkoder UltraScaler node controls
Transkoder UltraScaler node controls

7.7.6. Zoom

A legacy simple Zoom node with pan X and pan Y support. It uses a fast linear filter for scaling.

Zoom

Zoom into the image with the maximum rate of 400%.

Pan X

When zoomed in, you can shift your image along the X axis. With the minus values, move the input image to the left, while with plus values to the right.

Pan Y

When zoomed in, you can shift your image along the Y axis. With the minus values, move the input image to the top, while with plus values to the bottom.

Preset

Choose an output resolution preset (e.g. more variations for iPhone, NTSC, PAL, HD, 2K, 4K, 8K, etc.).

Transkoder Zoom node controls
Transkoder Zoom node controls

7.8. Other Tools

7.8.1. Analyze

The Analyze node is an auxiliary part of the functionality of the Analyzer. The Analyzer will show the values of the video at the actual placement of the Analyze node along the render pipeline. For more information on the use of one or even multiple Analyze nodes at the same time, please see the Using Analyzer with Analyze Node chapter.

Transkoder Analyze node controls
Transkoder Analyze node controls

7.8.2. Calibration

Transkoder Calibration node controls
Transkoder Calibration node controls

7.8.3. FrameBlend

Transkoder FrameBlend node controls
Transkoder FrameBlend node controls

7.8.4. Burn-In

The Burn-In node is used to configure burn-ins. Six different text based burn-ins, and one mask based burn-in may be configured this tool. Up to two different texts may be specified for each text based burn-in using the Text1, and Text2 fields respectively. The content maybe chosen from a list of keywords or manually entered. The Font, Position, Offset, Alignment, Transparency and Intensity for both text and background, and Drop Shadow are all configurable. The burn-ins also may be cofigured to appear for the length of the entire timeline or fade after a specified ForFrames number of frames (flash burn). Specifying -1 for the value of the ForFrames field will make the burn-in appear for the full length of the render. Specifying a positive whole number, will make the burn-in fade away afterthe specified number of frames. If zero is specified, the burn-in will appear and imediately start to fade. The fade is smooth and takes some time to completely finish.

Transkoder Burn-In node controls
Transkoder Burn-In node controls
Custom Text as Burn-In

To manually configure the burn-in text, select the Custom option from the Text1, or Text2 field and then double click the Text that appears in the main display window. You can use keywords here in the $keyword$ format.

video_$TimeCode$_$ClipName$
Please see the list of supported Keywords.

It is common that the last or first few characters of the keyword need to be trimmed from some metadata names when burning in. The following examples show how to do this:

$ClipName.trim(3,0)$ -> trim the first 3 characters
$ClipName.trim(0,-2)$ -> trim the last 2 characters
$ClipName.trim(1,-1)$ -> trim one character from both sides

7.8.5. Subtitle

The Subtitle node is used to adjust the physical characteristics of the subtitles or captions. Adjustments may be made to the Position, Alignment, Border, Shadow, various Colors, and Offsets. In addition, it can be used to apply the changes all the subtitles or captions by pressing the to all spots, specify whether to encode the text as either Subltitles or Captions, and whether to show the subtitles only on the forced display via the OnlyForcedDisplay field.

Transkoder Subtitle node controls
Transkoder Subtitle node controls

7.8.6. SubtitleHDR

The Subtitle HDR node is used to adjust the physical characteristics of the subtitles or captions for HDR content. Adjustments may be made to the Position, Alignment, Border, Shadow, various Colors, Offsets, Color Space, and Brightness. In addition, it can be used to apply the changes all the subtitles or captions by pressing the to all spots, specify whether to encode the text as either Subltitles or Captions, and whether to show the subtitles only on the forced display via the OnlyForcedDisplay field.

Transkoder SubtitleHDR node controls
Transkoder SubtitleHDR node controls

7.8.7. PSNR

The PSNR node is used to determine the fidelity of the rendered result image as compared to the original source image. It used a Peak Signal-to-Noise Ratio methodology to make this determination. For more information on this tool including a description of workflow, please refer to the PSNR Analysis section of the QC Tools chapter.

Transkoder PSNR node controls
Transkoder PSNR node controls

7.8.8. ICtCp

The ICtCp node is useful to compare two video frames using the delta ICtCp metric. The ICtCp node is an alternative to the existing PSNR node, which produces an output that correlates with the human perceived color difference between two inputs.

ICtCp is a color representation format specified in the Rec. ITU-R BT.2100 standard. ICtCp node measures code differences in the Rec.2020 color space, where a unit difference between any two component is more or less the same throughout the complete range of colors. This means that the maximum and average difference values of the intensity and the chromaticity coordinates provide a better representation of the visual difference between two images than PSNR values.

ictcp
The ICtCp Node and its node controls

As the calculation is performed in a specific color space, the operator needs to feed two images to the node in the same color space, and also indicate it by using the ColorSpace selector. The numerical indicator on the node shows the average and maximum difference between the two inputs.

7.8.9. Bitrate

When playing media files with bitrate data (e.g. IMF compositions) the operator can inspect the current numbers not only in the Graph Window, but also on the Bitrate node as well. It shows the bitrate information on the current frame, a moving window average and the average of all frames. The average numbers are based on the frames already played (and not the complete timeline).

Transkoder Bitrate node controls
Transkoder Bitrate node controls

The Bitrate Histogram window shows a historgram that is also computed based on the frames processed by the node, so it "builds up" data during playback.

Transkoder Bitrate Histogram
Transkoder Bitrate Histogram Window

7.8.10. Blend

The Blend node has two image inputs and a mask. If no mask is connected, the two images are blended together using the mix ratio set. If there is a mask connected, the node will use the mask image as the compositing matte. The Blend node is useful when the operator needs to cover some part of the image with another version, such as texts or texted images. To generate the input mask for that, please use the mask tools, for example the Shape or the Gradient nodes.

blend node
Composition with a Mask by using Blend Node
Invert

…​

Mix

Controls the ratio of the two input images used to create the output image.

Blend Mode

Controls the method of composing the two input images.

Add

Addition of the two images in the ratio set by the Mix slider. When Mix is set to zero, it shows the first, while it is 1 it shows the second input image fully.

Multiply

…​

Composite

…​

CompPreMult

…​

Subtract

Subtracts second image input from the first image. When Mix is set to zero, it shows the first image input fully.

Difference

…​

Transkoder Blend node controls
Transkoder Blend node controls

7.8.11. AlphaAdjust

Transkoder AlphaAdjust node controls
Transkoder AlphaAdjust node controls

7.8.12. Split

The Split node combines the two input images in different split modes, making it an ideal tool to compare different image processing pipes on the same input image.

Available options are the following:

  • Vertical

  • Horizontal

  • Butterfly

  • Checkerboard

  • Side-by-Side

Transkoder SplitView node controls
Transkoder SplitView node controls

7.9. Predefined Workflow Pipelines

7.9.1. ARRILook (Legacy) Workflow Pipeline

The ARRILook (Legacy) Workflow Pipeline is made up of three automatically created grading nodes. The PrinterLight node which is used for Log offset, the ARRILook node which is used for mapping from Log C to Rec.709, and the CDL node which is used for Rec.709 color correction.

7.9.2. Amira (Legacy) Workflow Pipeline

The Amira (Legacy) Workflow Pipeline is made up of an automatically created CDL grading node followed by the Amira node. The CDL node is used of Log C color correction, and the Amira node which is used for mapping from Log C to Rec.709.

8. Rendering

Transkoder has a hybrid CPU/GPU architecture. Image processing with floating point precision, and JPEG2000 encode and decode are all gpu accelerated. Other encoding processing such as Avid or QuickTime encoding is done on the CPU. The architecture also allows for multiple paralell result renders to be done at the same time with different framing, resolution, color space, etc. possible for each result.

To start a standard render, press Ctrl+R.

Transkoder - Render window
Transkoder - Render window

If, however, rendering DCP or IMF, it is recommended to use the Finalize Window instead.

8.1. Encode Page

Bring up the Encode page by pressing the P key. To add the desired encoding to a result, click in an empty row of the table to the right of the particular result. A structured two-level list of encoders will appear. Multiple encoders may be added to any result, and they may be removed by highlighting them and pressing the Backspace key. Encoders in the list include video, audio and metadata encoders.

Transkoder - Encoder selection on Encode Page
Transkoder - Encoder selection on Encode Page

8.1.1. Configure Encoders

Once an encoder has been added, the right-hand side will display a comprehensive and encoder specific configuration pane, which can be scrolled up or down by holding down the right mouse button while moving the mouse up or down.

Transkoder - Encoder pane on Encode Page
Transkoder - Encoder pane on Encode Page

The encoding table in the middle of the screen will contain a list of columns with the most frequently used settings which are common to most encoders. The columns from left to right are the following:

#

This indicates the encoder number for the particular result.

On

This indicates if the encoder is enabled or disabled for the particular render.

Format

This indicates the particular encoder format, such as Avid DNxHR HQX that was chosen from the encoder list when it was added to the result.

Container

This indicates the file container / wrapper format such as .mov, .mxf, etc.

Codec

This indicates the name of the actual codec used to encode the particular result render.

Resolution

This indicates the resolution of the particular result render. Possible values which maybe fixed or multi-select, are codec dependent, and may include Auto, Project, or a choice of specific resolutions. Choosing the Auto setting, will use the result’s resolution, while Project will use the project’s resolution.

Bitrate

This indicates the bitrate the encoder uses.

FPS

This indicates the frame rate of the video of the encode of the particular result render. Clicking on the field will cycle through the available values. If the difference between the encoder and Project/Timeline frame rate is 1000/1001, then the audio will automatically be adjusted and also be reflected in the Sound field. The operator can set the target fps to source for Avid deliverables. Supported rates include: 23.976, 24, 25, 29, 97, 30, 47.95, 48, 50, 59.94, 60, 119.88 and 120.

Selected

This indicates if the render should include All Takes, Selected.

(This is used in Transkoder, when the Select Flag, i.e. Circle Take is used to mark clips. This is not the same selection as highlighting a clip in the timeline window. )

Files

This indicates if the files created by the render will be Per Timeline, or Per Shot. The Per Timeline setting will create a single file for the result, while Per Shot setting will create a file for each clip.

The Render Shot Bounderies setting on Encode tab of the Settings Page (Tab) determines how the shot boundaries of the rendered files will be defined. The possible values are V1Track, VerticalEditing, and ReelMarkers. Please refer to the Encode tab section of the Settings Page chapter for a detailed explanation of this setting.

Sound

This indicates if the sound was Normal, Pulldown, or Pullup.

Mix

This indicates which audio mixdown will be used by the result render. The possible values are the same as the actual mixdown list. XML indicates the predefined mixdown in the encoder preset.

Slate

This is used in Transkoder in legacy VTRoll workflows. If the encoder has this option enabled, the automatically generated slate shots (bars, black, slate) will be included in the render.

PostProcess

Custom post-processing scripts.

File

This indicates the file path and the name of the file that will be created by the result render.

Double click on the File field in the table or single click the FILE PATH field in the encoder configuration pane to edit this value. A popup window will appear with the top half containing an Edit Render Path section for editing the file path and name using keywords and a browse button for modifying the render target folder.

The bottom half contains a Final Render Path section previewing the actual resulting render path and file name. The preview only shows the name of the first shot if using the Per Shot value of the File setting.

8.1.2. Forensic Watermarking

An invisible forensic watermarking is supported by many of the encoders in Transkoder. It is used to trace the origin of a media asset. NexGuard (Civolution) forensic watermarking is the watermarking method integrated into Transkoder.

To enable the forensic watermarking, set the WATERMARKING field in the encoder configuration pane to On, and specify the desired ID in the PAYLOAD ID field.

The forensic watermarking feature may only be enabled if the Transkoder workstation also has a NexGuard (Civolution) license installed.
Civolution NexGuard Watermarking Manager Interface

There is an interface in Transkoder for managing the communication with a remote NexGuard manager. Using this interface, the operator has access to a database of registered titles and thus can query the relevant key for the content. The encode process also communicates to the manager and reports the render progress and status.

First the manager needs to be enabled on the Tab Settings Page and the necessary address, user name and password needs to be entered. If this is not enabled the operator can manually enter the payload ID in the encoder settings on the P page.

Configuring the connection to the NexGuard Manager on the Encode tab
Configuring the connection to the NexGuard Manager on the Encode tab

When adding any encoder that supports invisible watermarking (Avid DNxHD and DNxHR, H.264, DPX, ProRes, IMF) please enable the WATERMARKING switch on the P page. The manager interface will be shown, with all selectors auto-populated based on information fetched from the server. The user can select item from the list under Title, Customer or Recipient or start typing in, and the system will auto-fill. When the text entry starts the list of items are automatically filtered in the drop-down list, so it should be easy to find any item even if the initial list of options is long.

The interface for the NexGuard Manager
interface for the NexGuard Manager

The Job Name field is populated based on the file name defined in the encoder, but the user can edit this. During the encode process this Job Name will be communicated back to the server along with the render status.

8.1.3. Timecode and Reel Name in Rendered Files

The majority of encoders support timecode metadata, and most of them also support reel name metadata as weel.

For any encoder that supports timecode metadata, it is possible to set which timecode metadata will be added to the result renders by specifying the desired timecode in the Rendered Media TC field on the Encode tab of the Settings Page. The available timecodes are the following:

Rendered Media TC Description

SourceTC

The timecode of the source clips currently on the timeline.

VTRollTC

The timecode used for legacy tape based dailies workflows.

RecordTC

The timecode as present on the actual project timeline.

RecordTC30DF

The timecode as present on the actual project timeline when being rendered with pulldown.

Manual

The timecode as specified in the Custom Encode TC field value.

For any encoder that supports reel name metadata, it is possible to set which reel name metadata will be added to the result renders by assigning the desired reel name to the Rendered Media Tape Name field on the Encode tab of the Settings Page. The available reel names are the following:

Rendered Media Tape Name Description

SourceContainer

Name of the folder media assets on the timeline.

VTRoll

It is used for legacy tape based dailies workflows.

CamRoll

This option may be used if the source file metadata contains the CamRoll values.

Filename

The file name of the media assets on the timeline.

SourceReelName

This option may be used if the source file metadata contains the ReelName values.

RenderFileName

It is the result render file name, as configured in the File fields of the Encode Page.

8.2. Encoder Categories

The Encode Page in Transkoder contains an Encoder List with the following encoder categories:

Encoder Category Description

Avid

High Definition related AVID encoders

AvidSD

Standard Definition related AVID encoders

QuickTime

Apple ProRes related encoders

Disc

DVD and Bluray related encoders

HEVC

High Efficiency Video Coding a.k.a. H.265 encoders

Web

Web related encoders

Apple

Apple related encoders

Broadcast H.264

H.264 a.k.a AVC related encoders

Image

Image related encoders

Audio

Audio only related encoders

DCI

DCP related encoders

IMF

IMF releated encoders

JPEG2000

JPEG2000 related encoders

AS02

AS-02 related encoders

Metadata

Metadata only related encoders

Others

Other type of encoders

Below is a detailed breakdown of the encoders listed in each category and their descriptions.

8.2.1. Avid Encoder Category

The Avid encoder category in the Encoder List includes codecs that support the MXF (OP-Atom) file type, which is the native file format for Avid Media Composer and contains the necessary metadata for an Avid Media Composer based workflow.

File types that can be selected from the FILE TYPE menu in the encoder pane are the following:

Avid File Types
MXF File

This OP-Atom file type will render separate, but cross-referencing .mxf files for both audio and video assets

Op1a File

This file type will render a single .mxf file containing both audio and video assets

MOV File

This QuickTime file type will render a .mov file containing both audio and video assets

Using any of the encoders in the Avid encoder category will result in the rendering of an ALE file, which is a plain-text .csv type file containing a comprehensive list of the metadata available for the project. For more information on configuring the ALE file, please refer to the Encode Settings section of the Settings Page. The Rendered Media TC field, also in the Encode Settings section, allows for configuring which TC to use as the main timecode for the rendered files.

All the encoders in this category, with the exception of the Legacy SD encoders, support either 16- or 24-bit audio bit depth. The DNxHR encoders provide an automatic bitrate adjustment depending on the desired frame rate.

By default, all Avid encoders use legal range (aka video range) for color processing, but may be modified to full range. Transkoder uses full range for all internal processing and will only remap to video range during rendering if the encoder setting is configured as video range.

It is possible to cycle between the different encoders of a particular family such as DNxHD and DNxHR by selecting the desired encoder from the TYPE field in the Encoder configuration pane on the right-hand side.

If files need to be placed into Avid Interplay after rendering, please configure the Interplay related fields as described in the Avid Interplay Setup.

Avid Encoder Category

Avid DNxHD (Digital Nonlinear Extensible High Definition)

Encoder Name/Family

Project Format

Pixel Format

Note

720p

720p 59.94fps

8-bit/10-bit YUV 4:2:2

145Mbit/220Mbit

HD

1080p 23.976fps/24fps/59.976fps

8-bit/10bit YUV 4:2:2

36Mbit - 220Mbit

HD

1080p 23.976fps

10-bit RGB 4:4:4

440Mbit

Avid DNxHR (Digital Nonlinear Extensible High Resolution)

Encoder Name/Family

Project Format

Pixel Format

Note

DNxHR

2K/UHD/4K 23.976/24/25/29.97/48/50/59.94fps

8-bit/12-bit YUV 4:2:2

DNxHR LB/SQ/HQ/HQX

DNxHR

2K/UHD/4K 23.976/24/25/29.97/48/50/59.94fps

12-bit RGB 4:4:4

DNxHR 444

Avid Legacy SD ( Standard Definition)

Encoder Name/Family

Note

DV25

Used for Avid Legacy DV SD workflows

Audio

Audio only (AvidAudio) rendered into a separate .mxf file

QuickTime Avid PAL/NTSC DV

Used for Legacy Avid PAL/NTSC SD workflows

Avid Uncompressed

Encoder Name/Family

Project Format

Pixel Format

Note

1:1 10bit RGB

1080i 59.94fps

10-bit RGB 4:4:4

Used for archival purposes

Avid XDCAM

Encoder Name/Family

Project Format

Pixel Format

Note

XDCAM 50mbit

1080p 23.976fps/24fps/25fps

8-bit YUV 4:2:2

Avid XDCAM 50mbit

8.2.2. AvidSD Encoder Category

The AvidSD encoder category is a list of lower than HD resolution encoders which only support the MXF (OP-Atom) file type that is used as the native file format for Avid Media Composer. They are mostly NTSC and PAL format types with together with a 1080p proxy format. This is useful for workflows which require high levels of compression of both audio and video to deal with lots of source material such as reality shows.

AvidSD Encoder Category

Encoder Name/Family

Project Format

Note

H264 1.5Mbps

PAL 25/NTSC 29.97

352x288/352x240

10:1

PAL 25/NTSC 29.97

720x592/720x496

20:1

PAL 25/NTSC 29.97

720x592/720x496

15:1s

PAL 25/NTSC 29.97

352x296/352x248

10:1m

PAL 25/NTSC 29.97

288x296/288x248

800kbps

Proxy 1080p

480x270

8.2.3. QuickTime Encoder Category

The QuickTime encoder category is a list of Apple ProRes and associated Quick Time encoders which all use the MOV (Quick Time) file type which is the native format of Final Cut. To get a comprehensive list of the metadata available for the project, add the desired encoder (Final Cut X or Final Cut 7 format XML) from the Metadata Category in the Encoder List.

The Avid NTSC DV and Avid PAL DV encoders are the two exceptions. They are optimized for Avid Media Composer compatibility. Please use the NTSC DV and PAL DV encoders if not using an Avid workflow for maximum compatibility.

Apple ProRes Encoder Configuration Pane Options

If a 23.976 fps project needs to be delivered as 59.94i, the PULLDOWN field value needs to be adjusted to the desired setting. Possible values are 2:3, 2:3:3:2, 3:2, and Off. The 3:2 pulldown pattern is the one most commonly used. It is only possible to enable pulldown for 4:2:2 encoders and doing so will make the result render interlaced. The FRAME TYPE field allows for the selection of either interlaced field 1 or interlaced field 2 when deciding on the dominant frame, meaning whether interlaced field 1 or interlaced field 2 will be displayed first.

If using pulldown with a dropframe enabled timecode, and there is a need to start the interlace cadence, meaning to set the start of the first interlaced frame at a particular time, the KEEP START TC field value should be set to On, and the desired start time should be entered in to the PDSYNC TC field.

To enable the forensic watermarking, set the WATERMARKING field in the encoder configuration pane to On, and specify the desired ID in the PAYLOAD ID field.

The forensic watermarking feature may only be enabled if the Transkoder workstation also has a NexGuard (Civolution) license installed.

By default the ProRes encoders map all the incoming metadata (that is metadata from the source file, visible in the Shift+U Metadata Inspector) to the header of the output file. This can be disabled using the Copy Metadata option.

All the encoders in this category support a choice of either 16 or 24 bit audio bit depth which can be set in the AUDIO BITDEPTH field in the encoder configuration pane.

The byte order of the audio may also be selected in the AUDIO ORDER field. The possible options are LittleEndian, BigEndian, LittleEndianV2. LittleEndianV2 is only to be used when there is a need to maintain legacy compatibility.

The Apple ProRes encoders support PCM, AC3 and E-AC3 audio types, and the Audio Type field may be used to select the desired type. If AC3 or E-AC3 is selected, the audio bitrate may be set with the AUDIO BIT RATE field.

The MULTI TRACK field is used to enable or disable the placement of audio tracks into either multiple individual tracks or a single track with multiple channels respectively.

The Range field is used to flag the signal range of the encoded file. Options are None, Full, and Video. This setting only effects metadata and changing it will not apply any range conversion.

To turn on HDR processing, select the desired HDR processing mode from the MASTERING field. Possible options are Rec2020_PQ_1000, P3D65_PQ_1000 or custom defined modes. The Framing field value also needs to reflect the project framing so that Transkoder can correctly calculate the light levels, etc needed for HDR processing. Possible options are Full frame, Pillarbox (1.33), Side Matted (1.66), Letterbox (2.35), Letterbox (2.39), Letterbox (2.40), and Matted (1.85). A PDF report containing HDR processing and Lighting information may be generated by selecting the PDF value for the HDR REPORT field.

The ProRes codec has an internal encoding using 12 bit YCC images, but for faster encoding speed the user can "feed" it with 10 bit images. To change this adjust the BIT DEPTH setting of the ProRes encoder.

QuickTime Encoder Category

Apple ProRes YUV Color Space

Encoder Name/Family

Project Format

Pixel Format

Note

Apple ProRes 422 Proxy/422LT/422/422HQ

SD-8K+

8-bit/10-bit YUV 4:2:2

Visually lossless for 4:2:2 material.

AMAZON UHD HDR ProRes 422 HQ (Auto)

UHD

8-bit/10-bit YUV 4:2:2

Apple ProRes 422HQ (Auto) which is configured to AMAZON’s UHD HDR requirements.

Apple ProRes RGBA Color Space

Encoder Name/Family

Project Format

Pixel Format

Note

Apple ProRes 444

SD-8K+

8-bit/10-bit/12-bit with 16-bit for Alpha Channel RGBA 4:4:4:4

Visually lossless for 4:4:4:4 material.

Apple ProRes XQ

SD-8K+

8-bit/10-bit/12-bit with 16-bit for Alpha Channel RGBA 4:4:4:4

Visually the same as Apple ProRes 444 with almost double the data rate to support HDR images.

Quick Time Legacy SD ( Standard Definition)

Encoder Name/Family

Resolution

Note

PAL DV/NTSC DV

720x576/720x480

Used for Legacy PAL/NTSC SD workflows

Avid Avid PAL DV/NTSC DV

720x576/720x486

Used for Avid Legacy PAL/NTSC SD workflows

8.2.4. Disc Encoder Category

The Disc encoder category is a list of encoders which are used for creating disc based media such as DVD and Blu-ray discs.

The DVD authoring workflow in Transkoder was created for dailies review allowing for the creation of a simplistic dailies reivew dvd. Transkoder is NOT a DVD mastering tool.

DVD Generation

For generating a DVD ISO image, first, the desired clips need to be rendered in to MPG deliverables on Encode Page.

Once the MPG deliverables have finished rendering, bring up the Finalize Window and specify DVD as the value for the ENCODER TYPE field.

In the Source field, specify the source of the DVD, which can be either the actual Timeline content, or any Transfer Date.

The DVD may be made to play automatically by specifying On for the value of the AutoPlay field.

If Transfer Date is specified for the value of the Source field, the desired Transfer Date may be selected from list of possible dates in the Date field. Since the DVD authoring workflow is intended for dailies work, the Date field will not be automatically populated with any possible transfer dates when specifying Transfer Date for the value of the Source field in Transkoder.

A simple autogenerated menu may be added by specifying On for the value of the Menu field.

The generated clips may either remain unsorted, or be sorted according to the method specified in the Sorting field. Possible values are the following:

  • Unsorted

  • Scene/Take

  • CamRoll/Take

  • Scene/Slate/Take

  • Scene/CamRoll/Take

  • CamRoll/Scene/Slate/Take

The Final Name field is used for entering the ISO image name together with the .iso file extension. If no name is defined for the DVD, then timeline name or the date is added to the DVD ISO file name in the following manner:

  • DVD_VT001.iso where VT001 is the VTRoll name.

  • DVD_A67B.iso where A67B is the tape/labroll name.

  • DVD_Screener1.iso where Screener1 is the screener name.

  • DVD_20120123.iso 20120123 is the transfer date.

After the DVD finalization process is finished, the ISO file is placed in the "<project name>\<transfer date>\generated" folder, or the folder named in a custom defined render path.

BluRay Generation
Disk Encoder Category
Encoder Name Resolution Note

MPG

720x480

DVD MPG

MPG 4:3

720x480

DVD MPG

M2V

720x480

M2V MPEG

M2V 4:3

720x480

M2V MPEG

BluRay

1920x1080

BLURAY

8.2.5. HEVC Encoder Category

The HEVC encoder category is a list of encoders which use the High Efficiency Video Coding (HEVC) codec. It is also called H.265, or MPEG-H Part 2.

HEVC encoded results of any custom resolution may be rendered by setting the resolution to Auto. The default encoding is variable bitrate. Constant bitrate encoding may be set by specifying identical values for the BITRATE and MAXBITRATE fields.

The PERFORMANCE field is used to specify the desired encoding performance and file size versus image fidelity. Available values are Fastest, Fast, Balanced, Better quality, and Best quality, with the highest performance, smallest file size and lowest image fidelty for the Fastest value to the slowest porformance, largest file size and highest image fidelity for the Best quality value.

The NETFLIX HEVC QHD TS encoder uses a 10-bit 2160p setting with the possibility to encode in HDR. The MASTERING field is metadata used to specify the color primary of the mastering display.

The Mastering field is metadata only and does not do any color primary conversion. The field value must accurately reflect the actual color primary of the input picture asset.

Rendering HDR will result in a HDR report and light level statistics being gathered. For more information on HDR and IMF, please refer to the chapter titled High Dynamic Range (HDR).

The Framing field is used to specify the aspect ratio of the active image area for use in calculating the accurate light level statistics for HDR results and as needed metadata by the IMF packaging standard.

Setting the High Tier field value to On will enable HEVC Level 5 High Tier encoding. It is set to Off by default.

HEVC Encoder Category

Encoder Name/Family

Color Depth

Note

MP4

8-bit

480p/720p/HD/QHD

MP4 10 bit

10-bit

480p/720p/HD/QHD/QHD HDR

TS

8-bit

480p/720p/HD/QHD

TS 10 bit

10-bit

480p/720p/HD/QHD

NETFLIX HEVC QHD TS

10-bit

QHD

MKV

HD/QHD

8.2.6. Web Encoder Category

The Web encoder category is a list of encoders which use a variety of encoding methods to create preview or proxy media with high compression rate, suitable for Internet based streaming / download. Some of these templates are based on the specification of dailies review systems and have limited configurability.

The x264 opensource h264 encoder is an alternative encoder to the H264 encoder and is more configurable in this encoder category. The DAX, PIX, and COPRA encoders are automatically rendered with the necessary sidecar metadata used in their relative systems. The PIX encoder will even generate an executable file that can be used to upload the media assets to the PIX servers. For the PIX encoder it is possible to add TC information into the metadata by setting the Time Code For Web field value to On in the Encode Settings section of the Settings Page. VP9 was initially used mostly for YouTube, but is now also used by an increasing amount of websites including Wikipedia, and Netflix which is aiming VP9 at mobile users.

Web Encoder Category

Encoder Name/Family

Project Format

Note

COPRA4

576p/1080p

Used for the COPRA dailies online collaboration system.

DAX|D3

360p/720p/1080p

Used for the DAX dailies online collaboration system.

H264

540p/720p/1080p

H264 used for generic dailies online collaboration systems.

PIX

720p/1080p

Used for the PIX dailies online collaboration system.

MpegTSH264

540p

H264 is inserted into an MPEGTS container.

vp9

720p/1080p

Used by YouTube, Wikipedia and Netflix mobile.

X264

720p/1080p/UHD

Opensource alternative to H264 and is a bit more configurable. It can be used wherever H264 is not a hard requirement.

8.2.7. Apple Category

The Apple encoder category is a list of encoders for AppleTV, iPad, iPod, and the ProRes. The AppleTV, iPad, and iPod encoders are all H264 set at specific Apple compliant resolutions. There are versions with preset 1920x1080 and Auto resolutions. The 1920x1080 versions will automatically render Full HD 1920x1080 without having to add a resize node to the pipeline on the Node Page. The Auto versions will render at the same resolution as the input or source material.

The Apple ProRes encoders support any resolution from SD to 8K and beyond at full resolution. The Apple ProRes 422 Proxy encoder is intended for low data rate offline workflows with full resolution video at 30% of the data rate of Apple ProRes 422. The Apple ProRes 422LT encoder is intended for workflows where storage and data rate are important considerations with a data rate of 70% of Apple ProRes 422. The Apple ProRes 422 encoder’s image quality is almost as good as Apple ProRes 422HQ, but at only 66% of the data rate. The Apple ProRes 422HQ encoder is visually lossless for 4:2:2 material. The Apple ProRes 444 encoder is visually lossless for 4:4:4:4 material. The Apple ProRes XQ encoder is visually the same as Apple ProRes 444 with almost double the data rate to support HDR images.

Apple ProRes Encoder Configuration Pane Options

If a 23.976 fps project needs to be delivered as 59.94i, the PULLDOWN field value needs to be adjusted to the desired setting. Possible values are 2:3, 2:3:3:2, 3:2, and Off. The 3:2 pulldown pattern is the one most commonly used. It is only possible to enable pulldown for 4:2:2 encoders and doing so will make the result render interlaced. The FRAME TYPE field allows for the selection of either interlaced field 1 or interlaced field 2 when deciding on the dominant frame, meaning whether interlaced field 1 or interlaced field 2 will be displayed first.

If using pulldown with a dropframe enabled timecode, and there is a need to start the interlace cadence, meaning to set the start of the first interlaced frame at a particular time, the KEEP START TC field value should be set to On, and the desired start time should be entered in to the PDSYNC TC field.

To enable the forensic watermarking, set the WATERMARKING field in the encoder configuration pane to On, and specify the desired ID in the PAYLOAD ID field.

The forensic watermarking feature may only be enabled if the Transkoder workstation also has a NexGuard (Civolution) license installed.

By default the ProRes encoders map all the incoming metadata (that is metadata from the source file, visible in the Shift+U Metadata Inspector) to the header of the output file. This can be disabled using the Copy Metadata option.

All the encoders in this category support a choice of either 16 or 24 bit audio bit depth which can be set in the AUDIO BITDEPTH field in the encoder configuration pane.

The byte order of the audio may also be selected in the AUDIO ORDER field. The possible options are LittleEndian, BigEndian, LittleEndianV2. LittleEndianV2 is only to be used when there is a need to maintain legacy compatibility.

The Apple ProRes encoders support PCM, AC3 and E-AC3 audio types, and the Audio Type field may be used to select the desired type. If AC3 or E-AC3 is selected, the audio bitrate may be set with the AUDIO BIT RATE field.

The MULTI TRACK field is used to enable or disable the placement of audio tracks into either multiple individual tracks or a single track with multiple channels respectively.

The Range field is used to flag the signal range of the encoded file. Options are None, Full, and Video. This setting only effects metadata and changing it will not apply any range conversion.

To turn on HDR processing, select the desired HDR processing mode from the MASTERING field. Possible options are Rec2020_PQ_1000, P3D65_PQ_1000 or custom defined modes. The Framing field value also needs to reflect the project framing so that Transkoder can correctly calculate the light levels, etc needed for HDR processing. Possible options are Full frame, Pillarbox (1.33), Side Matted (1.66), Letterbox (2.35), Letterbox (2.39), Letterbox (2.40), and Matted (1.85). A PDF report containing HDR processing and Lighting information may be generated by selecting the PDF value for the HDR REPORT field.

The ProRes codec has an internal encoding using 12 bit YCC images, but for faster encoding speed the user can "feed" it with 10 bit images. To change this adjust the BIT DEPTH setting of the ProRes encoder.

Apple Encoder Category

Apple Embedded Devices

Encoder Name/Family

Project Format

Note

AppleTV

720p/1080p

H264

iPad

576p/720p/1080p

iPod

640x360

Apple ProRes YUV Color Space

Encoder Name/Family

Project Format

Pixel Format

Note

Apple ProRes 422 Proxy/422LT/422/422HQ

SD-8K+

8-bit/10-bit YUV 4:2:2

Visually lossless for 4:2:2 material.

AMAZON UHD HDR ProRes 422 HQ (Auto)

UHD

8-bit/10-bit YUV 4:2:2

Apple ProRes 422HQ (Auto) which is configured to AMAZON’s UHD HDR requirements.

Apple ProRes RGBA Color Space

Encoder Name/Family

Project Format

Pixel Format

Note

Apple ProRes 444

SD-8K+

8-bit/10-bit/12-bit with 16-bit for Alpha Channel RGBA 4:4:4:4

Visually lossless for 4:4:4:4 material.

Apple ProRes XQ

SD-8K+

8-bit/10-bit/12-bit with 16-bit for Alpha Channel RGBA 4:4:4:4

Visually the same as Apple ProRes 444 with almost double the data rate to support HDR images.

8.2.8. Broadcast H.264 Category

The Broadcast H.264 encoder category is a list of high bitrate, high quality encoders that use the H.264 codec and conform to various broadcast standards. It is also called Advanced Video Coding (AVC), MPEG-4 Part 10, or MPEG-4 AVC. These presets are often 10- or 12-bit formats, thus are not supported by some consumer players, such as Windows Media Player or QuickTime Player.

The DirecTV 10bit MPEG-TS encoder uses AC3 encoded audio to provide for dual audio streams of up to 5.1 channels and stereo. In the encoder configuration pane, an .scc format closed caption file is also supported to source title tracks by specifying CC as the value in the CCSOURCE field. For transport stream multiplexing, in the TSMUX TYPE field, there is a new, more reliable method that can be chosen by selecting the Ver2 value, while Ver1 is the legacy version provided for backward compatibility.

To enable the forensic watermarking, set the WATERMARKING field in the encoder configuration pane to On, and specify the desired ID in the PAYLOAD ID field.

The forensic watermarking feature may only be enabled if the Transkoder workstation also has a NexGuard (Civolution) license installed.

The XAVC encoder is a Sony developed recording format that uses level 5.2 of the H.264/MPEG-4 AVC. The CLASS field, found in the encoder configuration pane, is used to specify the desired resolution, encoding bitrate, and whether the bitrate should be constant or variable. The available values range from HD Class50 CBG to 4K Class 480 CBG, with VBR options for 4K resolutions. The COLOR SPACE field is used to set a metadata flag. Available otions are Rec.709, Rec.2020 HLG, Rec.2020 PQ, and P3 D65 PQ. Forensic watermarking is also available for the XAVC encoder by enabling the WATERMARKING field.

The Panasonic AVC Ultra encoder is a 4K format that uses up to the 4:4:4 Intra Predictive profile of the H.264/MPEG-4 AVC standard. The CLASS field is used to specify the desired resolution and encoding bitrate. The available values range from 8-bit 4:2:0 720p-1080p AVCProxy to 12-bit 4K 2160p 4:4:4 AVCUltra4K.

Broadcast H.264 Encoder Category

Encoder Name

Pixel Format

Note

HD/QHD H264 10bit

10-bit

5 Mbps/10 Mbps/15 Mbps HD/QHD MP4

HD/QHD H264 10bit

10-bit

5 Mbps/10 Mbps/15 Mbps/120 Mbps HD/QHD TS

DirecTV 10bit MPEG-TS

10-bit

120 Mbps QHD TS that complies with DirecTV standards.

XAVC

RGB 8-bit/10-bit/12-bit 4:2:0/4:2:2/4:4:4

HD-UHD MXF

Panasonic AVC Ultra

YUV/RGB 8-bit/10-bit/12-bit 4:2:0/4:2:2/4:4:4

720p-UHD MXF

8.2.9. Image Category

The Image encoder category is a list of encoders which render each frame to an individual image file in various formats.

File sequence numbering may be customized by editing the value of the File Sequence Numbering field on the Encode Settings section of the Settings Page. Possible values are the following: Start with 0, Start with 1, Start with custom, From source, Source TC, and Record TC. If choosing the Start with custom value, the desired starting value needs to be set in the adjacent File Sequence Custom Start field.

The file name and path may be further customized by clicking on the …​ button of the File Path field on the encoder configuration pane and specifying the desired naming pattern in the Edit Render Path field. Keywords and leading zero padding may be specified in this field. The default leading zero padding is set to 7 and is written as %07d with d meaning decimal based integer value.

For the DPX encoder, it is possible to enable the use of the source DPX image header metadata by setting to On the Use Source DPX metadata field on the Encode Settings section of the Settings Page. If this setting is enabled, the source DPX image header is copied over to the result DPX image unchanged. If it is disabled, Transkoder generates the appropriate metadata from scratch.

To enable the forensic watermarking, set the WATERMARKING field in the encoder configuration pane to On, and specify the desired ID in the PAYLOAD ID field.

The forensic watermarking feature may only be enabled if the Transkoder workstation also has a NexGuard (Civolution) license installed.

For the EXR encoder, it is possible to set the desired compression method by specifying it in the Compression field on the encoder configuration pane. The methods affect the file size, rendering time and system resource utilization during playback. The possible compression methods are the following:

Compression Method Compression Type Description

Uncompressed

Uncompressed

No compression will be used.

RLE

Lossless

Run Length Encoder is a simple lossless image compression format, and is best used for encoding images mostly made up of sequences for repetitive data, such as computer graphics, logos, etc.

ZIPS

Lossless

A deflate type compression method used on individual scanlines. It is not based on the popular ZIP file format.

ZIP

Lossless

It is an extension of the ZIPS compression method used on 16 scanlines at a time. It is the best method for computer rendered images that do not have added film grain.

PIZ

Lossless

It is a combined wavelet and Huffman compression method, and is the best method for images with film grain.

PXR24

Lossy/Lossless

It converts data to 24-bit using rounding, and then uses deflate compression. It is lossless for 16-bit floating point and 32-bit integer type images, and is slightly lossy for 32-bit floating point type images.

B44

Lossy/Lossless

This compression method compresses blocks of 4x4 pixels from 32 bytes into 14 bytes. Images of 16-bit floating point type are compressed, but 32-bit integer and 32-bit floating point type images are uncompressed. Compressed file size is identical for all images at a given resolution, making it easier to plan file storage requirements and process real-time playback.

B44A

Lossy

It is an extension to the B44 type compression method, in which 4x4 pixel blocks with identical values are compressed into 3 bytes instead of 14 bytes. It is a more efficient method for imgages with large uniform areas.

DWAA

Lossy

It is a JPEG-like compression format, which compresses 32 scanlines at a time, and provides for balancing the requirements of image fidelity and size.

DWAB

Lossy

It is an extension of the DWAA compression method used on 32 scanlines at a time.

For a more detailed explanation of EXR technical specifications including compression formats used, please refer to OpenEXR Technical Introduction pdf.

The EXR encoder also supports various metadata injection methods specified in the Inject Metadata field on the encoder configuration pane. The available injection methods are: All, Custom or Off.

Off

No metadata will be embedded.

All

The encoder provides the complete list of metadata extracted into the EXR header. It is the same set of attributes as listed in the Shift+U metadata inspector. In the unlikely case of missing width and height metadata in the source, Transkoder automatically adds it as well.

In this mode, if the source file is an EXR file, the source header metadata is copied to the target header as is.

For ARRI type source media assets, enabling metadata injection with the All setting will enable the use of ARRI provided metadata mapping as long as the Arri Metadata Bridge setting is enabled in the Arri Camera section of the Settings Page.

Custom

When using the Custom metadata injection method, the encoder will look for an XML file that maps the metadata types and values into OpenEXR metadata. The advantage of such an approach is that you are allowed to predefine your metadata names, which can be based on the client’s naming conventions. Only the attributes defined in the mapper XML template will get exported, so it is up to the user which subset of available metadata will be injected.

The XML files need to be placed in the C:\Program Files\OSD2018\pymodules\scripts\EXRMetadata\ folder by default, but this folder may be changed by specifying a custom folder path using the Metadata Path field on the encoder configuration pane. If a valid path is set, the injection will look for the XML metadata files inside the custom folder. If the path is invalid, the encoder will revert to the default path.

The name of the XML file is based on the source camera format, which is is displayed on the left-hand side of the Node page in brackets. The XML file must be named according to the following syntax: <cameraformat>_metadata.xml. The value used for <cameraformat> must be lower case and must not contain spaces.

Example

If the camera format is ARRI RAW, the name of the metadata mapper XML would be arriraw_metadata.xml. For more details please check the examples that get installed under the EXRMetadata folder above.

The injection workflow supports additional metadata that are not from the source. In order to define custom metadata, the ForceInject attribute value needs to be set to True in the XML file. This will override the existing source metadata value if present. In the example below, <Aspect> will get exported as 2.0, overriding the value coming from the footage. ForceInject will add the metadata if it does not exist in the source.

Sample EXR Metadata OpenEXR XML file
<Aspect AttributeType="float" AttributeNameInTranskoder="Width" ForceInject="True">2.0</Aspect>
<Attribute1 AttributeType="int" ForceInject="True">1</Attribute1>
<Attribute2 AttributeType="float" ForceInject="True">2.0</Attribute2>

The JPEG2000 encoder is divided into two groups: DCI compliant and IMF compliant. For more information on the specifics of each, please refer to the DCI Encoder Category and the IMF Encoder Category.

Image Encoder Category

Encoder Name/Family

Project Format

Resolution

Note

DPX

Same color space as input 10-bit/12-bit/16-bit

Same as result render

EXR 16-bit half

Same color space as input 16-bit floating point half

Same as result render

JPEG2000 DCI compliant

XYZ/YUV 10-bit/12-bit

2K scope/flat/full + 4K scope/flat/full

JPEG2000 IMF compliant

Mono/YUV/RGB 10-bit lossy/lossless

HD-UHD+

Including Rev (App2E+)

TIFF

Same color space as input 8-bit/16-bit

Same as result render

1FPS JPEG

Same color space as input Same color depth as input

360p

MOV (QuickTime)

8.2.10. DCI Encoder Category

The DCI encoder category is a list of encoder defaults of a JPEG2000 based encoder, which creates DCI compliant DCP packages. The CIE 1931 XYZ color space is used with 4:4:4 chroma sampling and 12-bit color depth.

Although it is possible to render DCI compliant packages manually by using the encoders in this category, a much more convienient method is to render DCI packages via the Finalize Window. For more information regarding DCP or IMF rendering, please refer to the DCP and IMF Basics chapter.

For general JPEG2000 encoder settings, please refer to the JPEG2000 Encoder Category section.

If the TRANSWRAP field value is enabled, which is the default, and if the requirements are met for transwrapping, no JPEG2000 encoding will be done: the source will just be rewrapped into new MXF container. For more information on transwrapping, including what conditions are required, please refer to the Transwrapping section of the DCP and IMF Basics chapter.

The GUARD BIT and DWTRESOLUTION fields represent JPEG2000 parameters and automatically have default values set, depending on which encoder has been selected. To ensure maximum compatibility during playback, it is recommended that all assets in a package have matching values. If other assets in the package have values that deviate from default, the guard bit and dwtresolution fields' values should be adjusted to match those.

If the POST FINALIZE field is enabled, Transkoder will automatically finalize the DCP or IMF package applying the values entered in the Finalize window, even if rendering with Ctrl+R. If you need to render only the assets, please disable the Post Finalize field.

To enable Dolby Atmos® for DCP packages, please set the ATMOS field value to Enabled, and make sure to add the .mxf containing the Atmos sync signal to the 14th audio track on the timeline. For more information, please refer to the Dolby Atmos Audio section of the Audio Handling chapter.

DCI Encoder Category

Audio only MXF

Encoder Name

File Type

Note

Audio MXF

MXF

This will render an audio only MXF file.

Picture MXF

Encoder Name

Resolution

File Type

Note

2K DCP Scope

2048x858

MXF

2K DCP Flat

1998x1080

MXF

2K DCP Full

2048x1080

MXF

4K DCP Scope

4096x1716

MXF

4K DCP Flat

3996x2160

MXF

4K DCP Full

4096x2160

MXF

8.2.11. IMF Encoder Category

The IMF encoder category is a list of encoder defaults of a JPEG2000 based encoder which creates IMF compliant packages.

For general JPEG2000 encoder settings, please refer to the JPEG2000 Encoder Category section.

The MAX BITRATE field is used to specify the maximum bitrate to be used for the encoder. The PROFILE field is used to specify the JPEG2000 Operating Level. The SUB LEVEL field is used to specify the JPEG2000 Operating Sublevel. The COMP TYPE field is used to specify the JPEG2000 Profile. The PIX TYPE field is used to specify the chroma subsampling and color bit depth.

To get a lossless result render, make sure to set the value of the SUB LEVEL field value to 0 and the COMP TYPE field value to IMF Reversible Profile. The MAX BITRATE field will automatically be disabled.

For more information regarding the possible field values and relationships between them, please refer to the Encoding Profiles chapter.

The COLOR SPACE field is metadata used to indicate the color space of the picture asset. Options Color3:HD709, and Color5:Rec2020 are Standard Dynamic Range color spaces, while options Color6:P3D65 (HDR), and Color7:Rec2020 (HDR) are Perceptually Quantized High Dynamic Range color spaces.

The COLOR SPACE field is metadata only and does not do any color space conversion. The field value must accurately reflect the actual color space of the input picture asset.

The MASTERING field is metadata used to specify the color primary of the mastering display. If left as the Default value, then this is controlled by the COLOR SPACE field value. Otherwise the desired color primary and nit (brightness) level must be specified with the MASTERING and MASTER BRIGHTNESS fields. Just like for the COLOR SPACE field, these values are metadata only and no conversion is done.

The QUANTIZATION is used to specify in metadata if the picture asset is using QE.1 (Legal range) a.k.a head range, or QE.2 (Full range) According to the IMF standard, all YUV encoded picture asscets are Legal range, while RGB encoded picture assets are almost alwys Full range. Occasionally it is necessary to create a Legal range RGB IMF package, such as the type used by the Warner Bros HDR preset. This is also a metadata only value and no conversion is done.

Rendering HDR will result in a HDR report and light level statistics being gathered. For more information on HDR and IMF, please refer to the chapter titled High Dynamic Range (HDR).

The Post Finalize field if enabled, will automatically finalize the DCP or IMF package using the current values entered in the Finalize window even if rendering with Ctrl+R. If only the assets need to be rendered, please disable the Post Finalize field.

The FRAMING field is used to specify the aspect ratio of the active image area for use in calculating the accurate light level statistics for HDR results and as needed metadata by the IMF packaging standard.

The DOLBY VISION field is used to embed Dolby Vision™ metadata into the IMF result renders. For more information, please see the High Dynamic Range (HDR) chapter section titled Dolby Vision Workflow.

If the Transwrap field value is enabled, which it is by default, and if the conditions are met for transwrapping, no JPEG2000 encoding will be done, the source will just be rewrapped into new MXF container. For more information transwrapping including what conditions need to be met, please refer to the Transwrapping section of the DCP and IMF Mastering chapter.

The CPLLIGHT field is used to specify whether to add the light level statistics to the CPL. This field is only relevant for HDR workflows. Custom values may be specifed in the MAXFALL override and MAXCLL override fields of the Advanced IMF Settings of the Settings Page.

The APPLICATION VERSION field is used to specifiy the relevant IMF standard that is to be rendered. Available options are App2 (legacy standard) and App2E+ (Current standard published in 2016). The default option is *App2E+. This is needed for IMF validation. For the IMF Cinema Mezzanine preset, the hardcoded option is App4 which was developed for cinema archival purposes. App4 uses 16-bit color depthRGB 4:4:4 chroma sampling with the XYZ color space to archive media with super high fidelity.

Colorfront is at the forefront of supporting new IMF application standards and developement work is started as soon as they are published, with actual implementation done almost immediately. Naturally, Transkoder is able to validate all new IMF application standards.

IMF Encoder Category

Audio and Video

Encoder Name/Family

Project Format

Resolution

Note

YUV Lossless MXF

YUV 8-bit/10-bit/12-bit 4:2:2

HD/QHD

App2

YUV Lossy MXF

YUV 8-bit/10-bit/12-bit 4:2:2

HD/QHD

App2

RGB Lossless MXF

RGB 8-bit/10-bit/12-bit 4:4:4

HD/UHD

App2

RGB Lossy MXF

RGB 8-bit/10-bit/12-bit 4:4:4

HD/UHD

App2

FOX SDR YUV UHD Lossy MXF

YUV 4:2:2

UHD

App2E IMF package which complies with FOX standards.

FOX RGB Lossless MXF

RGB 8-bit/10-bit/12-bit 4:4:4

HD/QHD

IMF package which complies with FOX standards.

FOX HDR Lossless MXF

Same as Result Render

APP2E+ IMF package which complies with FOX standards.

SONY YUV Lossy MXF

YUV 4:2:2

HD/QHD

App2 IMF package which complies with SONY standards.

SONY HDR

Same as Result Render

App2E+ IMF package which complies with SONY standards.

NETFLIX YUV QHD Lossy

YUV 4:2:2

QHD

App2E IMF package which complies with NETFLIX standards.

NETFLIX HDR

Same as Result Render

App2E+ package which complies with NETFLIX standards.

ARD YUV HD Lossless MXF

YUV 4:2:2

HD

App2 IMF package which complies with ARD standards.

IMF REV

RGB 10-bit 4:4:4

Same as Result Render

App2E+

WARNER SDR/HDR Service Master

Same as Result Render

App2E+ IMF package which complies with WARNER standards.

8.2.12. JPEG2000 Encoder Category

The JPEG2000 encoder category is a list of encoders that render results into the JPEG2000 image format. The JPEG2000 sequence, DCP and IMF encoders all share the same base settings to control the encode quality and size. The enoder settings may be configured on the encoder configuration page.

The Resolution field is used to set the resolution of the result render images. For DCP, it is based on the DCI standards (2K/4K scope, flat, full), while for IMF it is flexible up to the maximal resolution of the current IMF application. To make sure that no frames in the generated asset will exceed a certain maximum bitrate, specify the desired bitrate (Mbit) in the Bitrate field.

Encoding Strategies

The Mode field specifies which JPEG2000 rendering mode Transkoder will use to render the results. These modes vary in performance, as well as what quality or size constraint they attempt to meet. One should consider changing the rendering mode from the default, if the size of the generated files should be small, the compression quality should be as good as possible, and render speed is not a primary concern. The available rendering modes are Constant Quality, Single-Pass Average, Multi-Pass Average, and PSNR-based Average.

Constant Quality

The Constant Quality mode is the Standard JPEG2000 rendering mode that uses fixed quality settings for each frame. It is not an adaptive method, so it is the fastest option of all. The two main settings for this render mode are the following:

Quality

It specifies the JPEG2000 quality level to be used for the encode process. Higher quality setting will result slower rendering speed and higher overall bitrate. If the majority of the encoded frames hit the Max Bitrate (see below), it is pointless to increase this setting. If the result of the compression process is a relatively low bitrate file, due to lack of detail in the source images, one can increase the quality to achieve a higher bitrate output with better fidelity to the source.

Bitrate

It specifies the maximum data size per frame. Regardless of the quality setting or image content, no encoded frame will exceed this size limit. As this limit is achieved by discarding excessive chunks of data after encoding, it does not affect performance.

Single-Pass Average

The Single-Pass Average mode is an adaptive method, where the compression routine dynamically adjusts the quality level per frame. The goal of this approach is to converge to the Average bitrate setting. This is achieved by internally adjusting the encoding settings based on the already processed recent frames. Adjustable settings for this render mode are:

Average bitrate

It specifies the average bitrate of the asset that is the target of this adaptive algorithm.

Bitrate

It specifies the maximum bitrate limit.

Quality

It specifies the default JPEG2000 quality level to use for the encoding process.

Multi-Pass Average

The Multi-Pass Average mode is the slowest adaptive method. In the first pass, the encoder analyses each frame to determine how well they compress, and how much high frequency detail each frame has. Then, in the second pass, during the actual encoding process, it can choose the ideal quality level for each frame in such a way that the final bitrate of the output will converge to the "Average bitrate" setting while maintaining the highest possible quality. The multi-pass approach is capable of distributing the bitrate in an optimal way compared to the single-pass approach, where knowledge of the complexity for the full sequence is limited. Adjustable settings for this render mode are the following:

Average bitrate

It specifies the target average bitrate of the output of this adaptive algorithm.

Bitrate

It specifies the maximum bitrate limit.

Quality

It specifies the default JPEG2000 quality level to use for the encoding process.

PSNR-based Average

The PSNR-based Average mode uses a method that dynamically adjusts the quality of each frame, so it is as small as possible while maintaining a minimum visual fidelity to the original source. This fidelity is measured in PSNR, Peak Signal-to-Noise Ratio, with the target being between "PSNR" and "PSNR"+"PSNRTolerance" whenever it is possible.

The PSNR is one way of measuring the difference between the original and the encoded image. Larger PSNR values indicate a better match or higher similarity between images. Adjustable settings for this render mode are:

PSNR

which specifies the low PSNR limit, below which the render process increases the encoding quality if the PSNR of a frame indicates lower fidelity;

PSNRTolerance

which defines the top PSNR limit by PSNR + PSNRTolerance, where, if the image fidelity exceeds this limit, the quality setting is decreased resulting in more compressed frames; and

Bitrate

which specifies the maximum permitted per frame bitrate.

If the Generate PSNR field value is enabled, Transkoder will generate the PSNR data during the encoding, which gets saved into the render folder. This allows for the per-frame PSNR-based visual fidelity analysis of the encoded assets. Please note that this mode will affect the speed of the encoding, since decoding of the rendered frame is necessary as well.

By default, audio and video assets are generated, however, this behavior can be overridden by the Media Type field to render video or audio files only.

To enable the forensic watermarking, set the WATERMARKING field in the encoder configuration pane to On, and specify the desired ID in the PAYLOAD ID field.

The forensic watermarking feature may only be enabled if the Transkoder workstation also has a NexGuard (Civolution) license installed.
DCP Specific Settings

For DCP specific settings, please refer to the DCI Encoder Category section.

IMF Specific Settings

For IMF specific settings, please refer to the IMF Encoder Category section.

JPEG2000 Encoder Category

DCI

Encoder Name/Family

Resolution

Note

2K Scope/Flat/FUll

2048x858/1998x1080/2048x1080

J2KPSNR/J2KGPU

4K Scope/Flat/FUll

4096x1716/3996x2160/4096x2160

Mono 10bit Lossless

1920x1080

IMF

Encoder Name/Family

Project Format

Resolution

Note

YUV 10bit Lossless

YUV 10-bit 4:2:2

HD/QHD

YUV 10bit Lossy

YUV 10-bit 4:2:2

HD/QHD

RGB 10bit Lossless

RGB 10-bit 4:4:4

HD/QHD

RGB 10bit Lossy

RGB 10-bit 4:4:4

HD/QHD

REV

Same as Result Render

App2E+

8.2.13. Audio Encoder Category

The Audio encoder category is a list of encoders which render the project audio to various audio formats. The Mixdown encoder will use an xml defined stereo mix by default, while the Multichannel encoder will use source audio files directly just like the Same As Source option in the Audio Mixdown window. It is possible to overwrite these default mixes for both Mixdown and Multichannel encoders by specifying a different value in the MIX field on the encoder configuration pane.

For Example

Specifying the sameAsSource value in the MIX filed of the Mixdown encoder will make it behave identically to the Multichannel encoder.

It may be useful to think of the Mixdown and Multichannel encoders as though they were the same encoder with different names which are configured by default to reflect this difference in names. mized to use any audio mixdown, irrespective of the encoder name.

Any audio mixdown may be specified for any of the encoders irrespective of the encoder name. Please refer to the section titled Audio Mixdowns for more information on creating custom audio mixdowns.

All of the encoders create separate files Per Shot, with the exception of the Mixdown and MP3 encoders which create files Per Timeline.

Audio Encoder Category

Encoder Name

Bit Depth

File Type

Note

Avid Audio

16-bit / 24-bit

MXF

This is used for creating Avid Media Composer compatible audio files.

DCI Audio MXF

24-bit

MXF

This is used for creating separate audio MXF files for DCP packages.

Mixdown

16-bit

WAVE

This will create an automatic stereo mixdown of the audio found on the timeline in 16-bit audio.

Mixdown 24 bit

24-bit

WAVE

This will create an automatic stereo mixdown of the audio found on the timeline in 24-bit audio.

Multichannel

16-bit

WAVE

This will use the source audio tracks on the timeline to create a 16-bit multichannel audio asset.

Multichannel 24 bit

24-bit

WAVE

This will used the source audio tracks on the timeline to create a 24-bit multichannel audio asset.

MP3 Audio Mixdown

NA

MP3

This will create an automatic stereo mixdown of the audio found on the timeline in MP3 format.

8.2.14. AS02 Encoder Category

The AS02 encoder category is a list of AS02 compliant encoders. Rendering AS02 assets is a two step process. First, the picture and sound are rendered into separate MXF files using the standard (Ctrl+R) rendering process. Next, a master or parent MXF which references the separately rendered MXF assets needs to be created one folder level up by opening up the Finalize Window, specifying AS-02 as the value for the ENCODER TYPE field and clicking on the Generate button.

When creating AS02 assets for HBO, it is necessary to make that there are 12 audio channels on the timeline. Rendering will result in a separate picture, and 6 stereo audio MXF assets. Finalizing as described above will create the necessary hierarchical master MXF.

AS02 Encoder Category

Encoder Name/Family

Project Format

Resolution

Note

YUV Lossless MXF

YUV 4:2:2

HD/QHD

AS02

YUV Lossy MXF

YUV 4:2:2

HD/QHD

AS02

RGB Lossless MXF

RGB 4:4:4

HD/QHD

AS02

RGB Lossy MXF

RGB 4:4:4

HD/QHD

AS02

HBO YUV HD Lossy MXF

YUV 4:2:2

HD

AS02 IMF package which complies with HBO standards.

8.2.15. Metadata Category

The Metadata encoder category is a list of encoders which render the project metadata to various metadata formats.

The LUT type encoders are used to "bake-in" the color processing of the current timeline. This is used when the dailies colorist work is finished with the color processing, and the visual effects (VFX) work needs to match this result. The LUT is exported by the colorist, and imported in to the VFX workflow on top of the raw footage. It is also used when a specific look is designed by colorists for a motion picture. The LUT is exported for import onto the monitoring equipment for use during shooting.

LUTs may vary in size from 17x17x17 to 33x33x33 and the size is denoted in the encoder name.

The Node Pipeline XML may be exported by right clicking anywhere on node page and selecting the Tools  Load Pipeline from XML menu option in addition to using the Node Pipeline XML encoder. The Node Pipeline XML may be imported by right clicking anywehere on the node page and selecting the Tools  Load Pipeline from XML menu option.

Metadata Encoder Category

Miscelaneous Encoders

Encoder Name

File Type

Note

CDL

CDL

This will extract CDL data present on the Node page pipeline, such as from the CDL, Printerlight, and RGBlin nodes.

ALE

ALE

This will render into desired file and folder the Avid metadata used in Avid Media Composer based workflows. This metadata is also automatically generated when using the Avid Category encoders, so this encoder is useful if the metadata needs to be made available in a distinct file and folder.

PSNR

PSNR

This encoder is needed for PSNR analysis. For more information on how to conduct PSNR analysis, please refer to the QC chapter section titled PSNR Analysis.

Bitrate data

Bitrate

Alexa XML

XML

This is used to create a metadata file for use with pre-ALF-2 Alexa workflows.

FCP 7 XML

XML

This is used to create a metadata file for use with Final Cut Pro 7 workflows.

Copra XML

XML

This is used to create a metadata file for use with Copra online review based workflows. This metadata is also automatically generated when using the Copra encoder, so this encoder is useful if the metadata needs to be made available in a distinct file and folder.

Colorfront EMD

EMD

Colorfront’s own sidecar type of metadata, containing all of the metadata from the database and the color correction pipeline as well. Please refer to the section titled EMD (Express Metadata) of the Metadata Appendix chapter.

Node Pipeline XML

XML

This is used to create a direct copy of the native metadata used in the entire Transkoder color pipeline. It is an excellent method of backing up or the archiving of the color processing pipeline for any given project.

LUT Encoders

Encoder Name

File Type

Note

3D Mesh LUT 17

3DMESH

A3D LUT 17

A3D

Lustre 3D LUT 17

LUT

Autodesk 3D LUT 17

3DL

Cube 3D LUT 17

CUBE

Panasonic 3D LUT 17

LUT

3D Mesh LUT 33

3DMESH

A3D LUT 33

A3D

Lustre 3D LUT 33

LUT

Autodesk 3D LUT 33

3DL

Cube 3d LUT 33

CUBE

ACES CLF LUT

CLF

8.2.16. Others Category

The Others encoder category is a list of various encoders which do not fit into the other main categories. The CUSTOM H264 and CUSTOM MPEG TS encoders allow for extensive customization of most of the encoder parameters directly from the encoder configuration pane. An even deeper configuration with total control over the encoder parameters for these two encoders is possible by manipulating the encoder XML files.

The Transwrapping encoder is used for file sequence image assets such as DPX, TIFF, and EXR, or single file formats RED, Sony MXF RAW and ProRes encoded assets to allow for the direct copy of the picture essence from the source to the result without the need to reencode and with no loss in image fidelity. This is especially useful when only shorter segment of a longer source asset needs to be created for a result, such as when sending specific scenes to the Video Effects teams. It is very high speed almost matching the speed of direct copying. For more information on transwrapping please refer the the section titled Transwrapping in the JPEG2000 encoding part of the DCP and IMF mastering chapter.

The Frame Light Level encoder will only provide appropriate frame light level statistics if the source picture essence is (SMPTE ST-2084) Perceptual Quantizer and Rec.2020 based. Additionally, the encoder must be run on full length uncut clips.
Others Encoder Category

Video

Encoder Name

Color Space

Chroma Sampling

Color Depth

Resolution

File Type

Note

Hypercube

7680x4320

Hypercube

This generates media compatible with the 8K HyperCUBE player by rendering four quadrants of the 8K image into 4K 422 HQ ProRes files.

MPEG4 For External Muxing Wide

1024x576

H264

MPEG4 For External Muxing

768x576

H264

H264 LPCM 1080p

1920x1080

H264

The H264 LPCM 1080 encoder allows for the creation of 16-bit / 24-bit LPCM wave audio stream together with a Full HD picture.

BLU-RAY 1080P

1920x1080

BLURAY

DirecTV 10 bit MPEG-TS

10-bit

3840x2160

MPEG-TS

Custom H264

1280x720

H264

The Custom H264 encoder allows for a very thorough customization of the H264 encoder directly from the encoder configuration pane.

MPEG2 1080p

1920x1080

MPEG2

The MPEG2 1080p encoder allows for a customization of the MPEG2 encoder directly from the encoder configuration pane.

Custom MPEG TS

960x540

MPEG-TS

The Custom MPEG TS encoder allows for a customization of the MPEG2 encoder in a TS container directly from the encoder configuration pane.

VC2

1920x1080

VC2

This is also know as Dirac Pro.

FOX YUV HD Lossless Archive MXF

YUV

1920x1080

FOXIMF

This will create a losless Full HD result render which includes both picture and audio assets in a single MXF which is compliant with FOX specifications.

HBO Op1a MXF

1920x1080

HBOOP1A

This will create a Full HD result render which will match HBO specifications.

FOX YUV QHD Lossless Archive MXF

YUV

3840x2160

FOXIMF

This will create a result render which includes both picture and audio assets in a single MXF which is lossless QHD and matches FOX specifications.

FOX Proxy H264

1280x720

H264

This will create a result render which includes both picture and audio assets in a single MXF which is 1280x720 Proxy and matches FOX specifications.

RAW YUV 420 8/10 in 16bit

YUV

4:2:0

8-bit/10-bit in 16-bit

1920x1080

RAW

This will create a headerless RAW file containing YUV color space picture essence with either 8-bt or 10-bit color depth encoded in 16-bit chunks.

XDCAM MPEG HD422

YUV

4:2:2

8-bit

1920x1080

MXF

This will create a Full HD result render which will match Sony XDCAM MPEG HD422 specifications.

RAW YUV 422 10bit - V210

YUV

4:2:2

10-bit

3840x2160

RAW

This will create a headerless RAW file containing YUV color space picture essence with 10-bit color depth.

Netflix Proxy 720p 2.4Mbit

1280x720

H264

This will create a 720p Proxy result render which will match Netflix specifications.

File Related

Encoder Name

Note

TransWrap

This is used for file sequence image assets such as DPX, TIFF, and EXR, or RED and ProRes encoded assets.

CopyFileSeq

The CopyFileSeq encoder is used when it is needed to copy a result to some other arbitrary location in while it is being rendered.

Other

Encoder Name

Note

Frame Light Level

The Frame Light Level encoder is used to manually generate the frame light level statistics that are required for HDR 10 workflows and Dolby Vision. HDR and Dolby Vision based encoders autmatically generated this information.

Benchmarking pipeline

The Benchmarking pipeline is used to measure the theoretical maximum encoding capability of the workstation. No result rendering is actually done.

8.3. Custom Encoders/Deliverables

8.3.1. Save Custom Encoder Preset

You can define any custom encoding template for frequently used presets that you can apply in any of your projects.

  1. Open the Encode Page by clicking the P key.

  2. Click anywhere on the highlighted empty row to bring up the Encoder List window.

  3. Select the preferred encoder from the menu, and modify any setting in the configuration Finalize window for IMF IMP on the right side of the screen.

  4. Go to Render  Save Selected Encoder Preset to save it as a custom preset.

  5. Add the preferred name and click enter to save it.

You can find the saved custom preset in the customEncoders.xml file under the C:\Program Files\OSD2018 folder. You can make any changes in the template by modifying this XML file.

To delete any saved custom encoder template, open the Encoder List and click on Custom to see all custom presets. Click on the small delete label to remove it from the list.

Custom encoders in the Encoder List
Custom encoders in the Encoder List

8.3.2. Modifying the XML Template

You can also define custom encoders by modifying the XML template, for which you need to do the following:

  • Open C:\Program Files\OSD2018\pymodules\lib\encodeTemplates.xml to see what the default templates are.

  • Create a new custom version of the encode file (with the same XML structure as encodeTemplates.xml) and save it as C:\Program Files\OSD2018\customEncoders.xml. To make sure the new encode job names do not conflict with the default ones, you may use numbers over 100 (e.g. job100, job101, etc.).

<encode>
    <job100 type="EncodeJob">
        <Enabled type="int">1</Enabled>
        <Description type="string">INHOUSE_FORMAT</Description>
        <ShortDescription type="string">MPEG2 HD</ShortDescription>
        <Container type="string">M2V</Container>
        <EncoderName type="string">MPEG2</EncoderName>
        <Profile type="string">high</Profile>
        <Width type="int">1920</Width>
        <Height type="int">1080</Height>
        <VideoBitRate type="int">200000000</VideoBitRate>
        <AudioBitRate type="int">0</AudioBitRate>
        <ResultFileName type="string">x:/inhouse_gen/$TransferDate$/$UniqueID$.m2v</ResultFileName>
        <ShotsToDiffFiles type="int">0</ShotsToDiffFiles>
        <ResultName type="string"> </ResultName>
        <FrameRate type="string">23.976</FrameRate>
        <PostProcessName type="string">upld.bat "$ScriptPath$" $FilledFileName$</PostProcessName>
        <PostProcess type="string"></PostProcess>
        <Sound type="string">no sound</Sound>
    </job100>
</encode>
It is possible to override the location of this XML containing the additional templates by editing the Additional Templates field of the Encode Settings section of the Settings Page.

To customize the encoders added to a specific project, edit the project’s encode.xml under ..\<project name>\osddata\<project name> which contains the actual deliverables. It has the same structure as customEncoders.xml shown above.

8.4. Create Multiple Results on Node Page

The node page is a freely configurable rendering pipeline. Bring up the node page by pressing the Enter key, and then moving the mouse cursor to the right-hand side and pressing the A key to add a new result. To add a new node, move the cursor to the middle of the node page, and press the A key. The nodes can then be connected to the desired results. The grading and viewing templates can be saved by pressing Shift+T. Please refer to the Node Page chapter for more information on proper configuration of nodes. Results with various resolutions, burn-ins, color space, etc. can all be configured for parallel encoding by using the appropriate nodes.

8.5. Setting Up Custom Pre- or Post-Scripts

There are two options to launch external scripts upon encoding: upon creating each encoded file, or upon finishing the complete encode job.

8.5.1. Upon creating each encoded file

If you open the pymodules\lib\encodeTemplates.xml under the install directory you see all the default encoders users can use. The encoders may have a "PostProcessName" tag or a "PreProcess" tag defined, which refers to a script to be triggered upon rendering an individual clip. A good example would be either the DVD encoder or any of the Avid MXF encoders. These short names refer to factory default scripts defined in \pymodules\scripts\scriptsdef.xml under the install home, but direct refereces to external batch files or executables are also allowed.

Do not change any files installed by the application installer!

To add your own post script that runs after generating each file we recommend to define your own deliverable template first. Please read the section "Adding your custom deliverable format" first.

As soon as you have your own deliverable template and want to add your own post-processing, you need to follow these steps:

  • create a batch file with your script

  • You can pass metadata to the batch files by using keywords like $VTRoll$ or $TransferDate$. Please check the scriptsdef.xml for examples.

  • Add the script path to the encoder’s PostProcessName tag in the customEncoders.xml you defined your own format in. (Do NOT edit the "PostProcess" tag, it is a temporary container for system use!) An example would look something like this:

<PostProcessName type="string">customEmail.bat "$ScriptPath$" "$FilledFileName$" "$Length$"</PostProcessName>

8.5.2. Upon finishing the complete encode job

If you press TAB and open up the "encode" tab, there is an EncoderPostBatch parameter. Here you can enter your batch file (with full path) and can also pass parameters using keywords like $ClipName$, $LabRoll$. The metadata will be taken from the first shot of the rendered timeline.

8.6. Render Menu

The Render menu is used to specify what gets rendered and also start, stop and restart the render job. The specific menu options are the following:

Render Timeline

This will render the Timeline as is and can be started using the Ctrl+R keyboard shortcut as well.

Render Shot

This will render the shot at the playhead position.

Render Highlighted

This will render all the highlighted shots. (Use CTRL click to select multiple shots.)

Marked Clips

This is used to manually render clips marked while DCP or IMF finalizing, if during the finalizing process auto rendering was turned off.

Render Selects

This is used in a dailies workflow to render clips which have been marked as Select in the metadata.

Render Non-Selects

This is used in a dailies workflow to render clips which have not been marked as Select in the metadata.

Continue Rendering

This is used to restart a stopped render job.

Stop Rendering

This will stop the active render job and can be done using the Esc keyboard shortcut as well.

Reveal Rendered File

This option opens one or multiple standard operating system file browser windows showing the location of the recent rendered files or current render targets on the Encode page.

Save Selected Encoder Preset

This is used to save any encoder as a custom preset by selecting the encoder on the Encoder page (P) and clicking on this menu option. The preset will be saved into the CustomEncoders.xml file and will be appended to the bottom of the Custom encoder category when adding a new encoder.

Export Transkoder Receipt

This will export sample files needed for rendering using the REST API and can be done using the Ctrl+Shift+X keyboard shortcut as well. Please refer to the TKD API Guide for more information.

8.7. Background Rendering

In Transkoder because rendering is done with heavy GPU utilization, background rendering must be either done using a separate dedicated render node, or the user must enable rendering of the queued jobs after exiting the interactive application. Rendering in the background while Transkoder is running in the foreground on the same workstation is not supported.

If only one workstation is available for both interactive work and rendering, one can queue up multiple render jobs by first disabling the render queue icon (grey color indicates the inactive state) and then submitting the tasks with Alt+R.

8.8. Render Queue

The Render Queue is a list of current and upcoming render jobs that have been queued up for processing. To access the render queue, navigate to the following address: http://$central_db_ip_address/render/ where $central_db_ip_address is the IP address of the central DB server. If the local host is being used for the render queue, then the render queue service tray application may be used to view the render queue on the local host. To access the render queue from the server or workstation that it is actually running on, navigate to the http://localhost/render/ address.

If using a dedicated (linux) db server for the render queue, please refer to the Appendix section titled Central DB Configuration for step-by-step setup instructions.

9. DCP and IMF Basics

Transkoder is able to play back, generate, verify and analyze IMF (Interoperable Master Format) packages (i.e. IMPs) and Digital Cinema Packages (DCPs): original versions as well as supplementals (version files). While the generation of such packages is often a single-click process, there are many ways to customize both the encoding and the packaging process.

For a more detailed explanation of DCP concepts and terminology, please refer to Cinepedia.

The following chapters describe all the different aspects of rendering and package creation process. Furthermore, in the subsequent chapters, DCP and IMF detail the corresponding specific mastering tools, such as encryption and KDMs for digital cinema or HDR options for IMFs.

9.1. Setting up the Timeline

To create a DCP or an IMF package, the timeline needs to be set up reflecting the final composition, which means audio, video, and, if needed, subtitle or closed caption components need to be in sync. For DCPs, audio and video events need to be cut identically to reflect the reel structure of the output. IMF packages do not have this constraint.

9.1.1. Import CPL

You can import a CPL file (either IMF/IMP or DCP) from the load page, or from the Timeline  Import CDI or IMP CPL menu option. Import will generate a new timeline on the project page. The timeline’s name will be inherited from the corresponding PKL file’s content title metadata.

You can also import a CPL into an existing timeline. If the timeline is empty, Transkoder will insert the CPL data in it.

Importing DCP and IMP packages from menu
Importing DCP and IMP packages from menu

9.1.2. Reel Markers

Reel markers are in and out editing points on the timeline to define reels, i.e. segments of track files that make up a temporal portion of a clip to be played for the viewer. It is very common to have video and audio segments, such as slates and counters, that are supposed to be in the encoded MXF files, but not in the final composition defined by the CPL. For this, reels can be used. Anything outside of the reels, defined by reel markers, will not be played back when playing the final CPL.

Reels can be set up manually by defining in points by pressing the I key, and out points by pressing the O key, or the Timeline  Markers  Set Reel Markers In and Set Reel Markers Out menu items.

Reels can also be set up in the Batch Edit window, activated by Ctrl+E, by adding the start frame number (in point) and the end frame number (out point) to the given fields.

Reels can be edited by selecting them in the Reels track on the timeline. Click Ctrl+Backspace to delete the highlighted reel or reels.

Timeline with three reels and reel markers
Timeline with three reels and reel markers

9.1.3. Positional Markers

Positional markers are special labels that can be added to specific frames of the composition to specify certain elements on the timeline, such as FFOC for First Frame of Composition, or LFCL for Last Frame of Company/Production Logo. Positional markers are accessible through Timeline  Markers  DCI Markers for DCP packages and IMP Markers for IMF packages. By clicking any of these labels in the list, a positional marker will be added to the current frame of the timeline.

After adding positional markers to the timeline, they will be saved in the CPL of the composition. When importing a CPL, positional markers will also be added to the corresponding place.

In case there are reel markers on the timeline, only positional markers inside the in and out editing points will be considered when creating a package.
Adding positional markers to the timeline
Adding positional markers to the timeline

You can edit positional markers by selecting them in the Markers track on the timeline. Highlight multiple markers by clicking Ctrl while selecting them with the mouse. Grab the highlighted marker to move it to another position, or press Ctrl+Backspace to delete them.

.Editing positional markers on the timeline
Editing positional markers on the timeline

See below the list of positional markers. The list contains all IMP markers, but only markers in bold letters are for DCI as well.

Marker labels Description

FFBT

First Frame of Bars and Tone

FFCB

First Frame of Commercial Blacks

FFCL

First Frame of Company/Production Logo

FFDL

First Frame of Distribution Logo

FFEC

First Frame of End Credits

FFHS

First Frame of Head Slate

FFMC

First Frame of Moving Credits

FFOB

First Frame of Ratings Band

FFOC

First Frame of Composition

FFOI

First Frame of Intermission

FFSP

First Frame of Digital Sync Pop

FFTC

First Frame of Title Credits

FFTS

First Frame of Tail Slate

FTXC

First Frame of Textless Title Credits

FTXE

First Frame of Textless End Credits

FTXM

First Frame of Textless Material Segment

LFBT

Last Frame of Bars and Tone

LFCB

Last Frame of Commercial Blacks

LFCL

Last Frame of Company/Production Logo

LFDL

Last Frame of Distribution Logo

LFEC

Last Frame of End Credits

LFHS

Last Frame of Head Slate

LFMC

Last Frame of Moving Credits

LFOB

Last Frame of Ratings Band

LFOC

Last Frame of Composition

LFOI

Last Frame of Intermission

LFSP

Last Frame of Digital Sync Pop

LFTC

Last Frame of Title Credits

LFTS

Last Frame of Tail Slate

LTXC

Last frame of Textless Title Credits

LTXE

Last Frame of Textless End Credits

LTXM

Last frame of Textless Material Segment

FPCI

Fixed Point Candidate Insertion

FFCO

First Frame of Commercial Overlay

LFCO

Last Frame of Commercial Overlay

FFOA

First Frame of Action

LFOA

Last Frame of Action

FFDC

First Frame of Dubbing Credits

LFDC

Last Frame of Dubbing Credits

9.1.4. Building DCPs with Reels at Each TC Hour

Transkoder also supports a workflow where DCPs are created using timelines where reels start at hours. It is also possible to round-trip these DCPs, e.g. to import CPLs in a way that the MXF track files are automatically placed to the corresponding hour in the timeline based on the embedded timecode. This way, it is very easy to drop in inserts or apply changes to individual reels in a non-destructive way, when making iterations to the OV package, or when creating supplemental (VF) packages.

9.2. Finalize Window

The Finalize window is where the DCP and the IMF/IMP package settings, including audio, subtitles and metadata can be configured. To bring up the Finalize window, press the Shift+D key combination.

The Finalize window is a more elaborate configuration space of the Encode Page, where not only rendering parameters, but basic metadata and various package settings, as well as render strategies can be defined.

Transkoder - Finalize window
Transkoder - Finalize window

First, on the Settings tab, the encoder type has to be selected to determine what kind of package needs to be rendered. The Finalize window is primarily for generating DCP or IMF packages, but it can also create DVD and Blue-Ray discs, among others. Then, the package configuration has to be specified (on additional tabs in case of IMF packages).

Once package specifications are defined, rendering and package finalization may be started by pressing the Generate button. By pressing this button, Transkoder will check the media on the timeline, assets already rendered, and, if needed, executes the render process and finalizes the package.

For more information regarding DCP or IMF rendering, please refer to the DCP and IMF Mastering chapter.

9.3. Render Strategies

There are two important areas in the user interface to control the mastering process: the timeline itself to design the final composition, and the Finalize Window to drive the packaging.

There are different workflows for typical use cases that apply to IMF and DCP mastering as well. There are two types of packages to be rendered: complete Original Version (OV) and Version File (VF) packages. Version File packages (i.e. supplemental packages) are IMPs with playlists (CPLs) that contain assets from other packages. Original Version IMPs, on the other hand, are self-contained and have all assets inside the IMP.

The strategies in case of OV packages may be:

Individual MXFs

In case source is set to V1 track. The number of rendered files will equal to the number of clips on the timeline.

V1 track driven MXFs

In case source is set to V1 and selected tracks. The number of rendered files will equal to the number of clips on the timeline, but inserts on the V2 track will be rendered into the clips on the V1 track.

Per timeline MXFs

The whole composition will be rendered into one MXF file. The CPL will carry the metadata referring to reels, if there are any.

Per marker MXFs

Only reels will be rendered to as many files as there are reels on the timeline.

The strategies in case of VF packages may be:

Individual MXFs

The number of rendered files will equal to the number of inserts on the timeline.

One MXF with all changes per movie

In case source is set to V1 and selected tracks, one MXF file will be rendered for all the inserts, along with a CPL containing the corresponding metadata.

These different workflows may require specific configuration of the clips on the timeline, and the Finalize window also needs to be set up correctly. The following sections detail these strategy workflows and the corresponding configurations.

9.3.1. Original Version Packages

Original Version packages (i.e. OVs) are complete packages with all assets needed for the composition. When the timeline is set up, open the Finalize Window and set the Package type to Original Version. An OV is created either from some video - DPX or TIFF mostly - and audio master files, or by replacing parts of another OV.

OV from Master: Single Reel

When working with original masters to create a package, the timeline needs to be set up on the V1 video track, and in the Finalize Window the Source should be "V1 track". If the final package should consist of a single picture and audio segment, set the Strategy to Per timeline MXF. This is a typical approach for IMF packages. In this mode, all frames are encoded into a single video MXF file, just like audio into a single audio MXF file. If you want to create an Original Version IMF with multiple audio tracks, go to the IMF audio configuration section for more details.

Single reel OV mastering
Single reel OV mastering
OV from Master: Multi-Reel

For multi-reel packages set the Strategy to Individual MXFs to preserve the clip structure of the original timeline. Reel markers can be set up if the video and audio master contains parts, such as frame counter and 2-pop, that are not intended to be seen in the final result. Make sure to enable Reels in the Finalize Window. Without reel markers the reel/segment structure of the final result will be defined by the edit points of the timeline.

Multi-reel OV mastering
Multi-reel OV mastering
OV by Changing Another OV: Single Reel

Often the Original Version package to be generated is based on another DCP or IMP. In this workflow it is recommended to have the original timeline on the V1 track and add the new video replacements to the V2 video track. This can be easily done by loading an original CPL, saving it as a new timeline and adding a new, empty video track for the changes, by pressing Alt+V. A typical use case would be the localization of a trailer: having the English master version on V1, the user can add the localized replacements to V2 and also potentially replace the audio. In this scenario, in the Finalize Window the Source needs to be set "V1 and selected tracks". If the final package is meant to contain a single audio and a single video asset, the Per timeline MXFs render strategy needs to be selected in the Finalize window.

When rendering a timeline where parts of the source is coming from an already encoded package, either a DCP or an IMF package, Transkoder automatically detects for each frame, whether it needs to be encoded, or it can reuse the original essence. This is called automatic transwrapping.

Changing parts of an OV to create a new OV (single reel)
Changing parts of an OV to create a new OV (single reel)
OV by Changing Another OV: Multi-Reel

In case of working with multi-reel original packages, Transkoder can preserve the reel structure. DCPs by default are loaded with reel markers, meaning the complete assets are visible on the timeline, and reel markers are automatically set up corresponding to the composition. This allows an easy workflow to change the package without losing any part of the assets, such as 2-pop, that is essential to checking video and audio sync.

Regardless of using reel markers or not, by choosing the V1 track driven MXFs option, you can finalize a package that maintains the reel/segment structure of the original package.

Changing parts of an OV to create a new OV (multiple reels)
Changing parts of an OV to create a new OV (multiple reels)
Rendering Based on Reel Markers

There is a finalize strategy called Per marker MXFs. Using this strategy, the operator can control the clip structure of the final IMF by adding reel markers to the timeline. These markers will drive what frames need to be rendered into separate MXF video track files.

As seen in the example below, reel markers can be used to flexibly render a number of clips into one MXF. The first reel marker spans over the first three video events. These events will be rendered into one video MXF. The rest of the timeline is another marker indicating the second MXF’s in and out points.

Per Marker MXF render strategy
Per Marker MXF render strategy

9.3.2. Version File Packages

Version File (VF) packages provide a supplement to other DCPs or IMF IMPs. A VF defines a composition that uses assets from the package, as well as assets from other packages, typically an Original Version. Thus, the VF package can be used to fix or alter another package without the need of media duplication. VF packages are most often used to define localized versions of titles, potentially with parts of the video or the complete audio replaced, or subtitles added.

To create a Version File package in Transkoder, the user first has to load an OV package. It is advised to save it as a new timeline, not to destruct the OV timeline. Then, necessary edits need to be added to this timeline. To replace parts of the video, there are two options: either by adding the inserts to the V2 video track, or by directly editing the V1 track. The first approach has the advantage that the OV timeline remains intact. The proper approach needs to be selected from the Source popup menu in the Finalize window.

Regardless how video inserts are defined, there are two render strategies to choose from: rendering individual MXF files with the Individual MXFs option, or all changes rendered into a single MXF with the One MXF with all changes per movie strategy. The first approach will generate a new MXF asset for each video insert. Please note, the VF package will not contain the assets of the OV package on the basis of which it was created.

Supplemental mastering with inserts on second track, rendering individual MXF files
Supplemental mastering with inserts on second track, rendering individual MXF files
Supplemental mastering from an edited timeline, rendering individual MXF files
Supplemental mastering from an edited timeline, rendering individual MXF files

To minimize the number of asset files in the package, the user can choose to render all inserts into a single, concatenated MXF file. This asset will contain all the frames of the inserts, with no gap. It is the CPL that defines the composition and ensures the proper sections of the video asset is edited into the OV timeline.

Supplemental mastering from an edited timeline, rendering a single MXF file with all inserts
Supplemental mastering from an edited timeline, rendering a single MXF file with all inserts
Supplemental mastering with inserts on second track, rendering a single MXF file with all inserts
Supplemental mastering with inserts on second track, rendering a single MXF file with all inserts
Create Version File from VF Timeline

If generating a Version File package from a timeline that is a supplemental package itself, the Assetmap of the original (OV) package needs to be manually specified in order to create a new proper Version File.

Load Version File for Editing

When loading a VF for editing, it is necessary to either load an OV first, or set Transkoder to search for missing assets by setting the Search Missing VF Assets field to On, and also specifying the appropriate search level in the VF Search Path Levels field in the DCI and IMF sections of the Settings Page. The level selected here will specify how many levels above the VF asset folder (currently imported CPL) will be searched. This search is recursive, meaning it will search all sub-folders as well.

Create Version File Relative to OV

In the DCI and IMF sections of the Settings Page, it is possible to use the VF Asset Selection option to create a VF relative to the Original Version by selecting RelativeToCPL or RelativeToSelect.

RelativeToCPL

It compares the new timeline to the original CPL, and the packaging process will be drived by this comparison.

RelativeToSelectedIMF/DCI

The operator will always be prompted to select the OV Assetmap, and only new assets will be hard linked.

Use Case 1: Creating a VF from a VF timeline
  1. Create an English OV.

  2. Create a German subtitled version VF.

  3. Load German VF and make some editing changes.

  4. Create another VF like this, but without the English files.

Use Case 2: Renaming a VF
  1. Load a VF.

  2. Rename assets by refinalizing with different naming.

  3. Include the same track files in the new package as in the original VF.

9.3.3. Deliver Complete Packages by Merge

The CPL Merge Window is for delivering complete packages by merging multiple CPLs of OV and/or VF packages into one complete IMP or DCP.

A typical use case is creating a package from an Original Version and multiple Version File packages, e.g. various supplementals for different languages and subtitles.

Another use case is when we have a complete IMP or DCP with one OV and several VFs for different languages, and we want to create a separate package from only one (ore more) VF and send it without the Original Version. In this case, we select the desired VF (or VFs) to generate a package, thus allowing complete package delivery from supplemental (VF) packages.

The Include All Assets option in the CPL Merge Window
The Include All Assets option in the CPL Merge Window

First, open the CPL Merge Window with Alt+D. All the CPLs of the current project will be listed in the window. Then, select the CPL file(s) you want to merge. Selected packages will be highlighted in yellow. If you select a CPL of a single VF, the complete package will be generated by merging the assets of the selected Version File with the MXF files of the corresponding Original Version.

Below the CPL list, you can configure the following fields:

MasterTitle

Title will come from selected files' content title metadata, but it can be modified to any title of your choice

Issuer

Add name of issuer

Destination

Add a destination folder of your choice or leave as default

When enabling the Include All Assets option, Transkoder merges the CPLs of the selected packages, and all assets listed in the CPL, i.e. MXF and XML files, which are required for playback, will be delivered into the ProjectBasePath\Delivery\<Title> subdirectory. When this option is Off, however, only assets included in the assetmap(s) of the selected package(s) will be delivered into the folder, while other assets necessary for playback will only be referenced in the CPL.

On the bottom right side of the CPL Merge Window there is an option called Keep PKLs. When this option is On, in case of multiple PKL files, each of them will be delivered into the folder and the new package will have multiple PKLs. If this option is Off, PKLs will be merged, and the package will be generated with a single PKL file.

9.4. JPEG2000 Encoding

The JPEG2000 render engine is GPU accelerated and has a built-in transwrap feature to rewrap the assets if possible without deencoding and reencoding. It is also possible to manually fine tune the encoder settings. All that is necessary is to bring up the Encode Page, and add a JPEG2000 format encoder. For more information on the JPEG2000 encoder configurable options please refer to the JPEG2000 encoder category section.

9.4.1. Transwrapping

By default all JPEG2000 encoders in Transkoder attempt to transwrap the video essence on a per-frame level. This means if the source frame is some J2K media, the encoder checks if that image essence is compatible with the encode parameters, and if so there is no decoding and encoding but the same essence is used for the output. This automatic transwrapping can be disabled on the Encode Page encoder configuration pane by setting the Transwrap field value to No wrap.

The criteria for transwrapping in case of DCP:

  • There must be no nodes on the node page altering the image (with the only exception of the Subtitle node)

  • Resolution should match

  • Color space should be DCI XYZ

The criteria for transwrapping in case of IMF:

  • There must be no nodes on the node page altering the image (with the only exception of the Subtitle node)

  • RGB vs YUV color encoding should match

  • Resolution should match

  • Color sampling and Quantization should match

  • Lossless vs lossy profile should match

As an example Transkoder can transwrap extremely fast DCPs from/to SMPTE or Iterop as well as from/to encrypted or non-encrypted.

9.4.2. Validation

Both DCP and IMF IMPs can be validated. Open the Validation window by pressing CTRL+d and start the process by pressing the Start button. Upon completion a detailed validation report will be generated, open it by pressing the Open button. The detailed validation steps are described in the QC Tools chapter in the section titled DCP and IMF Validation.

9.5. Bit-rate and PSNR Analysis Tools

The video content of DCP or IMP packages is typically encoded using a lossy J2K algorithm. By adjusting the quality setting of the encode process, either globally or per-frame, one can create different video assets of different sizes and quality. Being able to analyze the per-frame size and image fidelity of the packages is essential to be able to deliver optimal DCPs or IMFs. Transkoder allows the user to analyze any J2K asset using bit-rate and PSNR graphs, regardless what hardware or software generated the media.

9.5.1. Graph Window

The Graph Window, which is opened by pressing CTRL+COMMA, allows the detailed inspection of the size and quality aspects of video assets, at a per-frame frame. When looking at bit-rate data the horizontal axis represent time - the play head indicator is also shown - and vertical axis correspond to the bit-rate values. That is, the required bits to encode a single frame. Higher values indicate larger corresponding image chunks.

Graph Window with Bit-rate Data
Graph Window with Bit-rate Data

On the right panel the exact bit-rate value is also shown for the current frame, as well as average, minimum and maximum values for the whole timeline. To inspect the bit-rate curve closely please hold CTRL and right-click drag the mouse up to zoom in and down to zoom out.

To be able to analyze the fidelity of the encoded frames in conjunction with the bit-rate data PSNR information is needed. This is a numerical representation where high values indicate close match, and low values indicate significant difference between the original and the encoded frames. Unlike to bit-rate data, PSNR data is not auto-generated for the MXF video assets as it requires extra decoding steps affecting the render performance. When rendering video MXFs one can turn on PSNR data generation on the Encode Page, but it may be generated separately as well. For details see section "Generate Bit-rate and PSNR Data for External Assets".

When PSNR and bit-rate data is loaded into the Graph View the size vs. quality balance of the encoded video asset can be inspected. The bit-rate graph above indicates how many bits are required to encode a frame (higher values indicate larger frames) while the PSNR graph below indicates the quality of the encoded frames (higher values indicate better match). Please note: Although a higher PSNR generally indicates that the reconstruction is of higher quality, in some cases it may not.

Graph Window with Bit-rate and PSNR Data
Graph Window with Bit-rate and PSNR Data
PSNR
Peak signal-to-noise ratio, often abbreviated PSNR, is an engineering term for the ratio between the maximum possible power of a signal and the power of corrupting noise that affects the fidelity of its representation. PSNR is most commonly used to measure the quality of reconstruction of lossy compression codecs (e.g., for image compression). The signal in this case is the original data, and the noise is the error introduced by compression. When comparing compression codecs, PSNR is an approximation to human perception of reconstruction quality. source: wikipedia

9.5.2. Analyze Transkoder Generated Assets

Upon encoding any DCP or IMF asset Transkoder saves the per-frame bit-rate of the files. Thus right after rendering, or when loading the assets to the timeline, the Graph window will automatically show the relevant data. To generate PSNR data the Generate PSNR switch needs to be turned on in the encoder parameters (P page).

The render performance is significantly hit when Generate PSNR is enabled, as the encode process needs to decode the encoded frames to be able to compare them to the original.

9.5.3. Generate Bit-rate and PSNR Data for External Assets

One can load any J2K encoded MXF asset to the timeline, add a Bit-rate metadata encoder to the P page and launch the render. This will generate a bit-rate file that is automatically loaded right after rendering.

To generate PSNR data create a timeline with two video tracks and load the reference / source images to one track, the encoded asset to the other. Enable the Timeline / Timeline Mode / Enable All Tracks As Source and then open the node page. You will notice that the other track is available as an input now in the node page. Add a PSNR node to the tree, and connect its two inputs to the two sources (representing the two video tracks). If multiple events are on the timeline press SHIFT+T to apply this node tree to all the shots. Now one can visually inspect the PSNR values by looking the the tool HUD and playing the footage, and also one can add a PSNR encoder on the P page and generate a PSNR data file by rendering the timeline. This resulting PSNR file will be associated with the clip on the V1 track, and will be automatically loaded upon render completion.

9.6. Inspect Metadata

All original file metadata for the current CPL file can be viewed by opening the Clip Metadata Inspector window with Shift+U. In the TL tab, you can examine all timeline metadata for CPLs (either IMP or DCP) in a structured way, such as annotation, package size, language, creation date, loaded date, audio channels and layout, standard, etc.

The Clip Metadata Inspector has four tabs:

  • Video: original file metadata for video tracks

  • Audio: original file metadata for audio tracks

  • D: Dolby Vision metadata values from XML

  • TL: metadata for CPLs

TL tab of the Clip Metadata Inspector window displaying metadata for CPLs
TL tab of the Clip Metadata Inspector window displaying metadata for CPLs

10. DCP

The Digital Cinema Package (DCP) is a collection of digital files sent to a cinema. Transkoder is capable to play back, generate, verify and analyze all the various DCP formats:

  • encrypted and non-encrypted

  • Interop and SMPTE

  • flat, scope or full image aspect

  • 2K or 4K resolution

  • standard and high frame rate packages

  • subtitles support includes MXF, PNG, XML InterOp (Texas Instruments CineCanvas) and XML SMPTE (D-Cinema SMPTE 428-7)

  • support for Close Captions (MXF SMPTE Timed Text)

  • animated subtitles including depth

  • 2D or stereoscopic 3D packages

  • new original version or version file (supplementary packages)

  • various audio configurations and Dolby Atmos® support

  • KDM generation, including forensic watermarking

10.1. Auto-Finalizing DCPs

First create a timeline with picture and audio elements and make sure the timeline itself represents a valid cinema package, e.g. all video events have corresponding audio events, and there is no audio cut point within the video events. Subtitle files can be imported into the master bin of the project just like any other media files, and these subtitle files can be dragged to the timeline creating subtitle tracks. These subtitle tracks can be edited just like any other media. The Subtitle option of the Finalize Window controls if the subtitle elements are included in the created package or not.

When using subtitles please make sure to have a subtitle node in the node pipeline to be able to see the subs or to render them into the picture. If the subtitles are not meant to be rendered into the video asset make sure to disable subtitles before auto-finalizing or rendering.

Picture and audio elements on the timeline may be already encoded, valid DCP assets - in this case the finalize process will not reencode them - or may be of any format (e.g. TIFF or DPX sequences, WAV files or any other sources ). The finalize process automatically encodes the events that do not comply the DCP specification to be generated. If no reel markers are used, the reels in the DCP will correspond to the video events on the timeline.

After the timeline is set up and it represents the DCP one wish to generate, the Finalize window should be opened by pressing Shuft+D. The configuration of the DCP to be generated can be done in the finalize window. The following options can be set also from the window. Territory, Rating, Movie Title, Annotation text, Issuer, Creator, Studio, Facility, Audio and Subtitle language. These parameters will be presented in the DCP xmls. DCI standard DCP Title is generated automatically with the given metadata, but user can also edit it manually.

Finalize window for DCP
Finalize window for DCP

The audio layout option sets the metadata describing the audio layout of the DCP. By default it is on Auto, meaning based on the number of channels Transkoder will use the most common corresponding audio layout. For instance finalizing a DCP with 8 channels of audio in auto mode will set the audio layout to "7.1 (SDDS)", however the operator may need to override this if the audio configuration is actually "5.1 with HI and VI".

If all parameters are set up as desired one can press the "Generate" button to launch the finalize process. If necessary, Transkoder will start the encode process to render and wrap the needed MXF files. As soon as the rendering is done, the necessary folders and XML files will be generated. The DCP .xml files (PKL, CPL, Assetmap and Volindex) along with the symbolic links of MXF files (and subtitle files if present) will be generated and copied into <sourceFolder>/dcp/<Title> subdirectory. Upon finishing the process, Transkoder will automatically import the CPL to the project Bin window.

10.2. Generating DCP Assets Manually

If the user needs to control the render process manually, or custom encode settings are needed, one can add manually a DCP encoder to the Encode Page and adjust the encode parameters. One can launch the render process by pressing Ctrl+R. Next, the DCP can be finalized in the Finalize Window without further encoding. Alternately, the Finalize window can be used to auto-render the necessary assets with the current encoder. To add a DCP encoder, open the Encode Page, and click on an empty slot of the encode table from the DCI group, and add the required DCP encode preset.

The DCP specific encode parameters are the following:

DCI Type

Selects Interop or SMPTE DCP package type.

Encrypted

Encryption mode of the package.

Atmos

Enables Dolby Atmos® compatible packages.

10.3. Asset and XML Naming

The DCP Finalize Window contains a drop-down selection to control the asset naming in the final package. The options are the following:

ContentTitle

MXF track file names will be derived from the package Content Title

Default

Very similar to the Content Title preset, this is the current factory default.

SameAsSource

This will preserve the names of the source .mxf files as determined on the Encode Page. If this needs to be customized, set the Mix field value for the IMF encoder to the desired mix and modify the file name in the Audio Path field as described in the IMF’s audio section.

UUID

This will use the UUID in the .mxf file names.

It is possible to adjust these naming patterns or introduce new ones by editing or adding new XML files to the folder …​\pymodules\dci\NamingPatterns. These XMLs contain the naming schemes for all kinds of files in the package, the file name itself is the name of the preset shown on the UI. A sample is shown below.

Naming Pattern XML definitions for DCP
<pattern type="Container">
  <user type="Container">
	<!-- user defined default/forced/custom keywords and values | [a...z,A...Z,0...0,_,-]	max 170 char-->
	<ext>mxf</ext>
	<!-- TKD keywords: $UUID$, $ContentType$, $Territory$, $Rating$, $MovieTitle$, $Issuer$, $AudioLanguage$, $TTLanguage$, $Supplemental$, $MediaType$, $ClipNumber$, $OriginalFilename$ -->
  </user>
  <assets type="Container">
	<Video type="string">$ContentTitle$_$ClipNumber#2d-1$.$ext$</Video>
	<Audio type="string">$ContentTitle$_$ClipNumber#2d-1$_audio.$ext$</Audio>
	<Atmos type="string">$ContentTitle$_$ClipNumber#2d-1$_atmos.$ext$</Atmos>
	<Subtitle type="string">$ContentTitle$_$ClipNumber#2d-1$_sub.$ext$</Subtitle>
	<Caption type="string">$ContentTitle$_$ClipNumber#2d-1$_cc.$ext$</Caption>
  </assets>
  <cpl>CPL_$UUID$.xml</cpl>
  <pkl>PKL_$UUID$.xml</pkl>
</pattern>

10.4. Color Transformations

By default Transkoder is configured to work in the P3 color space. This means all non-P3 image inputs need to be converted to the P3 color space using the CSC node. The DCP encode process will auto-apply the floating points precision P3 to DCI-XYZ color conversion, as well as the DCP decode process will generate a P3 image directly. For example when working with standard 709 video images a CSC node needs to be added to the image processing pipeline with the following settings:

  • InputCurve: Gamma 2.4

  • InputColor: Rec709

  • OutputCurve: Gamma 2.6

  • OutputColor: P3

Alternatively one can work in the native DCI XYZ color space, eliminating any color transformation. This is the recommended workflow for those working with DCDMs already mastered in the DCI-XYZ color space. For this please disable the DCI/XYZRGBConversion in the Settings Page. In this mode images will be displayed as XYZ on the computer monitor and on the SDI output, and all non-XYZ sources need to be mapped to the DCI XYZ color space for encoding.

Other supported color spaces are the following:

  • Canon C.Gamut

  • V-Gamut

  • VLog (new in the output options)

  • CanonLog2 (new in the output options)

10.5. Dolby Cinema DCP Mastering

Transkoder can add the Dolby EDR transfer function metadata to SMPTE DCP CPL files indicating PQ color encoding, which is necessary for Dolby Cinema playback. To enable this, set the value of the Dolby EDR transfer function option on the Finalize Window to PQ10K.

Dolby Cinema Flag
Dolby Cinema Flag

This option enables the corresponding metadata indication in the CPL file. The source DCDM should be already in the proper PQ color space, mastered for 108 nits.

10.6. Stereo 3D DCPs

Please follow the next steps to create a Stereo 3D project and DCP in Transkoder:

  • Create a new project

  • Enable the stereo flag (TAB/Base -Stereo = On) right after creating the new project

  • Add left and right eye media to the V1 and V2 tracks of the timeline window. Make sure they are in sync.

  • Enable All tracks as source timeline mode in the Timeline Menu

  • Open the Finalize window and check if the stereo flag is enabled

Note: the Stereo flag is ENABLED, but it is always greyed out in the Finalize Window.

When working with 3D stereo MXF video track files there is only one track on the timeline, left and right eye are not separated on the V1 and V2 tracks. Because of this creating supplemental packages with video inserts should be done in two steps: first the operator needs to render the new 3D stereo MXF assets on a timeline with V1 and V2 containing the left and right eye DCDM. Then the rendered new assets can be used to edit them in to create a supplemental package.

10.7. KDM Generation for Encrypted DCPs

Auto-generated KDMs

Upon generating encrypted DCPs Transkoder auto-generates KDM files as well: one self-KDM which allows Transkoder to open the DCP and also there will be KDMs generated for every certificate (.pem file) listed in the KDM Certificates folder. KDMs generated for these user defined certificates will be placed in the deliveredKDM folder withing the project home, organized such that each CPL will have it’s corresponding subfolder. The location of these folders, as well as the validity period of the KDMs are configured on the DCI tab of the Settings Page.

By default the self-KDM is issued for a global Transkoder certificate, thus it will allow any Transkoder system to play the DCP, and corresponding private key file is also present on the file system. This should be used only for testing. To generate a self-KDM that certifies the workstation only, dongle-locked certification is needed. For this please refer to section "Installing Dongle Locked DCI Certificate".

KDM Generator Window

Open the KDM Generator window from the Window  KDM menu or by pressing CONTROL+SHIFT+d keyboard shortcut. In this window the operator can choose any CPL from the project, configure the validity period - either by start and end date, or by start date and length and set the (UTC) timezone. Before pressing the generate button the certificate file(s) need to be selected, either a single one or a group of certificates.

Trusted Device List for KDMs

The KDM Generator window also includes a Trusted Device List. A TDL defines peripheral equipment connected to the digital cinema server, which may also have certificates for themselves. This feature allows the generation of KDMs for cinema players with dual projection system.

KDM Generator Window
KDM Generator Window

10.7.1. Importing Encrypted DCI Package

To successfully import and playback an encrypted DCP package with KDM, place the KDM into the master KDM folder of the project, which is by default <projectHome>/KDM. Encrypted packages, or assets generated by Transkoder do not necessarily need a self-KDM to be playable in the system: during encoding Transkoder saves the keys for each asset as an encrypted key file - called cfdigest - in a folder within the project home directory.

10.7.2. Installing Dongle Locked DCI Certificate

To generate a DCI certificate that is valid only for a current workstation with the Colorfront dongle connected, contact support (support@colorfront.com), and request a DCI certificate. Please include in your email your system ID (same as the one used for licensing) and the HASP Dongle ID. These can be accessed from the File  DCI  DCI License Info menu option. Colorfront will send a package of certificate files, which may be installed via the File  DCI  Import DCI License menu option or manually in the following way:

Manual installation of Dongle Locked DCI Certificate
  1. Make sure that the HASP dongle is plugged into the workstation and use the Import DCI License menu option to import the dci license file to the dongle.

  2. Copy the certificate pem files to <install_home>\pymodules\dci\certs\production

  3. If the plugged dongle contains a valid DCI license then the certificate under <install_home>\pymodules\dci\certs\production is used. This certificate is tied to the host and previously issued KDMs work only on the original workstation.

KDMs can be generated using the leaf.signed.pem from <install_home>\pymodules\dci\certs\production.

If the dongle is not plugged in or does not contain a valid DCI license, the preinstalled example certificate is used from <install_home>\pymodules\dci\certs\example. This certificate is not tied to the host so your issued KDMs will work on any transkoder machine.

KDMs can be generated using the leaf.signed.pem from <install_home>\pymodules\dci\certs\example.

The same certificate is used as the signer and server certificate.

10.7.3. Exporting DCI Certificate

To export the Transkoder station’s certificate navigate to File  DCI  Export DCI Certificate menu option.

If your Transkoder operates without a valid production certificate it will export the EXAMPLE certificate which is not secure. The exported file name will contain the type of the dci license (production/DEMO).

10.7.4. PKL Signature

Transkoder is capable of digitally signing DCPs. An advanced setting called Write Signature is added to the DCI section of the Settings Page to control the signing for the merged PKL when delivering DCPs.

Transkoder - Write Signature setting in the Settings Page
Write Signature setting in the Settings Page

The setting has three options:

On

This option will sign both encrypted SMPTE and Interop DCPs. The <Signer> metadata will be included at the end of the PKL file.

EncryptedOnly

This option will only sign encrypted SMPTE DCPs.

Off

The generated DCPs will not be signed.

If the option Keep PKLs is turned On in the DCP Merge Window, Transkoder will not write signature.
Some older players may have problems with signed Interop DCPs.

10.8. Global DCP Settings

The following DCP parameters can be adjusted globally in the DCI Settings section of the Settings Page.

  • DCI Encryption

If ON, then new DCI deliverables added to the Encode Page will be initialized accordingly. This setting can later be adjusted on the Encode Page.

  • DCI Type

Defines the default standard of the DCI deliverables - SMPTE or INTEROP. This can be manually adjusted on the Encode Page for each encoder.

  • DCI Encryption Key Home

Location of the - encrypted - key files for encrypted content generated by Transkoder. Please note that without these files a KDM is needed to open the encrypted assets.

  • Import After Finalize

Enables the automatic ingest of generated DCPs.

  • Package Asset Creation

To save storage Transkoder by default does not copy the media assets into the generated DCP folders but creates hard links to the files. This way the packaging is faster, no extra storage is needed.

  • Package Asset Naming

If this is set to ContentTitle, Transkoder will rename the asset files upon DCP generation so they match the content title text. If set to AssetUUID, TKD will rename the assets based on their UUID.

  • KDM Home

Defines the project specific path where Transkoder looks for the KDMs when playing back a DCP.

  • XYZRGBConversion

If this is set to On, Transkoder makes conversion from DCI XYZ to P3 RGB in case of DCP decoding and from P3 RGB to DCI XYZ in case of DCP encoding.

11. IMF

The Interoperable Master Format (IMF) is a single, interchangeable master file format based on J2K compression to minimize storage requirement, allowing flexible versioning of the content. A single IMF Package, i.e. IMP may contain a variety of versions of the same content: different language, aspect ratios or playlists. It is loosly based on the DCP concept.

Transkoder supports the following IMP features:

  • HD, QHD and 4K

  • Application 2, Application 2 Extended+, Application 4, Application 5 and ProRes

  • YUV and RGB

  • supporting wide range of frame rates for packages: 23.976, 24, 25, 29.97, 30, 48, 50, 59.94, 60

  • 10, 12 and 16bit

  • Lossy and Lossless profiles

  • available deliverables meet the requirements for the following IMF specifications: NETFLIX, DISNEY, FOX, SONY, HBO, HULU, WARNER.

11.1. Finalize IMF Package

After creating a timeline with the image and audio content, an IMF package (IMP) can be generated by using the Finalize Window. The rendering of the video assets is GPU accelerated. For optimal performance, multiple high performance GPUs are recommended. The finalization process consists of generating the necessary XML files for the IMF package and creating the package itself.

There are several strategies to render the MXF resources into an IMF package. You can render clips on the timeline separately, the whole timeline composition, or only reels defined by reel markers. There are separate render strategies for creating an Original Version and a Version File package. Read about this in detail in the Render Strategies chapter.

The Finalize Window for IMF consists of multiple tabs with various configurable settings. The tab parameters are presented in the IMP’s XMLs. See the following sections for details.

IMF/IMP parameters can be adjusted globally in the IMF section of the Settings Page

11.2. Settings Tab

The first and and most essential tab of the Finalize Window is the Settings tab. Elementary settings can be adjusted here, such as the render strategy, title and encoding format.

Pressing Generate will generate an IMF package. The IMP’s XML files, i.e. PKL, CPL, OPL, Assetmap and Volindex, along with the symbolic links of the video and audio MXF files will be generated and copied into the ProjectBasePath\generated\IMF\IMP\<Title> subdirectory. After the generating process, Transkoder will automatically import the CPL timeline.

Transkoder - Finalize window
Settings Tab in the Finalize Window configured for IMF packages

The options of this tab are the following:

Package

Two types of packages: Original Version or Version File

Source

Two types of sources: V1 track or V1 and selected tracks

Strategy

See details on rendering strategies for each package type.

Reels

Turn On to include reels in the package

Content

Content type, e.g. feature, trailer, short, etc.

Territory

Target country/territory

Rating

Content rating for selected territory

Movie Title

Add any title

Issuer

Entity issuing the package

Content Originator

Creator of original content

Title

Package identifier, generated automatically with the given metadata, but also editable by user. Press Calculate to refresh text.

Annotation

Annotation text in package. Press Calculate to refresh text.

Audio

Language of audio

Layout

Audio layout controlled by Audio tab

Subtitle

Subtitle description if available

Version

Option to add version number

OPL

Receipts can be defined here

The application selector selects the Application type for different encoding, which are the following:

  • App#2 J2K: First generation broadcast HD IMF specification

  • App#2E+ J2K: Latest JPEG2000 based application, supporting UHD, HDR, etc.

  • App#4 Cinema: Cinema Mezzanine format based on XYZ J2K encoding

  • App#5 ACES: ACES image sequence in MXF

  • RDD45 ProRes: SMPTE RDD 45, Application ProRes

After the application is selected, a different list of encode recipes will appear in the Format pop-up menu. There are various encoding formats for specific IMF specifications such as for Netflix, Disney or Fox, with different profiles (lossy or lossless), color encoding systems (RGB or YUV), image resolutions (HD, QHD, UHD) and dynamic ranges (SDR or HDR) among others.

Parameters of the Encoder tab are also switched in line with the selected application.

11.3. Encoder Tab

All J2K encoding parameters are mirrored to the Finalize Window, so any custom setting and metadata attribute can be set and validated in one place. By selecting an IMF preset using the Format drop-down list the parameters are auto populated. It is still possible to use the Encode Page to adjust the encoder settings. The Encode Page J2K settings will automatically be synchronized with the Finalize window.

Encoder tab of the Finalize window to control IMF encoding
Encoder tab of the Finalize window to control IMF encoding

11.4. Color Tab

Metadata about color encoding and information on the HDR mastering display - which defines the brightness and color range of the content - is not available on the Color tab of the Finalize window. The Mastering setting is only available in HDR mode, when the Color Space is either Color-6 (P3D65) or Color-7 (Rec2020). If the Mastering field value setting is Auto, the mastering display properties are driven by the Color Space setting. The operator can inspect and if needed override these parameters.

Color tab of the Finalize window to control color encoding and mastering display metadata
Color tab of the Finalize window to control color encoding and mastering display metadata

Additional options on this tab:

  • Include Light Level Statistics in CPL: controls whether the MaxFALL and MaxCLL HDR10 metadata is included in the CPL.

  • Overwrite Light Levels Statistic: when On the MaxFALL and MaxCLL numbers can be overridden by the user. See this chapter for details.

  • Generate HDR PDF Report: optional HDR PDF report generation with gamut and brightness validation.

11.5. Audio Tab

Transkoder provides an interface to specify multichannel audio (MCA) metadata for IMF mastering.

Audio tab in the Finalize window to control MCA audio labeling
Audio tab in the Finalize window to control MCA audio labeling

The first step is to set up the proper audio configuration by using the Layout selector. This defines the number of audio groups, i.e. audio MXF track files to be rendered, and the channel configuration within each group. The layout options are the following:

Single Audio

Single 5.1 surround audio group

SameTracksAsSource

As many audio files as there are audio tracks on the timeline

Dual Audio

Two audio groups: 5.1 surround and a stereo

6StereoTracks

Six stereo audio groups

226AudioTracks

Three audio groups from which two is stereo and one is 5.1 channels

No Audio

No audio tracks on the timeline

Mono

Single audio channel

Stereo

Two separate audio signal channels

5.1 to Stereo

Converts 5.1 audio into stereo

The audio layout can also be initialized from the Audio Mixdown Window by selecting any mixdown from the left pane. The selected groups will define the audio tracks in the IMP to be created. For example, the default Dual Audio mixdown has two groups, so it will generate two audio tracks in the package. If the user wants to generate a custom IMF with three audio tracks, a new custom mixdown should be created with three groups.

When a new custom mixdown is composed in the Mixdown window, it will be automatically loaded into the layouts of the audio tab of the Finalize window. Then, when making any changes to this custom mixdown layout, by clicking on the RefreshLayout button next to the Layout selector on the audio tab, all adjustments will be automatically synced to here. These changes will also be reflected in the Loaded Layout field that displays the current audio layout of the timeline.

You can switch between audio groups using the Audio Group selector to inspect the metadata for each group (corresponding to MXF audio track files). All MCA information is audio group specific, so each audio MXF track file can potentially have completely different metadata set. The following MCA audio label parameters can be edited manually:

MCA Group Type

Soundfield group type, e.g. standard stereo, 5.1 SMTPE, 7.1 Film, etc.

MCA (Content) Title

Name of the overall program to which the audio essence track belongs

MCA Language

RFC 5646 language tag for spoken language, for this channel or soundfield group, e.g. en-GB

MCA Title Version

Version of the program to which the audio belongs, such as a specific cut of a movie

MCA Sub-Version

Sub-version of the program to which the audio belongs, such as a localized rendition that has the same cut as depicted in MCA Title Version

MCA Episode

Episode of a MCA Title

MCA Partition Kind

Partition kind of a complete program, such as “Part”, “Post Production Reel” or “Act”

MCA Partition No.

The position of the partition in the sequence of partitions that make up the complete program, such as “3” for Reel 3

MCA Content Kind

Description of the audio content kind, such as "Master"

MCA Element Kind

Technical description of the audio element, such as "Composite Mix"

In most cases, basic information, such as MCA title and sub-version should be the same for all audio groups, so make sure to first fill out these fields and press Apply to All so that any changes in MCA labels apply to each audio group.

The MatchSource button will set the audio configuration to match that of the timeline. If the MCA labels of audio assets in a track differ from the timeline metadata, the IMF package will fail on Photon validation. By pressing the Match Source button, MCA metadata will be extracted from timeline assets and applied to the whole timeline. Thus, the package will be automatically updated with MCA audio metadata, which is an essential requirement for IMF package validation.

In case there is an IMF on the timeline, or there are already rendered audio MXF track files, if the MCA labels in the files correspond to the MCA metadata in the audio tab of the Finalize window, Transkoder will generate a package without re-rendering.

By selecting a Group type, such as Standard Stereo, 3.0, 5.1 SPPTE, 7.0 Film, from the drop-down list, Transkoder configures the channel allocation displayed in the lower part of the window. These MCA tags can be edited manually for each channel in the bottom of the tab to create a custom configuration.

11.5.1. Inspect MCA Metadata

All MCA metadata for the current IMF file can be viewed by opening the Clip Metadata Inspector window. The audio tab contains all audio metadata, such as audio sampling rate, audio codec type, channel count, including the metadata for the MCA audio label parameters.

The audio tab of the Clip Metadata Inspector window displaying metadata for audio
The audio tab of the Clip Metadata Inspector window displaying metadata for audio

11.6. Version Tab

Version tab in the Finalize window
Version tab in the Finalize window

11.7. Naming Tab

The IMF Finalize window contains a dedicated Naming tab to control the asset naming in the final package making it possible to inspect and customize the naming of video, audio and subtitle track files as well as CPL files.

Naming tab in the IMF Finalize Window
Naming tab in the IMF Finalize Window
ContentTitle

MXF track file names will be derived from the package Content Title

FOX

This will name the .mxf files as required for FOX projects.

SameAsSource

This will preserve the names of the source .mxf files as determined on the Encode Page. If this needs to be customized, set the Mix field value for the IMF encoder to the desire mix and modify the file name in the Audio Path field .

UUID

This will use the UUID in the .mxf file names.

If multiple audio MXF files are generated each file can be assigned a custom naming pattern by selecting Group A, Group B, etc. By default all audio files use the Group A pattern, but this can be changed. By pressing the Save button the customized naming pattern can be saved for later use.

11.8. Sidecar Tab

During IMP mastering, sidecar assets, such as a text file containing a HDR report, images files for posters or stills, or any other materials, can be added to the package. An additional Sidecar Assetmap XML will be created that links these additional files to the composition. Simply drag and drop the files into the tab area, or on the bottom of the window click on the small ⊕ (plus sign) button to add files. Click the Clear button to remove all the added sidecar files at once.

Sidecar tab in the IMF Finalize Window
Sidecar tab in the IMF Finalize Window

Sidecar files can also be loaded from the original IMF package. For this, do the following:

  • Load the IMP containing any sidecar assets

  • Open the Finalize Window and go to the Sidecar tab

  • Click on the LoadFromCPL button

  • The sidecar tab will display all sidecar files that was attached to the orignal package

Loading sidecar files from CPL
Loading sidecar files from CPL

A single sidecar window, containing the same components as the sidecar tab, will be loaded when opening the CPL Merge Window with Alt+D, and then clicking on the Sidecar button.

Sidecar button on the CPL Merge Window
Sidecar button on the CPL Merge Window

11.9. Edit CPL

Transkoder allows the user to edit the CPL manually, which invalidates the package, and then fix the hash and size values to reflect the edited files. The FixXMLs button is used to to update all the hash and size values in the CPL, PKL and Asset Map XMLs.

Fix Hash of IMP XML files after manual edit workflow:
  1. Generate the package.

  2. Open the XML files and edit them if needed.

  3. In the Validation window run the Transkoder test. It is expected to fail.

  4. Press the FixXMLs button.

  5. Run the validation again. The package should validate.

Fix package after manual edit workflow
Fix package after manual edit workflow

11.10. Pre-Generate MXF files

MXF files for video and audio assets may also be created though the Encode Page by selecting an IMF encoder from the encoder list. The parameters of the IMF encoder can be configured on the encoder pane, on the right side of the screen. Then, pressing Ctrl+R will start the rendering process. The picture and audio resources at this point may contain extra head and tail. After pre-rendering the MXF track files with this method, either using foreground or background rendering, the user can open the Finalize Window and create the package instantaneously.

Encoder list on Encode Page for creating IMF packages
Encoder list on Encode Page for creating IMF packages

In general, package generation from the Finalize Window utilizes a "lazy" render strategy, meaning it only regenerates MXF track files when necessary. Any recent MXF files rendered from the project, or MXF track files on the timeline are automatically packaged if they conform to the specifications of the various tabs of the Finalize Window.

12. Audio Handling

Transkoder supports multichannel, and multi-mono broadcast .wav, and .mxf formats for standalone assets in addition to most native embedded audio formats.

12.1. Managing Audio Assets in the Bin Window

The Bin Window is the Transkoder GUI for managing media assets including audio assets.

12.1.1. Adding Audio Assets to the Bins

To add assets, such as clips, to bins via the file system, click on the small ⊕ (plus sign) button in the bottom left corner of the Bin Window, and a file system browse dialog will appear on the screen. Browse to the asset to be added, and click on the OK button.

Transkoder - Adding assets via file system
Transkoder - Adding assets via file system

If adding multi-mono audio assets, Transkoder will automatically recognize their relatedness, and they will appear as a single audio asset in the Bin Window. This feature may be disabled by specifying the Off value for the Group Multi File Audio field in the Advanced Settings section of the Audio Settings section on the Settings Page. With the setting set to Off the files will be displayed / treated separately.

Transkoder Multi-mono audio files will appear as a single audio asset in the Bin Window
Transkoder Multi-mono audio files will appear as a single audio asset in the Bin Window
Transkoder does not support > 4gb wav files.

12.1.2. Removing Audio Assets from Bins

To remove assets from bins, highlight the assets to be removed, and press Ctrl+Backspace while the mouse pointer is within the Bin Window, or click on the small ⊖ (minus sign) button in the bottom left corner of the Bin Window asset list.

12.1.3. Inspecting Audio Assets in the Bin Window

To hide or show audio assets in the Bin Window asset list, click on the Audio button on the top right-hand side. Click on it again to return to the previous state.

Transkoder Bin Window - Hidden Audio Assets
Transkoder - Bin Window - Hidden Audio Assets

The following fields appear when viewing Audio Metadata in Bin Window Details View:

Media Path

This is the media path to the audio file.

TCin

This is the starting TimeCode of the audio file.

Duration

This is the length in time of the audio file.

SoundRoll

This is automatically extracted from the broadcast metadata in the .wav file header, if present.

Samplerate

This is the number of samples of audio per second.

Name

This is the friendly name of the audio file.

Channels

This is the number of channels of a given audio file.

12.2. Audio on the Timeline

Audio Tracks added to the Timeline show up as a waveform and are displayed in a collapsed mode, by default, with all channels of a track combined into one and with A1..N, as the track label, where N represents the number of channels. Channels can also be displayed in an expanded mode with each channel shown as an individual track with A1,A2,A3…​, etc. as track labels.

To switch from the condensed view to the expanded view, double-click the A1..N track label, or to switch from the expanded view to the condensed view, double-click one of the A1,A2,A3…​ track labels.

Transkoder Timeline with Collapsed Audio Tracks
Transkoder Timeline with Collapsed Audio Tracks
Transkoder Timeline with Expanded Audio Tracks
Transkoder Timeline with Expanded Audio Tracks
For and in depth explanation on managing (audio) assets in the Bin Window, please see the Manage Assets within Bins chapter.

12.2.1. Managing Audio Assets on the Timeline

To add audio assets to the Timeline drag and drop them from a Bin to the Timeline or highlight the desired assets and press the V button. (Both methods behave the same way.)

When adding audio assets to a track on the Timeline with either the drag-and-drop method or the V button method, the inserted location will depend on whether the Timeline is set to POS or TC insertion mode. In POS mode, the audio asset will be placed at the current playhead position, and in TC mode, the audio asset will be placed according to the timecode metadata contained in the audio asset. To switch between POS or TC mode, double-click on the mode text which is located directly above the track list at the left hand side of the Timeline.

All audio assets added to a track must have the same number of channels.

To create a new audio track press ALT+SHIFT+v or select the Timeline  Add Track  Add Audio Track menu item.

To remove an audio track from the Timeline, select the Timeline  Del Track  Delete Audio Track menu item.

Right-click drag up or down in the middle of the audio track list on the left hand side of the Timeline to scroll the tracks up or down.

To group audio and/or video clips together, make sure to highlight the desired tracks, and select the Timeline  Group  Group selected clips menu item. To ungroup them, make sure to highlight the desired tracks, and select the Timeline  Group  Ungroup selected clips menu item.

12.2.2. Editing Audio on the Timeline

To slip an audio event along the Timeline, CTRL drag it to the left or right. To slip an audio clip with sample precision within its container, please open the Batch Edit window by pressing Ctrl+E. The Batch Edit window is also used when editing multiple audio tracks or performing a number of different audio edits.

To trim the begining or end of an audio track, click and drag the begining or end of the track. The mouse cursor is indicative of the current timeline interaction.

To split or cut an audio clip first highlight it by clicking on it, place the playhead at the desired positiont and press Shift+C to cut.

12.2.3. Audio Metadata on the Timeline

Audio Track Metadata

The Record TCin is displayed at the bottom left corner of an audio track, and the Record TCout is displayed at the bottom right corner. If space permits, the audio asset’s file name is displayed centered along the top.

Audio Track Label Metadata

The bottom left corner of the audio track label box indicates the language of the audio track using the standard 2 letter abreviation. This can be changed by double-clicking the text or from the Audio  SetLanguage menu.

The bottom right corner of the audio track label box displays the channel label (Boom, Center, etc) which is read from the track’s built metadata.

12.3. Adjusting Audio Speed with Pulldown and Pullup

Transkoder can change speed of audio clips on the timeline to compensate for the 1000/1001 speed change ratio, so audio matches the corresponding video. To adjust the audio speed from an integer TC rate such as 24 fps to a non-integer TC rate such as 23.977 FPS, select the Edit  Change Audio Speed  Pulldown menu item. A small triangle pointing downwards will appear to the right of the TC if pulldown has been applied to the audio track.

It is also possible to apply pullup/pulldown upon rendering, if the framerate if the timeline does not match the encoded FPS. For example rendering a 23.9 ProRes from a 24 project, the ProRes encoder will automatically readjust itself to apply the correct pulldown to the audio.

To adjust the audio speed from a non-integer TC rate such as 23.977 fps to an integer TC rate such as 24 FPS, select the Edit  Change Audio Speed  Pullup menu item. A small triangle pointing upwards will appear to the right of the TC if pullup has been applied to the audio track.

To revert to the original speed of an audio asset thereby removing any pulldown or pullup, select the Edit  Change Audio Speed  Normal

If Transkoder detects a 1000/1001 speed ratio difference between the project frame rate and the frame rate in the audio header, it will automatically apply the necessary pulldown, or pullup and indicate this with the appropriate triangle icon to the right of the TC.

Supported speed change scenarios:
  • 23.9 ↔ 24

  • 23.9 ↔ 25

  • 24 ↔ 25

12.3.1. Audio Conversion upon Frame Rate Conversion

To support workflows where 23.9 or 24 frame rate media is converted to 50 fps outputs (with duplicating frames) Transkoder automatically readjusts audio in a 50 fps project, as if it was a 25 fps project. This way the operator can simply drop in 23.9 media to the 50 fps project, change the audio speed to 50% and the picture and audio will be in sync at 50 frames per second.

12.4. Embedded Audio on the Timeline

When adding video clips with embedded audio to the timeline, the embedded audio will be added to the corresponding audio track automatically. These clips will also be automtically grouped together so all editing operations apply to both of them. To ignore embedded audio and not add it to the timeline, disable the embedded audio handling by selecting the Timeline  Insert Embedded Audio menu item or pressing ALT+b. All typical editing action such as trim, cut, slip, etc., will be done to both video and audio. To ungroup the audio and video clips, make sure they are highlighted and select the Timeline  Group  Ungroup selected clips menu item or press the ALT+SHIFT+g keyboard shortcut. To group the clips again, make sure they are highlighted and select the Timeline  Group  Group selected clips menu item or press the SHIFT+g keyboard shortcut.

Video assets with embedded audio will appear only as video clips in the Bin Window, there is not separate representation of the audio object in the Bin. However, the expanded metadata view does show audio metadata such as the channel number.
Transkoder does not do audio stretching to any length, and if the framerate of the video with embedded audio differs from the project framerate, the audio and video will not be in sync.

12.5. Audio Mixdowns

Audio mixdowns are used to define audio mixing and routing patterns for playback and for encoders. Transkoder defines default audio mixdowns including one for the workstation motherboard’s audio jack and one for the SDI embedded audio output.

12.5.1. Audio Mixdown Window

To open the Audio Mixdown window, press Alt+Y. The left-hand side contains a list of the defined mixdowns, with the following by default:

Audio Mixdown Window
Audio Mixdown Window
DefaultPlayback

This mix is routed to the workstation motherboard’s audio jack that is stereo, so the mix should be stereo as well. (This is symbolized by the speaker icon.)

DefaultVideo

This is a 5.1 multi-channel mix routed to the SDI embedded audio. (This is symbolized by the speaker icon with a frame around it.)

Dual Audio

This mix is used for IMF packaging and includes both a 5.1 channel and stereo channel.

To change the number of channels in the mixdown, double-click or right-click the channel number, such as, Channel 6 for 5.1 channels.

Another method is to use the channel preset drop-down menu located in the middle of the bottom menu of the Audio Mixdown window.

Audio routed through the workstation’s audio jack is indicated by the speaker icon, and audio routed to the SDI embedded audio is indicated by the speaker icon with a border around it.

The right hand side contains the actual mixdown routing table with the mixdown name, followed by the source audio channels, listed on the top from left to right.

Right below the source audio channel IDs is the Volume change indicator, displayed in db under each of the source audio channels. The volume change can be set by clicking in the field and typing a value, scrolling the mouse wheel in the field, or using the UP ARROW and DOWN ARROW keys. If manually entered, the value can be a floating point number.

Vertically is a list of the deliverable tracks with a two letter language abbreviation listed directly below the track label, and to the right of the track list is a list of the Channels for each track. This is followed by a table with columns corresponding to source audio channels and rows corresponding to deliverables' output channels. By enabling multiple columns within a single row one can mix multiple input channels.

At the bottom of the Mixdown Window from left to right are the following:

Under the mixdowns

⊕ (PLUS SIGN)

Create a new mixdown. (Double-click to rename.)

⊖ (MINUS SIGN)

Remove the highlighted mixdown.

SPEAKER ICON

Designate a mixdown for routing through the workstation motherboard’s audio jack.

SPEAKER ICON w/ BORDER

Designate a mixdown for routing through the SDI embedded audio.

Under the mixdown routing table

ADD GRP

Add a channel group to the mixdown. Groups will be labeled A,B,C…​, etc. This is used by encoders that support multi-group audio encoding such as IMF.

DEL GRP

Remove the hightlighted channel group from the mix.

CHANNEL PRESET

Indicates the number of channels to be used for the group. The value MOS may be used to define a mixdown without any audio. Such a mixdown can then be used to generate a render without any audio present.

LABEL

A custom label such as 5.1 DTS can be specifed here and will appear above the channel group. This can be used in IMF encoding audio asset naming.

LANGUAGE

Define the language for the channel group. This can be used in IMF encoding audio asset naming.

SET ALL

This will merge all source channels to one deliverable or output channel.

CLEAR ALL

This will clear all routings.

One To One

This will create a 1 to 1 mapping of each source channel to a unique deliverable or output channel.

The Audio  Video Mixdown menu shows the actual list of defined mixdowns and allows quickly switching between them. This is especially useful when doing qc, because it speeds up the testing of different mixdowns.

12.5.2. Channel Merging and Rerouting Examples

12.6. Audio on the Encode Page

On the Encode Page, results are on the left hand side, the result specific encoders are listed in tables in the middle of the page, and the encoder settings are in a pane on the right hand side. To select from the defined list of mixdowns, click on the Mix field in either the desired encoder table in the middle of the Encode Page, or in the encoder settings pane on the right hand side. A MixdownList drop-down will appear with the actual defined list of mixdowns.

The MixdownList drop-down includes:

Same As Source

This will add all tracks and all channels.

Same As Source A

This will add channels from the channel group A. (This list item is dynamically generated to match the mixdown channel groups.)

Same As Source B

This will add channels from the channel group B. (This list item is dynamically generated to match the mixdown channel groups.)

Same As Source C…​ etc

This will add channels from the channel group C. (This list item is dynamically generated to match the mixdown channel groups.)

Same Tracks As Source

This is only used for IMF packages and is a convenient way to use audio grouping as layed out on the Timeline.

Only the IMF encoder supports multiple channel groups. All encoders that do not support multiple channel groups will use the first group in the list if multiply channel groups are specified.

If using the Encode Page to drive audio file names in IMF packages, choose either a single group or a multi group entry instead of a Same As Source entry. This way the Audio Path setting becomes clickable.

When editing Audio Render Path, the following keywords can be used:

  • Language

  • Label

  • TrackName

  • AudioFilePostfix

12.7. Volume Metering

There is a small volume meter at the top of the Timeline window to the left of the Snap button. Double-click the small volume meter to make a larger Volume Meter window appear. The small volume meter can monitor up to 8 channels and the Volume Meter window can monitor any number of channels of the SDI embedded audio volume. To change what channels to monitor in this window please asjust the SDI output mixdown. To change between mixdowns it is not necessary to open the Mixdown Window, the SDI audio mixdown can be accessed directly from the Audio  Video Mixdown menu option.

Transkoder Volume Meter on Timeline window and separate Volume Meter window
Transkoder Volume Meter on Timeline window and separate Volume Meter window

12.8. DCP Audio Handling

DCP finalizing requires strict adherance to the the DCP standards, such as making the Picture Reel boundary identical to the Audio Reel boundary, having only 1 audio track, etc. In order for the audio to be valid for DCP, every frame of video must have a matching frame of audio, and both the video and audio reel markers must align. In addition only a single audio track may be included and there must be an even number of channels. If more audio tracks are needed, then the DCP package needs to contain multiple CPLs (Composition Play Lists).

12.8.1. DCP Audio Layout

By default the DCP audio encoders use a 5.1 mixdown with (L, R, C, LFE, LS, Lr) field layout. If a different audio layout is needed the user should create a custom mixdown and set it as the active mix in the Encode Page and also set the apropriate audio layout on the finalize page to indicate the audio channel configuration. For instance, if one needs to create a 7.1 multichannel DCP with Hearing Impaired audio channel, a 10 channel custom mixdown needs to be created in the Mixdown Window and activated in the Encode Page and the DCP Audio Layout popup button needs to be adjusted accordingly. The default Auto value indicates that Transkoder will use a layout setting that is matching the current channel number. However, for the same channel numbers there may be potentially multiple possible layouts.

Possible choices include:

Auto

Uses the most basic audio field configuration that matches the channel number of the current mixdown.

MOS

Silent audio

Wild Track Format

This will create the minimum required 14 channel layout needed for Dolby Atmos. If more than a 5.1 (6) channel mix is desired, make sure to create a custom mixdown in the Mixdown window and choose this mixdown for the Mix field value.

Various multichannel configurations

See the table below.

Configuration / Channels

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

2.0

L

R

-

-

-

-

-

-

-

-

-

-

Motion Data

Sync

Sign L.

-

5.1

L

R

C

LFE

Ls

Rs

-

-

-

-

-

-

Motion Data

Sync

Sign L.

-

5.1 with Hi

L

R

C

LFE

Ls

Rs

Hi

-

-

-

-

-

Motion Data

Sync

Sign L.

-

5.1 with Vi

L

R

C

LFE

Ls

Rs

-

Vi

-

-

-

-

Motion Data

Sync

Sign L.

-

5.1 with Vi and Vi

L

R

C

LFE

Ls

Rs

Hi

Vi

-

-

-

-

Motion Data

Sync

Sign L.

-

6.1 (5.1 + Center Surround)

L

R

C

LFE

Ls

Rs

Cs

-

-

-

-

-

Motion Data

Sync

Sign L.

-

6.1 (5.1 + Center Surround) with Hi

L

R

C

LFE

Ls

Rs

Cs

-

Hi

-

-

-

Motion Data

Sync

Sign L.

-

6.1 (5.1 + Center Surround) with Vi

L

R

C

LFE

Ls

Rs

Cs

-

-

Vi

-

-

Motion Data

Sync

Sign L.

-

6.1 (5.1 + Center Surround) with Hi+Vi

L

R

C

LFE

Ls

Rs

Cs

-

Hi

Vi

-

-

Motion Data

Sync

Sign L.

-

7.1 (SDDS)

L

R

C

LFE

Ls

Rs

-

-

Lc

Rc

-

-

Motion Data

Sync

Sign L.

-

7.1 (SDDS) with Hi

L

R

C

LFE

Ls

Rs

Hi

-

Lc

Rc

-

-

Motion Data

Sync

Sign L.

-

7.1 (SDDS) with Vi

L

R

C

LFE

Ls

Rs

-

Vi

Lc

Rc

-

-

Motion Data

Sync

Sign L.

-

7.1 (SDDS) with Hi+Vi

L

R

C

LFE

Ls

Rs

Hi

Vi

Lc

Rc

-

-

Motion Data

Sync

Sign L.

-

7.1 (DS)

L

R

C

LFE

Lss

Rss

-

-

-

-

Lrs

Rrs

Motion Data

Sync

Sign L.

-

7.1 (DS)+Hi

L

R

C

LFE

Lss

Rss

Hi

-

-

-

Lrs

Rrs

Motion Data

Sync

Sign L.

-

7.1 (DS)+Vi

L

R

C

LFE

Lss

Rss

-

Vi

-

-

Lrs

Rrs

Motion Data

Sync

Sign L.

-

7.1 (DS)+Hi+Vi

L

R

C

LFE

Lss

Rss

Hi

Vi

-

-

Lrs

Rrs

Motion Data

Sync

Sign L.

-

One can choose the Language for the DCP in the Finalize window, it is also auto-initialized based on the language metadata from the timeline. This information is used to provide the values for both the Title and Annotation fields.

12.8.2. Dolby Atmos Audio

Dolby Atmos® is a proprietary audio format that comes in an MXF container file added to the timeline. The MXF contains the spatial audio description metadata that is used for the dynamic rendering of audio, based on the available speaker layout and capabilities.

The Atmos audio asset that contains the Dolby metadata is not a separately playable audio content. Transkoder, however, displays the single channel audio waveform embedded in the file, which is essential to check the proper syncronization. Make sure that the normal audio asset is on the first A audio track, while the Atmos MXF is on the B track.

Atmos MXF on the timeline, use the visible 2-pop signal to check sync
Atmos MXF on the timeline: Use the visible 2-pop signal to check synchronization

In a Dolby Atmos DCP, the sync signal is generated on the 14th audio track in the regular audio asset. The Atmos option must be enabled in the DCP Finalize window in order for it to be added to the DCP package. If the current mixdown has less than 14 channels, the encoder extends it with silent channels so the sync track can be rendered.

Channel Number Selector

There is also an option to create a Dolby Atmos® DCP package with 16 channels for the main audio. You can set the default channel number for the main audio track in the advanced DCI section of the Settings Page. Here, you can select either a 14-channel or a 16-channel Atmos DCP package.

Selecting the default channel number for main audio for Dolby Atmos DCP packages
Selecting the default channel number for main audio for Dolby Atmos DCP packages

12.9. IMF Audio Handling

In order for the audio to be valid for IMF, every frame of video must have a matching frame of audio, but the video and audio reel markers do not need to align. IMF packages may also include multiple audio tracks such as the typical dual audio set up of a 5.1 channel track, and a stereo track (2 channel).

There are two approaches to managing audio IMF packages within a project. In one approach, the audio Timeline is set up to reflect the desired resulting audio in the IMF package. This approach uses the Same Tracks as Source option in the Mixdown field. When using this approach it is very important that number of tracks and channels remain the same along the entire Timeline.

Same Tracks as Source Use Case

You have a 6 channel and a stereo audio file and need to create an IMF package with a 5.1 channel and a stereo audio track. Add both files to their own separate tracks on the Timeline and choose .Same Tracks as Source when finalizing the IMF package.

In the other approach, the audio mixdown window is used to reroute the audio into the desired resulting audio in the IMF package. For each desired audio track in an IMF package, it is necessary to add an identical number of audio groups and match the appropriate source channels from the Timeline to the the desired final IMF package audio configuration as in the example below:

Channel Merging and Rerouting for IMF Mastering
Creating a custom mixdown to map 4 stereo inputs into a 5.1 and a stereo output
Creating a custom mixdown to map 4 stereo inputs into a 5.1 and a stereo output.
Routing specified in the "customMix" mixdown
Routing as specified in the "customMix" mixdown.

The ALT+SHIFT+v keyboard shortcut or the Timeline  Add Track  Add Audio Track menu item may be used to add additional tracks to a project timeline. For each desired The additional tracks may be populated just like track 1, by adding *.wav files, including multi-channel ones, or .mxf files.

The tracks may have their language labeled, either by double clicking in the bottom left hand corner of audio track on the timeline, or selecting the appropriate language from a list in the Audio  Set Language menu. The language label may be used later in rendered file names, and even as metadata in the created IMF packages.

Dual Audio Use Case with 4 Stereo Audio Files

You have 4 stereo audio file and need to create an IMF package with a 5.1 channel and a stereo audio track. Add the appropriate group of 3 stereo files and single stereo file to two separate tracks on the Timeline and choose .Dual Audio when finalizing the IMF package. Transkoder will automatically recognize the configuration and generate the correct 5.1 channel and stereo tracks when finalizing the IMF package.

Custom Mix Use Case with 8 Stereo Audio Files

You have 4 stereo audio files in English, and 4 stereo audio files in Spanish. Add the appropriate group of English 3 stereo files and single stereo file to two separate tracks on the Timeline and do the same with the Spanish tracks. Then in the Mixdown Window, create a new mixdown, with 4 groups, a 5.1 channel English group, a stereo English group, a 5.1 channel Spanish group, and a stereo Spanish group. Next, reroute each audio source channel to the desired result channel for the IMF package. When finalizing the IMF package, make sure to choose this mixdown.

12.10. ProRes Audio Handling

A new option has been added to the ProRes encoders to indicate the audio field layout. Available options are Auto, None, Mono, Stereo, 2.1, 2.2, 3.0, 3.1, 3.2, 5.1, 7.1, Discrete.

When Auto is chosen the number of output audio channels in the current mixdown is driving the field layout metadata.

12.10.1. Audio Asset Naming

Transkoder allows the user to configure the audio MXF names in the package to be created. For details see the Naming Patterns for IMF section.

12.11. Drive External Devices with LTC Time Code

Real-time LTC is generated into SDI embedded Audio which matches the current record timecode. This is for driving systems such as ProTools via LTC timecode. To enable this please set to On, the Generate LTC Setting in the Advanced Settings section of the Video Settings section of the Settings Page. Other controls on the TAB page include:

  • LTC Channel: 7 (default: 7), this is the audio channel number to write the signal into

  • LTC Offset: plus minus frames (applies immediately)

  • LTC Volume Multiplier: 1-255, adjust signal gain for expert users(default: 240)

Please make sure the current video out mixdown has enough channels so the channel identified by the LTC Channel setting exists.

13. Titles and Captioning

Transkoder supports the loading, autoconversion and export of various text or image-based (PNG) broadcast and cinema subtitle and caption formats. Tools allow the inspection and basic editing of subtitles, packaging into DCP and IMF formats or rendering (burning) into the picture.

For clarification, the definition of the related terms are the following:

subtitles

regular titles that are mainly used for the translation or the transcription of conversations

closed captions

titles including transcription or translation of the dialogue and also the description of non-verbal sounds to help hearing-impaired people, or anyone in a too noisy environment or in a place that must be kept quiet to understand the video content more accurately

titles

both subtitles and closed captions, there is no difference in how to work with them in Transkoder.

To adjust subtitle settings on the Settings Page, please see the Subtitle Settings chapter.

13.1. Importing Titles

Import title files to a project like any other types of media files with the exact same methods.

For the entire list of supported title formats, see Supported Subtitle Formats.

To add assets, such as clips, to bins via the file system, click on the small ⊕ (plus sign) button in the bottom left corner of the Bin Window, and a file system browse dialog will appear on the screen. Browse to the asset to be added, and click on the OK button.

Transkoder - Adding assets via file system
Transkoder - Adding assets via file system

In the Bin Window, Subtitles are represented by a white capital letter T on a black background. To find specific subtitles easily within a complex project please use the subtitle filter button on the top right corner and/or use search queries like EN or FR in the search bar. When viewing the subtitle assets in the Bin Window with details view, the basic properties of the files is displayed.

13.2. Using Titles

To add a title to your composition on the timeline, grab a title source clip from the bin and drag it to the timeline area. The possible title tracks will appear. T1 track is for regular subtitles, while CC1 is for closed captions. Drop your media to the required track and position on the timeline, or press V to insert it to the playhead position (POS) or any timeline (TC).

To move a title clip in the timeline to a different timecode or to the other title track, just grab and drop the clip to the required track and position.

For DCP projects, the subtitle asset can be added to the T1 track to define standard subtitle, or to the CC1 track for closed captions, or to both. You can also use the standard V shortcut, and the media insertion modes POS or TC to control where the timeline event is added.

The timeline representation of subtitle events also include the frame-specific spots shown with text.

13.3. Rendering/Viewing Subtitles

To render or view a subtitle asset in a project, a Subtitle node needs to be in the pipeline on the Node Page. With the Subtitle node selected, press the H key to bring up the Subtitle HUD. With this HUD, it is possible to change the position, color and alignment of the current spot. There is also a switch to decide if the subtitle track or the closed captioned track should be rendered.

It is possible to mute or disable the rendering of subtitles, by choosing Subtitle  Mute Subtitles menu option. This is very important if there are subtitles on the timeline but the user does not want to render (burn) them into the picture.

The font selection used for rasteriaztion is done in the Subtitle Settings section of the Settings Page. One can use either the font associated with the subtitle XML (if there is any), the system Arial font, or setting the Subtitle Font Source field value to CustomFont, will enable the use of the font located in the path as defined Custom Font File field. This also affects subtitle packaging for DCPs.

13.4. Editing Subtitles

Editing Subtitles can be done either on the main display screen, or via the Subtitle Table. With the Subtitle node active, editing the subtitles can be done by double clicking the subtitles shown on the main display screen, and modifying the text. To bring up the Subtitle Table, choose the Window  Show Subtitle Table menu option or by pressing Shift+S.

The Subtitle Table has the following columns from left to right: Spotting#, TCIn, TCOut, Duration, Text, HP (Horizontal Position), and VP (Vertical Position). Click in the Text field to modify the particular spot. Double-clicking one of the spot numbers will navigate to that spot on the timeline. Clicking anywhere on the timeline will scroll the Subtitle Table to the matching spot.

The physical characteristics of the subtitle are edited via the Subtitle node. Please refer to the Subtitle section of the Tools on the Node Page chapter for an indepth description.

Subtitle Offset Controls
Subtitle Offset Controls

The operator can move a subtitle by adjusting the AddOffsetV and AddOffsetH sliders and then use the Apply to all spots button to paste this change to all subtitle events on the timeline.

Transkoder does auto conversion of the input subtitle format color space into the appropriate DCI color space required for packaging. To disable this auto conversion and use the original color in the native color space, go to the Subtitle section of the Settings Page with the Advanced settings turned ON, click on the Bypass button in the Subtitle Color Conversion field.

Transkoder also does auto conversion of the input subtitle format into the one required for packaging. In some instances when creating DCP packages it is best to leave the original xml subtitle file untouched to preserve compatibility with legacy content delivery systems (i.e. movie theater projectors). In this case, to disable the auto conversion, go to the Subtitle page in the Settings Page, click on the Off button in the Re Generate Subtitles field. If auto-conversion if off, and the subtitle is not in the required format, the DCP or IMF packaging will fail.

13.5. Subtitles and Captions in DCPs

To package subtitles, closed captions or both into a DCP package first the apropriate timeline needs to be created. When editing subtitles please keep in mind that in Interop packages the reel and subtitle configuration should be identical. In SMPTE packages this is no longer a limitation and a single long-play subtitle can be used for a multi-reel composition. Make sure the CC track contains the closed captions, and the T track has the subtitles.

Upon finalizing there are two buttons to control the inclusion of the subtitles and captions respectively.

13.6. Captions in IMFs

Transkoder supports a single caption track within the composition in case of IMF mastering. When packaging, the subtitles will be converted to the IMSC1 format, unless auto conversion is disabled. Please use the Subtitle button in the Finalize Window to trigger caption/subtitle packaging.

13.6.1. Improved Automatic Font Selection

When rendering IMSC1 or other types of subtitles the system automatically selects the proper font using the following mechanism:

The default fonts directory for Transkoder subtitle fonts is <install home>/fonts/ - this is a new folder in install directory. Installer will always place/replace the default fonts into this folder, but the user can also put here any ".ttf" or ".otf" font files for subtitle rendering.

If the Subtitle / Subtitle Font Source project setting is set to System Font or if Transkoder cannot find suitable "Incoming" font in these font files, it will try to use any suitable font file from <install home>/fonts/ directory. Transkoder will parse font files in alphabetic order if more then one is present and will use the first working font. If it cannot find a character in this font file, it will fallback to the default international font set, which can be found in <install_home>/pymodules/guidata/fonts/NotoSans-Regular/. These fonts cover all the international unicode language character sets very well.

Specific font families can be found in <install_home>/pymodules/guidata/fonts/italic/ for italic characters or <install_home>/pymodules/guidata/fonts/monospace/ for monospace rendering. These fonts (as well as the international sets) can be replaced with any user desired ones, just delete the default and place a new one.

The Transkoder installer will place/replace default fonts into these folders.

13.7. SCC Captions

Transkoder supports the embedding of Scenarist Closed Captions (SCC) captions into H264 10-bit transport stream. For this, you need a subtitle source with SCC format that you add to your timeline composition. In this case, autoconversion does not work.

[IMPORTANT] Please, make sure that the subtitle source is set to CC on the Encode Page for the Broadcast H264 encoder.

13.8. Forced Captioning

For media in IMF IMSC format, Transkoder supports reading itts:forcedDisplay metadata, which is used to mark specific titles to display even when subtitles are disabled. Forced subtitles are commonly used in movies to provide translation when the characters speak a foreign language; when a sign, flag or other text in a scene needs translation; or when the location, year, time or any important attribute of the scene needs to be subtitled as well. To turn this feature on, use the OnlyForceDisplay option on the HUD of the Subtitle tool.

13.9. Safe Area

Safe area guides are displayed either as an overlay in the GUI, which if visible only on computer monitors, or as an overlay embedded into the image so it is visible in the SDI video ouput as well. The separate switches for these two modes are the following:

Safe area guides
Safe area guides
There are two different types of safe areas: the inner rectangle is the title safe area, while the outer rectangle is the action safe area.

The Aspect ratio selector allows control of the geometry of the guides, while the Opacity slider affects the transparency of the guides.

14. Video Output Settings

Find the explanation of some video output settings in this chapter. See the entire list of Video Settings as part of the the Settings Page.

14.1. HD/UHD Dual Video Out

The two independent video out mode can be enabled by selecting the Dual Video Out Mode in the Video Settings section of the Settings Page.

Available selections:

  • AUTO - will follow first video out mode

  • HD / 2K / QHD / 4K

  • DUALLINK HD / 2K / QHD / 4K

Every custom video out settings (like 3G or 12 bit color depth) are also available for the dual video out mode. Select the video tab and press Show Advanced to change these options. The secondary video output will be on SDI Out 5 cable and above.

To control which node page result is sent to which output channel please left-click in the small white rectangle above the result to set the first output, and Alt+click in the rectangle to select the second output, indicated by double dot.

Click and Alt+Click assigns the results to output one and two
Click and Alt+Click assigns the results to output one and two
Requirements: AJA Corvid88, Windows platform. This feature does not support 3D stereo mode.

14.2. Video Output Auto-Resize

The automatic resizing of video output is used to simplify workflows where HD/2K media is monitored on a UHD display or vice versa. Please enable the Resize option on the Video settings in the Tab page, in case if the video output being 2x or half the size of the current result the image is automatically resized. This generates a "meaningful" image on the monitor instead of being center cropped or being a small HD image in the center of the UHD frame. This feature is essential in the Content Mapping Unit workflow where the same UHD HDR output is sent to two different video results at different resolution (as of today the CMU only supports HD input).

14.3. UHD SDI Output on Two Cables (3G)

Available on Windows, Mac support is coming in Update 2. To turn on 2 wires QHD mode with enabling Aja QHD On Two Wires setting in the Video Settings section of the Settings Page. The SDI output will be on the 1st and the 2nd channels, and will work in QHD up to 30 fps or to 60 fps with halved video frame rate enabled.

14.4. Lock SDI Output to Reference Signal

Enable this with Video settings / Reference Signal parameter.

15. Analyzer

Transkoder has a variety of image analyzer tools to inspect pixel color values, brightness levels, signal range, etc. There are two different modes to use these tools: having a secondary UI monitor as a second head analyzer, or without a secondary monitor, as an internal floating window in the main UI.

The Second Head Analyzer requires a secondary UI monitor and can be enabled in the Settings Page. After turning on and restarting the application, it will be automatically enabled. There are different layout presets to choose from. See the following chapters for details.

The Second Head Analyzer is only available on the Windows platform.

When no secondary UI monitor is available, the floating Analyzer window can be opened in the main UI by pressing Ctrl+H. In this window, a waveform is shown in the left side of the window with either a vectorscope or a CIE xy gamut graph on the right-hand side.

Some features are exclusive to the Second Head Analyzer mode. Thus, when using the floating Analyzer window, the corresponding options are disabled in the Analyzer menu.

15.1. Key Features

  • Waveform and histogram graphs

  • Camera log / HDR / SDR analysis

  • Color picker

  • CIE xy gamut graph*

  • Single Line Mode*

  • HDR analysis with brightness and gamut check*

  • False color analysis including HDR brightness and gamut warning*

  • Optional LUTs for viewing and/or analysis*

  • Audio phase and level meter*

The items marked with asterisk (*) are available only when using a Second Head Analyzer on a separate monitor.

15.2. Analyzer Modes

The waveform monitor, the histogram, and the vectorscope are used for video signal levels inspection, as well as for peak brightness and gamut validation. See the following sections for details about currently supported Analyzer view modes. These options can be chosen from the Analyzer  Analyzer Mode menu options.

Waveform Luminance

Waveform showing only luminance (Y) and no color information. Each pixel on the graph shows the frequency of the corresponding luminance value in the given column. Higher luminance value means lighter image content, while lighter pixel means higher frequency of that luminance value.

Waveform Lumi Color

Waveform showing both luminance and color information. Here, in addition to the Waveform Luminance mode, the graph is colored according to color information. The hue and saturation of a pixel shows the sum of the dominant colors having the corresponding luminance value.

Waveform Color

Waveform showing the RGB channels superposed in one graph.

Waveform RGB Color

Waveform showing the RGB channels in a split view, where each graph is shown in its respective color.

Waveform RGB

Waveform showing the RGB channels in a split view, where each graph is shown in monochrome.

Waveform YCbCr

Waveform showing luma (Y), blue minus luma (Cb), and red minus luma (Cr) in a split view.

Waveform YRGB

Waveform showing luminance and RGB channels in a split view, where each color channel graph is shown in its respective color.

Histogram Luminance

Histogram Luminance is showing the frequency of luminance values throughout the current image content. Left-hand part is for darker, while right-hand part is for lighter image content.

Histogram Color

Histogram Colors showing the frequency of RGB channel values throughout the current image content, superposed. For each channel, the left-hand part is for image content including less, while right-hand part is for image content including more of that color component.

Vectorscope/Gamut

Vectorscope is showing the hue and saturation of all pixels throughout the current image content. The middle of the circle indicates a lower, while the edge of the circle indicates a higher level of saturation. The brightness of the graph shows the frequency of the respective color in the image content. A skin tone line is also available on the Vectorscope graticule. Thus, you can set your images of people in the most natural way.

Selecting the Analyzer mode from the menu

15.2.1. Luminance Waveform

Waveform Luminance
Waveform luminance

15.2.2. Luminance Waveform with Colors

Waveform LumiColor
Waveform lumi color

15.2.3. Color Waveform

Waveform Color
Waveform color

15.2.4. RGB Color Waveform

Waveform RGB Color
Waveform RGB color

15.2.5. RGB Waveform

Waveform RGB
Waveform RGB

15.2.6. YCbCr Waveform

Waveform YCbCr
Waveform YCbCr

15.2.7. YRGB Waveform

Waveform YCbCr in RGB Channels
Waveform YCbCr in RGB channels

15.2.8. Histogram Luminance

Histogram Luminance
Histogram luminance

15.2.9. Histogram Color

Histogram Color
Histogram color

15.3. Analyzer Color Space and Range

The color space and the range of the Analyzer can be configured with the options in Analyzer  Analyzer Color Space menu. You can choose from the options mentioned in the following sections.

Video Range

The waveforms are scaled in a way that black and white levels correspond to the legal range video levels, allowing non-legal code values to be shown.

Full Range

The waveform levels are configured to show pixel values only between the legal black and white levels.

To be able to inspect super-white and sub-black signal levels, it is recommended to leave the Analyzer in Video Range mode.

Selecting the current color space from the menu
Selecting the current color space from the menu

15.3.1. Camera Signal Mode

Several original camera signal curves are supported for processing media in the original capture color space. Supported formats are the following:

  • ARRI Log C WideGamut

  • CanonLog2

  • CanonLog3

  • Panasonic VLog/VGamut

  • Red WideGamut/Log3G10

  • Sony SLog3/BT2020

  • Sony SLog3/SGamut3

  • Sony SLog3/SGamut3Cine

  • ACEScct

In these modes, the waveform graticules are displayed in camera stops. To set up the base level and the warning level, adjust the following settings in the QC section of the Settings Page:

CameraBaseGrey

Base grey level in nits; the 0 exposure line will be drawn at this level.

CameraWhiteStopsOverGrey

Brightness warning level; it stops over base level.

15.3.2. SDR Mode

For working with SDR gamma encoded broadcast or cinema signal, set the Analyzer Color Space to SDR. Possible primary color options are the following:

  • Rec.709

  • P3 DCI

  • P3 D65

  • Rec.2020

  • XYZ DCI

15.3.3. HDR – HLG Mode

HLG mode supports the variable Hybrid Log-Gamma color encoding. In this mode, code values range from 0 to 1, and pixels do not have a specific nit level interpretation. Possible primary color options are the following:

  • Rec.709

  • Rec.2020 (with gamut warning for colors outside of P3)

15.3.4. HDR – PQ Mode

In this mode, SMPTE ST 2084 Perceptual Quantizer (PQ) transfer function is used, and thus pixels values have nit level interpretation. Peak brightness values above the maximum allowed nit level are indicated red on all graphs.

Analysis of PQ Signal with the CIE xy Gamut Graph, the Vectorscope and False Color Showing Areas with Illegal Peak Brightness
Analysis of PQ signal with the CIE xy gamut graph, the vectorscope and false color, showing areas with illegal peak brightness

Possible color primary options are the following:

  • Rec.2020 (with gamut warning for colors outside of P3)

  • P3 D65

  • Rec.709

By default, the peak brightness threshold is set to 1000 nits according to the HDR10 standard. When working with PQ masters of higher peak brightness, the corresponding maximum brightness needs to be adjusted on the QC section of the Settings Page.

15.4. Color Picker

There is also Color Picker functionality built in to the Analyzer window. Moving the cursor over any pixel in the video frame will show its RGB value under the waveform table.

15.5. Using Analyzer with Analyze Node

The analyzer can operate with or without an Analyze node on the node page. When used without a node, the analyzer will show the values of the currently active render result.

When used with an Analyze node, it will show the values of the video at the actual placement of the node along the render pipeline. Multiple Analyze nodes can be placed along the pipeline, and the Analyzer window values displayed can be switched between them by bringing up (H) the HUD display of the Analyze nodes and toggling the Active field button to ON. Enabling a particular Analyze node will switch all the others off.

15.6. Second Head Analyzer

When using a second GUI monitor, the second head analyzer can be enabled in the GUI section of the Settings Page. After restarting the application, the Second Head Analyzer appears in full screen view on the secondary monitor regardless of its size or position relative to the primary output. It is recommended to use a monitor at least with HD resolution.

Second Head Analyzer feature is only available on Windows.

15.6.1. Analyzer Layouts

The Analyzer uses layouts either with four quadrants, or a full screen single view.

Image

Showing the display image in full screen single view

Waveform

Showing the waveform in full screen single view

Gamut

Showing the gamut in full screen single view

Vectorscope

Showing the vectorscope in full screen single view

Combined with Audio

Showing four quadrants including the audio window

Combined with Pixel Picker

Showing four quadrants including the pixel picker window

Combined with Vectorscope

Showing four quadrants including the vectorscope

4 quadrants

15.6.2. CIE xy Gamut Graph

Default Analyzer layout with four quadrants

The CIE xy Gamut View can be used to check the encoded colors on a standard CIE xy graph, and to see if they are within the valid color range in case a specific gamut constrain is to be enforced. This is relevant when working in the Rec.2020 color space, so the operator can identify that the colors are outside of the P3 gamut. This is a requirement for several currently popular delivery formats.

CIE xy Graph with Brightness and Gamut Bars Showing out of P3 Colors.
CIE xy graph with brightness and gamut bars showing out of P3 colors.

This view also contains two bars. The one on the left is the brightness bar that indicates the brightest pixel detected on the actual frame. The one on the right is the gamut bar that shows if the colors of the actual frame are within the legal color gamut (P3 in case of Rec.2020 input color space). Please note, Transkoder uses thresholds when detecting these extremes, so while using the default settings, a few pixels are allowed to exceed the legal limits without triggering a warning.

15.6.3. Additional Analyzer Settings

Single Line Mode

When using a Second Head Analyzer, Single Line Mode can be enabled by the Analyzer  Single Line Mode option. If this mode is turned on, Transkoder will only analyze the selected line. The selected line number is printed in the upper right corner of the display image.

Single Line Mode
Single Line Mode
White Point Compensation

This option is to be enabled in DCI XYZ workflows, where the intended white point is not the standard DCI white point but D65. When this option is enabled, the color picker will show the same R, G and B values for colors on the D65 white axis, and the vectorscope will show the D65 white in the center.

There are two options:

  • D65

  • DCI

  • Off

False Color

It is possible to use Transkoder for checking illegal or near-illegal brightness and gamut levels by enabling the False Color mode. The available False Color display modes are the following:

  • Brightness

  • Brightness Warning

  • Gamut

  • Gamut Warning

False Color - Gamut Warning on the Left, Brightness Warning on the Right
False color: Gamut Warning on the left, and Brightness Warning on the right

In the Warning modes, pixels that are too bright or out of gamut are colored red. Values at 90% of threshold are orange. The rest of the pixels are displayed in grayscale.

YCbCr Matrix

For the purpose of showing YCbCr components in various graphs (such as the vectorscope or the waveform YCbCr), the result image is converted to the YUV space. Select from the following formats for the specific matrix transformation:

  • Rec.2020

  • Rec.709

White Point Sign on the CIE xy Graph

Transkoder represents the color space white point on the CIE xy graph by a small square, as shown in the following image:

White Point Sign as a small rectange on the CIE xy Graph, the cross is the picker
White Point Sign as a small rectange on the CIE xy graph, where the cross is the picker

Options are the following:

  • D65 - 6500 Kelvin white point

  • DCI - 6000 Kelvin white point

  • OFF - No marker shown

VectorScope Target

This function shows vectorscope targets at 75% (broadcast safe) or 100% intensity levels.

Options are the following:

  • 75%

  • 100%

Waveform Graticule

Waveforms have "smart" graticules based on the current color space selection. HDR waveforms use bit level values, while SDR modes use percentage. This behavior can be enabled and disabled by the following options:

  • Default: automatic, percentage or bit level display

  • Code Values: always shows 10 bit code value graticules

Lookup Tables

Transkoder allows Lookup Tables (LUTs) to be used for two different purposes:

  • Display LUT: This LUT is only applied to the image being shown and does not affect the actual analysis of the pixel values. This is for the operator to see a correct image on display devices that do not match the current video signal type. When using the Auto option, the LUT is automatically selected in a way that the incoming image looks correct on a Rec.709 SDR monitor.

  • Scope LUT: This LUT is applied to the video signal being processed. This way all scopes, waveforms, etc. will be based on the remapped image.

Analyzer Gain

The following options allow the operator to increase or decrease the brightness of the analyzers so that lower density areas are also visible:

  • Increase Gain

  • Decrease Gain

  • Reset Gain

Zoom

For better visibility, there is a setting to zoom into the Vectorscope graph:

  • VectorScope Zoom: zooms into the center of the vectorscope for better visibility of near-grey colors

Pixels View

This menu controls the pixel readout format when using a combined view with the pixel picker. To select any particular pixel for analysis, Ctrl+Click any pixel in the main viewer. Available options are the following:

  • Hexadecimal: 16 bit hexadecimal code values

  • Decimal: 10 bit decimal code values (0-1023)

  • Nits: nit level interpretation of the pixel code values

Combined View with the Pixel Picker
Combined view with the pixel picker

If the pixel picker is active, the selected pixel is indicated on all scopes, graphs and waveforms.

Pixel readout from the Location Marked with Ctrl-Click.
Pixel readout from the location, marked with Ctrl+Click.
Audio Phase Meter

Transkoder supports an audio phase meter, which can be configured to display either 2-channel (stereo) or 8-channel (surround) audio. When configured to display 2 channels, channels 1-2, 3-4, 5-6, or 7-8 may be displayed. When configured to display 8 channels, channels 1-8, or 9-16 may be displayed.

Audio Phase Meter
Audio phase meter
Audio Level Meter

Transkoder supports an audio level meter up to 16 channels, which can be used to monitor audio levels and also display peak DB levels.

Audio Level Meter
Audio level meter

15.6.4. Timecode on Image

Use this menu to set the timecode to be displayed over the image on the Second Head Analyzer display.

Timecode on Image on the Second Head Analyzer display
Timecode on image

16. QC Tools

Transkoder is equipped with a wide range of QC tooling. The QC tools for advanced packages such as IMF packages and HDR media in particular are very sophisticated and provide in-depth analysis far beyond anything else on the market.

16.1. Video Audio Playback

Transkoder guarantees the real-time playback of all supported audio and video formats when it is run on the hardware recommended in the System Requirements section matched with sufficiently fast storage. To achieve the guaranteed real-time playback, the output must be routed through professional output devices such as SDI output. This is necessary for proper QC playback in any case as it is not possible to display the playback with the correct refresh-rate or raster. If using non-professional output devices, the playback will still be in real-time but without the appropriate image fidelity.

To double check that the the playback is real-time, bring up the Performance window, by pressing Ctrl+P. The Playback field should indicate a steady fps value which matches the fps of the media being output. The Read Ahead field indicates the status of the fullness of the read cache and if this number starts to drop below 30, this is usually a good indication of the storage not being fast enough. If the Read Ahead field does not change but the Result Read Ahead field value starts to drop, this is an indication of a resource bottleneck either CPG/GPU or with the Transkoder engine itself. Please double check to make sure that Transkoder is being run on conformant HW, and if so, please refer to Colorfront support for additional help with troubleshooting.

On Mac platforms, because of the lower performing hardware, the playback of audio and video in the Transkoder GUI monitor only will potentially not be in real-time. This will not affect real-time playback through the professional output devices.

If playback is not real-time Transkoder will resync audio with video occasionally and this will cause an audible glitch sound.

16.2. Audio Playback

Transkoder defines default audio mixdowns including one for the workstation motherboard’s audio jack and one for the SDI embedded audio output. For QC purposes, please make sure to use the mixdowns for the SDI embedded audio output to ensure that the audio will be synchronized with the video being QC’d. Any desired mixdown can be directed to the The SDI embedded audio output by making sure to highlight it and clicking on the the speaker icon with a frame around it in the Mixdown Window, which can be opened by pressing Alt+Y.

Transkoder Mixdown Window
Transkoder Mixdown Window

Monitoring playback levels can be done via either the small volume meter located on the right-hand side just above the timeline, to the left of the Snap, Ripple, EDIT, and GRADE buttons. Double-clicking on the small volume meter will bring up the Volume Meter Window, which can be used to monitor up to 24 channels. Please see Volume Metering chapter for more information.

Transkoder Volume Meter on Timeline window and separate Volume Meter window
Transkoder Volume Meter on Timeline window and separate Volume Meter window

16.3. Manual QC Tools

16.3.1. Tagging Notes

In Transkoder, the Quality Check (QC) window, which is opened in Window  Show Quality Check, or by pressing Shift+Q, is used to mark frames and save QC notes. It is the primary tool used for the manual QC process.

To add a new entry, click the plus sign (+) button. This will capture the timecode at the playhead position and increment the QC count (located between the left and right ◀ ▶ arrows) for the current clip. You can delete any entry with the minus sign (-) on the QC window.

To jump the playhead to the previous or next QC location on any clip in the timeline, click on the left or right arrow respectively, next to the QC counter. This allows QC marks to be used as memory location points.

Transkoder - QC window
Transkoder - QC window

To give a rate to the entry, click on the RATE stars.

You can also manually enter any text as a comment by clicking on the COMMENTS field.

The POS pull-down menu will tag the area of concern (tags comprise ALL, UL, CL, LL, UC, C, LC, UR, CR, LR, where UL=upper left, CL=center left, LL=lower left, etc.). To change the POS tag, hold the Ctrl key and click on the particular area of interest on the screen.

The TIME drop-down menu selects between the following options:

TIME Option Description

SHOT

This refers to the entire shot and disables the timecode.

FRAME

This refers to the current frame.

RANGE

This has a start and end timecode.

Move the playhead into position, hold the Ctrl key, and click the timecode field in the QC window to update either the start or the end timecode.

16.3.2. Clip Metadata Inspector

All original file metadata for the current media or CPL file can be viewed by opening the Clip Metadata Inspector window with Shift+U. The Metadata Inspector has four tabs:

  1. Video: original file metadata for video tracks

  2. Audio: original file metadata for audio tracks

  3. D: Dolby Vision metadata values from XML

  4. TL: metadata for CPLs

    Video tab

    It contains all original metadata for the video file, such as camera format, width and height of image, file name, frame count, timecode, etc.

    Audio tab

    It contains metadata for the audio file, such as file name, audio codec type, camera type, camera format, etc.

    D tab

    It contains all Dolby Vision metadata values from the XML file, such as frame number, unique ID, offset, lift, gain, gamma, chroma, etc.

    TL tab

    It contains all timeline metadata for CPLs (either IMP or DCP) in a structured way, such as annotation, package size, language, creation date, loaded date, audio channels and layout, standard, etc.

Video tab of the Clip Metadata Inspector window
Video tab of the Clip Metadata Inspector window

16.3.3. QC Table

Press Ctrl+Q to open the QC Table that is a floating window to see all of the QC entries of the current timeline. To instantly jump to the marked frame, click any line in the table.

Transkoder - QC table and QC window
Transkoder - QC table and QC window

16.3.4. Quick Button Comment Labels

You can accelerate the QC workflow by using quick button comment labels. This way, you can quickly add a frequently used comment to the QC entry by clicking on any of the quick comment buttons. Quick button labels can be defined in the QC section of the Settings Page. Enter the preferred labels in the QC Comments line, separated by commas.

Restart Transkoder to see the newly added labels in the QC window.
Transkoder - QC section in the <<Settings Page>>
Transkoder - QC section in the Settings Page

16.3.5. QC Report

To generate a PDF report that includes thumbnails, basic metadata, and QC comments, select the File  Generate PDF Report menu option. The report is written to the default Reports folder on the system. The QC operator and other information included in the report header can be entered or modified in the QC tab of the Settings Page.

16.4. PSNR Analysis

Transkoder uses Peak Signal-to-Noise Ratio analysis to measure the difference between the signal of the original project image and the error or noise in the resulting render image, which is introduced by the codec during compression. The validity of this analysis is reliable only if images of the same content are being evaluated with the difference being the compression of one image versus the other. In Transkoder, the resolution of the images must be the same as well.

The smaller the PSNR value, the bigger the difference between the images. Identical images would produce a PSNR value of infinity. PSNR is used to roughly guess the level human perceptibility of the differences between images. The table below shows the commonly used ranges and their level of perception of image fidelity between the two images.

PSNR Range Perception of Image Fidelity

Infinity

Actually identical images

>100

Barely measurable differences between images.

>45-50

Differences between images are mostly imperceptible to humans

<40

Differences between images are starting to become apparent to humans

0

Images are completely different

To create a PSNR analysis, start by creating a result render of the timeline using the desired codec. Next add the rendered clip to the bin and using Alt+V add it also to the V2 track on the timeline. Then turn on the Timeline  Timeline Mode  Enable All Tracks As Source menu option. On the Node page (Enter), a second source will appear. A PSNR node needs to be added to the pipeline and both sources need to be added a inputs and result render needs to be connected as the output. If the V1 track of the timeline is made up of multiple shots, press a Shift+T to save as a template, before proceeding back to the main screen. If the V1 track consists of only a single shot, it is OK to return to the main screen without saving as a template (Shift+T not needed).

On the main screen, the image shown by default will be the difference between the two images, with a uniformly grey screen indicating no difference. Lighter areas reflect a positive numerical difference and darker areas reflect a negative numerical difference. To configure what is being shown on the main screen, bring up the PSNR HUD by pressing H. The PSNR HUD will display the PSNR value of the current frame, allow the configuration of the blend mode, and setting a diffusion multiplier. The diffusion multiplier is used to increase the visual difference to make it more noticeable for QC purposes. The possible options for the BlendMode values are the following:

Blend Mode Description

Difference

This is the default and is used to show the differences between the two images.

Picture 1

This will show the original image from the timeline V1 track.

Picture 2

This will show the result render image from the timeline V2 track.

Red Difference

This will show the differences between to two images in the Red color channel only.

Blue Difference

This will show the differences between to two images in the Blue color channel only.

Green Difference

This will show the differences between to two images in the Green color channel only.

Transkoder can also use the PSNR values to create a graph, which is very useful for QC purposes. Rendering errors like image blockiness are easily identified by looking at the graph. Any encoding issue causing significant difference between the encoded frames and original and indicated by drops in the PSNR graph.

To create such a graph, make sure to have the PSNR node connected, and go to the Encode Page by pressing the P key, and add a PSNR encoder to the result render from the Metadata encoder sub-list of the Encoder list. Leaving the File, aka FILE PATH value as default, will place the newly created *.psnr file into the necessary folder structure needed for Transkoder to automatically detect it. Finally, press Ctrl+R to render result.

Currently the PSNR encode must be done on a timeline which has the full length, meaning untrimmed, source clips. The values calculated in the *.psnr file will pertain to the whole untrimmed clips, but they will automatically be calculated for use in any subsequent timelines created in the project from the same source clips even if they have been trimmed.

Making sure that the V1 track from the timeline is setup as it was when the PSNR render was done, select the Window  Show Graph Window menu option, or press Ctrl+comma to bring up the Graph window. The Graph window will automatically display the PSNR values as a graph. The right-hand side of the graph will show the PSNR value of the current frame, and the Average, Minimum, and Maximum PSNR values of the entire selection, which is the entire timeline by default. To learn how to set a different range or otherwise navigate the Graph window, please refer to the section titled Navigating the Graph Window. For QC purposes, the Minimum value is the one that must not cross over what is considered the acceptable threshold, such as less than 45 since this for the frame with the greatest difference between the two images. On the graph, the Minimum value is automatically marked with a dotted red line.

16.5. DCP and IMF Validation

For DCP or IMF package validation, it is mandatory to manually inspect the entire result render both visually and acoustically during the QC process. Transkoder has a great tool set available to help with both the manual QC process and also for automatic technical validation. To open the Validation window, press Ctrl+D. Transkoder detects the loaded CPL type and offers the apropriate validation test plan.

16.5.1. DCP Validation

For DCP validation, this will run a standard validation library that verifies the validity of the requisite XML files, checksums, etc.

16.5.2. IMF Validation

For IMF validation, there is a choice to use either the Colorfront developed Transkoder validation library, or extensive Photon, an open source library, originally developed by Netflix. The desired library may be set in the Validation Library field, located at the bottom left-hand side of the Validation window. Currently, Transkoder is the only tool in the industry to have integrated Photon validation library support. Colorfront developed validation in Transkoder has been improved to be fully compliant to the latest specifications, including some intricate details that are not captured by Photon validation but are detected with Baton.

16.5.3. Validation Window

The Validation window has the following columns from left to right: #, Test, Description, and Result. The possible values for Result are Pending, OK, Warning, or Error and will be detailed in the report once the validation has been completed. The bottom left-hand side has a Browse button, with path to the right of it, and for IMF validation only, the Validation Library field with drop-down selector. The bottom middle left has the Start, Cancel and Reset buttons.

The bottom right-hand side has the Open button. The Open is used to open the validation report after it has been generated.

The bottom right-hand side has the FixXMLs and Open buttons. The Open is used to open the validation report after it has been generated. The FixXMLs button allows the user to update all the hash and size values in the CPL / PKL / Asset Map. This allows the user to edit the CPL manually, which invalidates the package, and then fix the hash and size values to reflect the edited files.

Validation Window
Validation Window
Fix Hash of IMP XML files after manual edit workflow:
  1. Generate the package.

  2. Open the XML files and edit them if needed.

  3. In the Validation window run the Transkoder test. It is expected to fail.

  4. Press the FixXMLs button.

  5. Run the validation again. The package should validate.

Fix package after manual edit workflow
Fix package after manual edit workflow

16.5.4. PSNR Analysis for DCP and IMF

JPEG2000 based encoders such as those used for DCP and IMF packages, may be configured on the Encode Page configuration pane to create a PSNR encode by setting the PSNR field value to ON. Immediately after the result render has been completed, the PSNR values may be checked through the Graph window. As an added bonus, the Bitrate data will also be display.

The Graph window is used to show the PSNR, Bitrate and other values and is opened by selecting the Window  Show Graph Window menu option or pressing CTRL+COMMA.

To move the graph around in the window, right-click and drag. To zoom in and out, hold down the CTRL key while right-click and dragging up and down respectively. To center the graph in the window, thus making the entire graph visible, press the c key. To move to any arbitrary image frame by driving the playhead around, left-click in either the graph window or along any place on the timeline. The current frame, or playhead position will be marked with a bold solid red line. To select a range, hold the SHIFT key and left-click and drag to define the desired range which will be marked in green.

To change what data is shown in the graph window, click on the gray button in the top right corner of the window. Possible data options are the following: PSNR, Bitrate, Light level, and PQ (L1 Metadata). To select data to be shown in the graph window, click on the box located to the right of the desired data. Highlighted means Enabled, and darkened means Disabled. It is possible to show 2 sets of data at the same time.

16.7. Validation Metadata

Transkoder stores some metadata such as PSNR, bitrate, and file integrity checksums (SHA1 hashes) in the file system instead of the database. These are stored in the ContentDescriptors sub-folder of the main project folder by default. If the default is left unmodified, Transkoder will know where to find this metadata and automatically load it when needed. The metadata stored here is in simple text format, and may be used by custom scripts, tools, etc.

17. High Dynamic Range (HDR)

Transkoder has a range of tools to manage HDR workflows, including:

  • a sophisticated color volume transformation tool for SDR to HDR, HDR to HDR, or HDR to SDR dynamic range remapping and legalizing

  • SDR simulation mode for monitoring SDR images on a HDR display

  • a set of advanced QC and image analysis tools designed for HDR

  • calculation and display of light level statistics

  • full support of static HDR10 and dynamic Dolby Vision and HDR10plus metadata, including metadata generation, adjustment and color remapping

  • delivering HDR masters, such as HDR10 or Dolby Vision IMPs, HDR10 or HLG HEVC or ProRes files, etc.

  • adding the Dolby EDR transfer function metadata to SMPTE DCP CPL files indicating PQ color encoding, which is necessary for Dolby Cinema playback

  • updated forensic watermarking algorithms optimized for HDR content

These tools are described in more detail in the following chapters.

17.1. SDR to HDR Expansion

Transkoder’s flexible node pipeline allows users to implement any kind of color space transformations by designing specific processing pipelines. The CSC (Color Space Convert) node is a general tool to change the color representation by defining the input and output primary colors and gamma curve. However, this tool is not designed to address dynamic range mapping.

SDR to HDR mapping with the HDRSDRRemap node
SDR to HDR mapping with the HDRSDRRemap node

To properly map SDR to HDR use the HDRSDRRemap node. In addition to the CSC node where the user specifies the input and output curves and color primaries, the intended display brightness level (nits) can also be specified. Transkoder uses a custom color volume transformation algorithm - based on human visual perception - to remap the dynamic range and color volume while preserving the artistic intent of the original source material. The tool also works well for HDR to HDR remapping. For example, 1,000 nits to 4,000 nits, or 4,000 nits to 1,000 nits.

Supported Curves and Color Primaries for the HDRSDRRemap node
InPrimary OutPrimary InCurve / OutCurve

Rec709, Rec2020, P3-D65, P3-DCI

Rec709, Rec2020, P3-D65, P3-DCI

PQ, HLG 1.2, HLG Variable, Gamma 2.4, Gamma 2.6

In case you want to remap from a source color space not included in the list of the HDRSDRRemap tool, an additional CSC node has to be added to the pipeline before the HDRSDRRemap node, as CSC supports all available color spaces.

For example, if you want to remap from a XYZ color space, first convert it with the CSC node to one of the supported source color spaces of the HDRSDRRemap node, then convert it again with the HDRSDRRemap to the color space of your liking.

17.1.1. Finetuning of HDR Remap

The SDR to HDR mapping is not content aware, thus the result may need manual tweaking. This kind of finetuning of the HDR remapping can be done using the below sliders on the HUD of the HDRSDRRemap node.

HDRRange:

This parameter affects the highlight range to be boosted. A minimal value means most of the tones will preserve their luminosity and only the top of the highlight range will be brighter.

HDRAmount:

This parameter controls the strength of the remap effect. The zero value will essentially generate an SDR output.

ColorVolume:

This parameter increases or decreases the volume of the color gamut.

These controls are shot specific, so the user may have to cut the long-play timeline into shorter segments corresponding to scenes or takes.

17.2. Simulation of SDR using a HDR Display

By turning off the HDRMode toggle, one can simulate the standard dynamic range Rec.709 output on a HDR monitor, having the output color space set up for the proper handshake with the given HDR monitor. This powerful option allows the user to check both the HDR and SDR version of the same content using a single HDR monitor.

SDR simulation on a HDR monitor with the HDRSDRRemap node
SDR simulation on a HDR monitor with the HDRSDRRemap node

17.3. HDR to SDR gamut mapping

The HDRSDRRemap node can be used to map HDR content to SDR as well. Set up the input and output color spaces and light levels, and the color volume transformation algorithm will take care of the proper color remapping.

17.4. Legalizing Images for HDR

Every now and again the HDR media on the timeline has a few pixels violating either the brightness or the gamut constraints. A practical example would be when working with compressed media, where the lossy codec is causing overshoots near high contrast edges in the image. For more information on how to check this, please refer to QC Tools.

All images processed by the HDRSDRRemap node pass through the color volume remapping engine, which is designed to never output illegal color values. Thus, if the input color and brightness is configured identical to the output, it acts as a legalizer.

17.5. Light Level Statistics (MaxFALL and MaxCLL)

Transkoder is capable of computing and storing the average and the maximum frame light level for each frame, so the user can inspect the statistics of these on a graph in the Graph Window. Whenever needed, the system calculates on the fly the Maximum Frame Average Light Level (MaxFALL) and the Maximum Content Light Level (MaxCLL) values for the current timeline. These values are based on the CEA Standard CEA2861.3 "HDR Static Metadata Extensions".

Light level calculations are designed to work on PQ encoded images, so make sure the result used to generate the statistics are in PQ.

The per-frame statistics are automatically generated using the input frames when the user renders an HDR deliverable as the following examples:

  • ProRes with the Mastering Display parameter set to PQ encoding

  • HEVC with the Mastering Display parameter set to PQ encoding

  • IMF App2E+ video MXF with color space set to "Color 7 (Rec.2020)"

You can generate the statistics for any content already encoded by adding a frame light level metadata encoder (select it from the Others category in the Encoder List) and executing a render. Per-frame statistics can be generated using the File  Generate Light Level Stats menu item from the File menu. The result can be inspected in the Graph Window, or in the PDF report generated automatically to the Documents folder of the system.

By encoding video with lossy codecs, the MaxFALL and MaxCLL values will be different when computing the statistics on the input frames, versus computing them on the encoded frames. Even if the visual fidelity of the encoding is considered to be very good, the MaxFALL and MaxCLL values may change significantly.

To display the per-frame light level statistics, please open the Graph Window by pressing Ctrl+, (comma). This window displays all available per-frame metadata of the current image content on the timeline (bitrate, light level or PSNR information). You can enable or disable these in the top-right corner selector.

Per-frame light level statistics in the graph window
Per-frame light level statistics in the graph window

The frame corresponding to the most extreme value (e.g. the frame containing the brightest pixel) is marked in the graph with a vertical dashed line. This makes it easier to navigate to the potential problem on the timeline.

17.5.1. Exclude Segments from Statistics

Very often the timeline contains parts (custom head formats with slates, logos and/or color bars) that should not be considered in the MaxFALL and MaxCLL calculation. This can be achieved by adding Reel Markers to the timeline. Press I to mark the in frame and O the out frame. Only frames within the markers will be considered in the MaxFALL and MaxCLL values. This works interactively, even after the per-frame statistics have been computed.

Having reel markers on the timeline also changes how IMF packages are created: the composition will only include the marked segments. Please disable the reel markers option on the Finalize window to avoid this.

17.5.2. Manually Override Light Level Values

When creating an IMF package, the user may have premeasured light level values that are different from the statistics defined by Transkoder. For this case, there is an option for manually entering the MaxFALL and MaxCLL values into HDR recipes. In the Advanced IMF Settings of the Settings Page, custom values can be specifed in the MAXFALL override and MAXCLL override fields to override the light level analysis of the software.

Overriding light level statistics in the Settings Page
Overriding light level statistics in the Settings Page

17.5.3. Cut Detection Based on Light Level Analysis

The Cut detection feature splits up the footage into scenes (in other words, shots) that are sequences of frames shot uninterruptedly by one camera. In Transkoder, the method for cut detection is based on the calculations of the light level analysis. Cuts will be added to the positions in the clip where one scene is replaced by another, defined by sudden statistical data change among the light level values.

The Timeline  Cut Detection based on LightLevel feature will automatically cut the timeline based on the available light level statistics. If there are no light level values calculated yet, you will be asked if you want to render the statistics. Then, the timeline will be cut into short segments with same light level values for each shot.

Cut detection based on light level statistics
Cut detection based on light level statistics
This feature works only on a Windows platform.
Manual Correction of Cut Boundaries

If cutting was not performed correctly by cut detection, you can manually modify the cut boundaries between scenes, as follows:

  1. Move the playhead (yellow vertical line) on the timeline to the desired position with the up and down ( and ) keyboard arrows. The playhead’s position will be the first frame of the following segment.

  2. Click on this subsequent shot to select it so it will be highlighted in yellow. You can also highlight the preceding shot (right before the playhead) by holding Ctrl and clicking on it, but it is not required.

  3. Click Edit  Join Clip, or press the Shift+J shortcut. The two shots before and after the playhead will be merged into one segment.

  4. Move the playhead to the position you wish to make a cut in, and select the shot by clicking on it.

  5. Click Edit  Cut Clip, or press the Shift+C shortcut to make the cut. The shot will be cut into two segments.

Manual correction of cut detection
Manual correction of cut detection

17.6. QC Tools in HDR Mode

17.6.1. Analyzer Window in HDR Mode

To verify if the light level for the timeline content does not exceed the maximum brightness constraints, first please specify the maximum light level in the Maximum Brightness for HDR field (in nits) on the QC Settings section of the Settings Page. Having this configured properly, one can open the Graph Window with Ctrl+, (comma), and inspect if any segment of the maximum frame light level is marked red, indicating values above the threshold.

By changing the Analyzer Range in the bottom of the Window menu to HDR, the Analyzer is configured for HDR Quality Check:

  • The waveform shows nit levels and marks over-the-threshold levels with red.

  • The standard vectorscope becomes a CIE xy graph, indicating the P3 and the Rec.2020 triangles, and also indicating colors out of the P3 gamut with red.

The HDR analyzer also has two bars with peak-holds. These bars indicate if the maximum brightness and the color volume limit is reached on the current image, as follows:

Brightness

This bar indicates the brightness of the brightest pixel in nits. The 100% level can be configured in the Maximum Brightness for HDR field on the QC Settings section of the Settings Page.

Gamut

If the Window  HDR Color Space menu option is set to Rec.2020, indicating that the source is in the Rec.2020 PQ color space, this bar shows how the most saturated pixel approaches or exceeds the boundary of the P3 color volume. If the Window  HDR Color Space menu option is set to P3, the bar never goes above 100%.

In HDR mode the analyzer shows a CIE xy graph with the P3 and Rec.2020 rectangles with Brightness and Gamut bars
In HDR mode the analyzer shows a CIE xy graph with the P3 and Rec.2020 rectangles with brightness and gamut bars

For the HDR Analyzer to be able to show the proper gamut information, the user needs to specify the current color space. This should be either Rec.2020 or P3D65, corresponding to the current result. The default value is Rec.2020, so please make sure to adjust it when working on native P3 HDR footage. This may be done using the Window  HDR Color Space menu option.

17.6.2. Tolerance Threshold

Very often the HDR statistical analysis is executed on compressed or YUV media. Both the compression artifacts and the step of the RGB reconstruction from YUV samples cause pixel values to appear that did not exist in the original source. This, in fact, "throws off" the MaxCLL computation, and causes the brightness and gamut tests to fail.

To avoid this issue, Transkoder uses configurable tolarence thresholds, and allows a small fraction of "outlier" pixels to be out of limits, thus the result is a better representation of the true HDR image. To adjust this, open the Settings Page and change the GamutTolerance or the BrightnessTolerance values, expressed in percentage of pixels in the image.

17.6.3. HDR Warning View

To find pixels above the maximum brightness limit, you have to turn on the HDR Warning View by selecting the View  HDR  Show HDR Warning menu item, or pressing Shift+F8. In this mode, the image is desaturated, and only the problematic pixels are displayed red.

HDR Warning View is designed to work on PQ encoded images, so make sure the active result is in PQ.

In HDR view mode, pixels with above-threshold PQ light levels are marked red in the GUI
In HDR view mode, pixels with above-threshold PQ light levels are marked red in the GUI

17.7. HDR Deliverables

There are multiple deliverables in the Encode Page that can be configured, or are already configured to handle HDR content. ProRes and HEVC encoders have a Mastering parameter indicating the Mastering Display metadata. If this is set to an HDR preset, Transkoder automatically:

  • Generates the per-frame light level statistics

  • Embeds the (HDR-10) Mastering Display metadata into the files including the MaxFALL and MaxCLL values into ProRes and HEVC

  • Embeds ST2086 metadata into the MXF header per ST2067-21:2016 standard, and the MaxFALL and MaxCLL into the CPL for App2E+ IMF

  • Generates an HDR PDF report with details on light levels and a screen grab of the frame with the brightest pixel

Options for the HDR ProRes encoder presets to control the Transfer Characteristics via the Mastering field are the following:

Off

This is used for standard 709 video.

Rec.2020 PQ 1000 nits

This is used for PQ encoding curve and Rec.2020 primaries.

P3D65 PQ 1000 nits

This is used for PQ encoding curve and P3 primaries at D65 white point.

Rec.2020 HLG

This is used for Hybrid Log-Gamma encoding curve and Rec.2020 primaries.

Mastering field to control the transfer characteristics metadata indicating the mastering display
Mastering field to control the transfer characteristics metadata indicating the mastering display

As a reference please see the AMAZON HDR deliverable with Rec2020_PQ_1000 Mastering Display. This sets the color metadata to Rec2020 PQ in the ProRes per-clip and per-frame metadata with applying the correct Rec2020 RGB→YUV matrix. Then when importing such clip into Transkoder it detects this color metadata and applies the correct Rec2020 matrix from YUV to RGB. This metadata may be viewed with MediaInfo.

Users can add custom mastering display metadata presets by renaming the sample file to MasteringDisplayMetadata.xml in C:\Program Files\OSD2017\user and editing or adding parameter sets.

17.7.1. Customizing ST2086 Metadata for IMF

Very often it is needed to customize the properties of the master display metadata or the minimum and maximum light levels in the metadata injected into the encoded files. By default Transkoder auto-selects the appropriate set of metadata based on the Color setting of the IMF encoder. However, it is possible to configure the IMF encoder such that the Color setting (e.g. Color7:Rec2020) is different from the master display setting in the ST2086 Metadata: for this the Mastering option should be adjusted to some value, other than the "Default".

For example if the encoding color space is Rec.2020 but all colors are within the P3 gamut, it may be required to set the Color parameter to Color7:Rec2020 and the Mastering parameter to P3D65.

To configure the minimum and maximum brightness of the mastering display, or readjust the primary chromaticities one needs to open and edit the \pymodules\imp\st2086.conf file.

17.7.2. Customizing ST2086 Metadata for HEVC

The HDR HEVC encoders have a Mastering parameter that control the embedded HDR metadata. Either choose from one of the presets, or edit the config file at \user\MasteringDisplayMetadata.xml under the install home to add more custom presets.

HEVC with HDR Metadata and SEI Message to Trigger HDR TVs

HEVC HDR metadata support includeds support for Alternative Transfer Characteristics SEI message to trigger devices reading this metadata into the proper display mode. This is driven by the same setting that controls the Transfer Characteristics (BT2020 and HLG) and the Master Display metadata setting.

17.8. HDR PDF Report

By rendering an HDR deliverable discussed above the light level calculation and the PDF report generation is automatically triggered. One can also generate an HDR PDF report by adding a light level metadata encoder and launching a render. Gamut information is also shown in the report to flag media where the colors fall outside the P3 color volume. Make sure the Analyzer color space setting is set up properly under the Window  HDR Color Space menu option.

HDR PDF report is generated with any HDR deliverable
HDR PDF report is generated with any HDR deliverable

The calculation of light level and gamut statistics is done so tolerance thresholds are incorporated. The factory default settings allow a small fraction of pixels to exceed the brightness and gamut thresholds, so operators are able to check compressed source files without being always flagged as FAIL.

17.9. Dolby Vision Workflow

Transkoder has a complete set of tools for the creation and the inspection of IMF packages with embedded Dolby Vision metadata, as well as image sequence masters and any other video format with corresponding Dolby Vision mastering XMLs.

The Dolby Vision technology developed by Dolby Labs provides an end-to-end HDR workflow based on dynamic metadata. The Dolby Vision metadata is bundled with some HDR content, typically mastered to 4000 nits. The scene-specific metadata, which includes statistical components (L1 metadata) and optional creative adjustments for one or more target displays to fine-tune the default remap (L2 metadata), allows the Dolby Vision algorithm to produce high-quality pictures on a wide variety of displays. This metadata travels all the way to the Dolby Vision enabled TV sets.

17.9.1. Dolby Licensing

Transkoder supports efficient Dolby Vision dynamic metadata creation, including the creative trim pass, IMF mastering, metadata QC and adjustments. By default, there is no additional license needed to use the following functions:

  • Importing and exporting Dolby Vision master metadata

  • Offset metadata track to match video content

  • Usage of the internal remapping algorithm for QC

  • Rendering the source to any target display

  • Creating and importing a Dolby Vision IMF

The only function that still requires a license from Dolby is to specify new L2 trim metadata, i.e. creative trimming. The license is granted by Dolby, so please contact them directly at https://www.dolby.com.

After receiving the license, please copy the .bin license file into the license folder of OSD2018. After restarting, the sliders for trimming will be unlocked.

17.9.2. Dolby Vision Remapping

By default, it is possible to do the color remapping within Transkoder and to preview the Dolby Vision remapping to specific target displays, as well as render these results.

When the remapping is enabled, the Dolby Vision tool applies the corresponding color transform. The user can inspect and validate the effect of the Dolby Vision node by clicking the Bypass button on its HUD. The remap works in the following scenarios:

  • Green timeline: In this scenario, the DolbyVision node is initialized, and the dynamic metadata is driven by the node itself. The green line indicates that the metadata editing is enabled.

  • Orange timeline: In this scenario, the metadata is imported by means of an XML file. In this case, the current L1 metadata, and the creative trims (L2 metadata) are not shown on the sliders of the HUD of the Dolby Vision node. Please use the D tab of the Clip Metadata Inspector window to see these values.

  • Yellow timeline: In this scenario, the metadata is embedded into an MXF file of an IMF or an IMF package (IMP). Similarly to the previous case, the remap works, but the metadata is not driving the sliders.

Both the timeline and the bin thumbnails show an orange line on clips with associated Dolby Vision mastering metadata XML. Similarly, a yellow line indicates an MXF file of an IMF package with embedded Dolby Vision metadata.

Orange line indicates Dolby Vision mastering metadata XML associated with the clip
Orange line indicates Dolby Vision mastering metadata XML associated with the clip

To get the expected output in terms of color, please make sure to adjust the following settings accordingly:

  • CMU Mastering Display: defines the target color space

  • PQ Analysis Colorspace: defines the encoding of the current input

17.9.3. Create Dolby Vision Metadata

The DolbyVision node is used to interactively adjust Dolby Vision metadata. First, add the DolbyVision node to the pipeline on the node page. See details on how to Edit Nodes.

If the metadata is from an external source, such as either in the form of a mastering XML, or a Dolby Vision IMF, and it is not to be changed, then the DolbyVision node is not required in the pipeline.

The Dolby Vision node
The Dolby Vision node

The second step of Dolby Vision metadata creation is the generation of statistics. Simply press the Init button on the HUD of the DolbyVision node, so Transkoder very quickly analyzes every single frame on the timeline. When it is done, the scene-based light level values will be updated in the Dolby Vision node, so a green line will appear on the timeline indicating valid Dolby Vision metadata.

After this, the timeline needs to be notched into smaller segments corresponding to individuals scenes, hence the Init button needs to be pressed again. At this time, Transkoder updates the scene-based statistics instantly from the precomputed per-frame data.

The initial statistical calculation needs to be executed before any editing, with the source media at full length on a timeline. When statistics are generated for all the frames, it is possible to edit the timeline.

As soon as the DolbyVision node has valid statistics, indicated by the green line, you see the remap result on the monitor. Make sure to set up the target display in the Settings Page so it matches the display device you are monitoring the remap results on. A typical workflow is to connect to a standard SDR display and set the target to 100-nit, Rec.709.

If the default mapping needs fine-tuning so the result is technically correct and represents the original artistic intent, the operator can adjust the creative controls, such as Lift, Gamma, Gain, etc., to achieve the best result. It is possible to further fine-tune the remapping for different display targets: simply choose another target display from the drop-down list on the HUD, and turn on the Enabled switch. See the DolbyVision chapter for details.

When creating the pipeline in your project, please make sure the DolbyVision node is within the grade box indicated with a red rectangle. This can be done by selecting the node, pressing V, and then saving it as a template for all shots with Shift+T.

17.9.4. Enable Metadata Editing

When working with an orange (XML metadata) or a yellow (IMF) timeline, no special node is needed. Per-frame metadata can be inspected in the Clip Metadata Inspector window, and the user is allowed to QC the metadata.

However, sometimes it is necessary to change the metadata, or simply to check if shot boundaries in the metadata are in sync with the editing points on the timeline. For this, one can enable Dolby Vision editing. First, add a DolbyVision node to the node pipeline. Then, use the File  Dolby Vision  Enable Metadata Editing menu option. Thus, the timeline will become green and it will be cut into shorter segments corresponding to cut boundaries in the metadata. In addition, the DolbyVision node will be initialized with all the necessary parameters.

Please do not use the Enable Metadata Editing feature if per-frame metadata is present in the original XMLs.

17.9.5. Import Dolby Vision IMFs

When loading an IMF with embedded Dolby Vision metadata, a yellow line will be drawn over the timeline window. This indicates valid metadata from an MXF source. It is then possible to check the metadata sync points in the timeline window, and also to export the XML from such a composition.

17.9.6. Export Dolby Vision Master XML Metadata

Having valid Dolby Vision metadata on the timeline, one can export the Master XML metadata via the File  Dolby Vision  Export XML Master Metadata from active track menu option. This function is essential when checking or creating Dolby Vision IMF packages, or saving metadata for further grading in any color grading software.

When working with multiple clips on the timeline, you can export a single XML for the whole timeline as well. In this case, the exported XML will include the metadata of all the events on the timeline.

Transkoder is also able to export the Dolby metadata XML for selected clips via the File  Dolby Vision  Export XML Master Metadata from selected clips menu option. This feature is useful when adding new inserts to a Dolby Vision package, where the user needs to apply the metadata from the original content to the new parts. For this, when adding a new insert to V2, the operator needs to make edits at the head and at the tail of the insert on the V1 track to create clips matching the inserts. See the Synchronize Dolby Vision Metadata chapter for details. Metadata that matches the length of the insert can be exported from these clips.

17.9.7. Import Dolby Vision Master XML Metadata

In Transkoder, you can also do quality checking, potentially modify and deliver metadata in case of Dolby Vision IMFs. It is done by loading a timeline with the source images, and applying a Dolby Vision master XML using the File  Dolby Vision  Apply XML Master Metadata to events on active track menu option. If there are multiple events on the timeline, and none of them is selected, the metadata will be applied to the clip where the playhead is positioned to.

When importing a Dolby Vision master XML to the timeline, to make sure each frame has metadata, the length of the metadata in the imported XML must be at least as long as the full length of the source content on the timeline. If the XML represents a longer sequence, the first frame is aligned with the beginning of the XML. Then, the operator can offset it if needed, and the sync between the metadata and the picture has to be checked subsequently. If the timeline event is longer than the selected Dolby Vision master XML, however, a warning message will appear and the application of the XML will fail.

In the Settings Page, you can also enable the option Force Dolby Length Match so that Transkoder rejects Dolby Vision XMLs that do not match the length of the original source media file. When this option is enabled, the metadata can neither be longer, nor shorter than the timeline event; only metadata XMLs with a uniform length are accepted. Otherwise, an error message will pop up indicating that the application of the metadata file failed because of length mismatch.

Force Dolby Length Match option in the Settings Page
Force Dolby Length Match option in the Settings Page

You can also import Dolby Vision master XML metadata when working with a yellow timeline with embedded Dolby Vision metadata. In this case, if the length of the timeline does not exceed the length of the selected XML, the embedded metadata will be overridden, and the timeline will become orange.

Dolby Vision XML metadata can be applied not only to timeline events, but to any clips in the bin. For this, first select the clip in the bin to which you want to apply the metadata, then click on the File  Dolby Vision  Apply XML Master Metadata to Clips in Bin menu option and select the XML file containing the metadata.

To apply the metadata XML to multiple clips at the same time, for instance when working with timelines with multiple events, make sure the number of selected clips in the bin or on the timeline matches the number of XMLs being applied. Transkoder will match the XMLs and the clips based on alphabetical order, e.g., first XML is associated to the first clip, and so on. To generate a Dolby Vision IMF from this timeline, the user needs to open the Finalize Window and double-check that the Dolby option is enabled.

You can also remove the imported metadata from the timeline with the File  Dolby Vision  Clear Master Metadata from Timeline Events menu option.

The orange in the timeline window indicates valid Dolby Vision metadata from an XML file, while a yellow line indicates embedded Dolby Vision mastering metadata in the video MXF file. The metadata for the current frame can be inspected on the D (Dolby Vision) tab of the Clip Metadata Inspector window.

Dolby Vision per-frame metadata from XML is shown in the _D_ tab of the Clip Metadata Inspector window
Dolby Vision per-frame metadata from XML is shown in the D tab of the Clip Metadata Inspector window

17.9.8. Generate Dolby Vision IMFs

The Finalize Window has the ability to enable the Dolby Vision option on the Encoder tab. This option enables the embedding of per-frame Dolby Vision mastering metadata into IMF video assets. If the source is an IMF without Dolby Vision metadata, the user can manually load the corresponding XMLs to the timeline, and inject the metadata very efficiently by transwrapping. Some IMF presets, such as the IMF NETFLIX HDR (App2E+), have this option enabled by default.

The Dolby Vision option has three settings:

If Available

Dolby Vision metadata is injected into the IMF package only if there are such metadata available.

On

Dolby Vision metadata is injected into the IMF package. If this setting is enabled, Dolby Vision mastering metadata must be associated with the timeline, otherwise the encode process will fail.

Off

Dolby Vision metadata will not be injected into the IMF package, even if available.

The above Dolby Vision settings are only available for certain HDR formats, selected from the Format drop-down menu in the Settings tab. For other formats, such as SDR, the setting will be Off.
Dolby Vision IMF option in the Encoder tab of the Finalize Window
Dolby Vision IMF option in the Encoder tab of the Finalize Window

Dolby Vision IMF can be generated from a timeline with XML metadata (indicated by an orange line), or a timeline with IMF track files with embedded Dolby Vision metadata (yellow line), or directly from a timeline with a Dolby node (green timeline).

17.9.9. Load Dolby Vision metadata XML as a sidecar file

In automated workflows it may be necessary to automatically associate the metadata XML with the image files. For this place a Dolby Vision Mastering Metadata XML in the same folder as the media files and name it dolbyvision.xml. This will be automatically parsed and associated with the clip.

17.9.10. Inject Dolby Vision metadata into existing App2E+ IMP

Load an IMP without Dolby Vision metadata and import the corresponding Dolby Vision mastering metadata XML to the timeline (use the File  Dolby Vision section). Make sure the metadata and the IMP reels do have matching lengths. Having this set up, one can turn on Dolby Vision metadata for the IMF encoder and create a new package by transwrapping the original image essence along with the per-frame Dolby Vision metadata embedded. To avoid reencoding of the video, make sure the IMF encoder parameters match the source IMP.

17.9.11. Inspect Dolby Vision metadata

The Clip Metadata Inspector window showing Dolby metadata on the D tab
The Clip Metadata Inspector window showing Dolby metadata on the D tab

The Clip Metadata window has a D tab, where the Dolby Vision mastering metadata can be inspected. The values reflect the metadata for the current frame, either coming from an XML file or embedded in the MXF itself. See details in the Clip Metadata Inspector chapter.

17.9.12. Show Dolby Vision Dynamic Metadata Scene Boundaries

When importing Dolby Vision XML Master Metadata to timeline events the operator can now see in the Timeline Window where the scene boundaries are in the metadata. This allows easier checking of picture and metadata sync, which is essential in Dolby Vision workflows: as part of the QC process one has to make sure the changes in the metadata correspond to the real editing points.

Scene boundaries in the Dolby Vision metadata shown in the Timeline Window
Scene boundaries in the Dolby Vision metadata shown in the Timeline Window

To jump to metadata scene change points please use the Alt+> (greater than sign) and Alt+< (less than sign) shortcuts.

17.9.13. Synchronize Dolby Vision Metadata

It is possible to apply Dolby Vision XML master metadata to timeline events even if the metadata length is longer than the source video. In this case, the user is able to offset the dynamic metadata. To validate that the sync is correct between the metadata and the image, please inspect the Dolby metadata edits (vertical dashed lines) on the timeline.

Manual Synchronization of Metadata

To manually offset the metadata, hold Ctrl+Shift and drag the video event with the mouse on the timeline. Provided that the XML metadata is longer than the video, the scene boundary markers should move as you slide the metadata manually.

Metadata Synchronization with Batch Edit

You can also offset the metadata by opening the Batch Edit window Ctrl+E. Enter 0 as the starting frame number into the Slip Dolby XML field to synchronize the metadata to the beginning of the video file.

Slipping the Dolby Vision XML Master Metadata in the Batch Edit window
Slipping the Dolby Vision XML Master Metadata in the Batch Edit window

17.9.14. Copy Metadata from V1

When working with timelines where, besides the original media file on the V1 track, there are multiple clips on additional video tracks (V2, V3, etc.) as well, you can apply Dolby Vision dynamic metadata to them in a simple way. Choose the File  Dolby Vision  Copy Metadata Up from V1 menu option to copy the Dolby Vision XML master metadata from the active V1 video track to the clips on V2, V3 or other video tracks.

Copy Dolby Vision metadata from V1 to V2 track
Copy Dolby Vision metadata from V1 to V2 track

17.9.15. GPU vs. Legacy Remap

In Transkoder, the Dolby Vision remapping is implemented on the GPU architecture, which is a much faster color remapping method than before, when a third party component was required for the process.

The GPU implementation of the color remapping is Dolby certified.

Occasionally, it may be favorable, however, to switch from the fast remap option, and use the legacy option for color mapping. For this, go to the Dolby Vision tab of the Settings Page, and change the Render option from Fast to Legacy.

In this case, you also have to install the Dolby Vision Professional Tool set containing the cm_offline.exe file component. Then, copy this .exe file into the C:\Program Files{app-installation-folder}\pymodules\scripts folder to enable the remapping through the legacy option.

Dolby Vision remap with Fast GPU and Legacy options
Dolby Vision remap with Fast GPU and Legacy options

17.9.16. Hardware Configuration with CMU

In some cases, a special hardware, a Content Mapping Unit (CMU) is needed from Dolby to be able to visualize the color remapping based on Dolby Vision metadata to different target displays. CMU is a one rack unit computer with SDI video I/O in HD resolution. As Transkoder is capable of outputting two independent video output signals with different resolution or color pipeline, it is possible to set up a configuration with an UHD HDR display (to inspect the original HDR master) and an HD SDR monitor connected through the CMU. For the two independent video outputs to work, a Corvid 88 card is needed.

The layout of the two possible CMU configurations
The layout of the two possible CMU configurations

To enable Transkoder to drive the CMU, make sure to adjust the following project settings:

  • Video output bit depth: 12

  • Video resolution: HD

  • Dolby Vision setting: enable CMU

  • Dolby Vision setting: fill the CMU Device 1 IP address with your CMU unit’s IP

17.10. HDR10plus Workflow

HDR10plus is an open standard for dynamic HDR metadata. In contrast to static HDR10, this standard provides a per-scene dynamic tone mapping curve so TV sets are able to apply a proper remapping of bright pixels that would clip without adjustment. Transkoder on the Windows platform can generate this metadata and also apply the remap itself to the image.

17.10.1. Generating HDR10plus Metadata

To generate HDR10plus metadata, the user needs to load the HDR master encoded with the PQ curve, and add it to the timeline. The metadata statistics are generated on the Encode Page with a metadata encoder called HDR10 Plus, selected from the Others list. After adding the encoder, the user needs to render the file with Ctrl+R. The render result will be a human readable JSON file containing the mathematical coefficients defining the tone map curve for each scene. By default, the encoder automatically detects cuts between scenes based on the light level statistics.

Encoder to generate the HDR10plus metadata
Encoder to generate the HDR10plus metadata
The location of the saved JSON file is shown in the Encoder table. You can also modify the final render path by clicking on the path name.

The metadata in the generated file targets a specific brightness level, defined in nits in the TargetLevel parameter of the encoder. The maximum brightness of the source master should be set in the MasteringLevel setting.

17.10.2. Quality Check HDR10plus Metadata

To visualize or quality check the result of the dynamic remap within Transkoder, please add a HDR10plus node to the node page first. Then the user can import the generated JSON file through the Timeline menu by clicking on Cut Detection with HDR10plus JSON. This operation adds the cuts defined in the JSON file, as well as populates the HDR10plus node with the appropriate metadata. As soon as the metadata is loaded into the nodes, the remap is processed. To validate it, park on a frame with bright pixels exceeding the nit level defined in the TargetLevel of the metadata, and toggle Bypass on the HDR10plus node.

HDR10plus node
HDR10plus node

17.11. Forensic Watermarking for HDR

In Transkoder, there are updated forensic watermarking algorithms optimized for HDR content. Forensic watermarking is available in HDR encoders, such as IMF App2e+, Amazon UHD HDR ProRes and 16-bit DPX. The new algorithm is automatically triggered by the HDR metadata.

See the Forensic Watermarking chapter for details.

Appendix A: Validation

Transkoder has a dedicated validation procedure for DCP and IMF packages as well in the Validation window (CTRL+d). The path at the bottom of the window should point to the package folder to analyze. The validation process consists of many steps involving checking the proper format of metadata as well as analyzing samples from the encoded assets. Should there be any failure please check the detailed error report in the current user’s Document folder, which can be opened by pressing the Open button after the validation finished.

A.1. DCP Validation

The following checks can be performed:

A.1.1. Metadata

Assetmap exists

Checks if the assetmap file exists in the root of the DCP volume and if it is correctly named.

Volume index exists

Checks if the volumeindex file exists in the root of the volume and if it is correctly named.

Assetmap is valid

Checks if the assetmap file structure is valid by checking it against the related xml schema.

Volume index is valid

Checks if the volumeindex file structure is valid by checking it against the related xml schema.

Assetmap contains PKL

Checks if the assetmap file contains at least one Packing List (PKL) entry.

Assetmap files are existing

Checks if every file listed in the assetmap exists in the DCP volume at the path specified in the assetmap.

Files exists in assetmap

Checks if each file in the DCP volume is listed in the assetmap.

Assetmap chunk length is valid

Checks if the referenced files sizes match the related assetmap files 'Length' entry.

A.1.2. Packing List (PKL)

PKL exists

Checks if the Packing List files listed in the assetmap exists in the DCP volume at the path specified in the assetmap.

PKL is valid

Checks if the Packing List file structure is valid by checking it against the related xml schema.

Assets exists in assetmap

Checks if every file listed in the Packing List is listed in the assetmap.

Asset sizes are valid

Checks if the Packing Lists 'Asset' elements 'Size' entry matches the size of the related file in the DCP volume.

PKL Signature is valid

Checks if the Packing List file signature is valid if signed.

Mimetypes are valid

Checks if the Packing Lists MIME type value is correct for the referenced asset.

Asset hash is valid

Checks if the PackingLists assets "Hash" entry matches the SHA-1 hash of the related asset in the DCP volume. This test may take some time dependent on the DCP data to test.

A.1.3. Composition Playlists (CPL)

CPL exists

Composition Playlists must exist in the root of the volume and must be named correctly.

CPL is valid

Checks if the Composition Playlist files listed in the assetmap exist in the DCP volume at the path specified in the assetmap.

CPL is valid XML

Checks if the Composition Playlist files' structure is valid by checking it against the related xml schema.

Durations are valid

Checks if every Composition Playlist assets 'IntrinsicDuration' entry matches the duration of the respective mxf track file.

CPL assets exist in assetmap

Checks if every file listed in the Composition Playlist is listed in the assetmap. Note: A valid DCP may have unresolved Composition Playlist asset references.

CPL assets exist in PKL

Checks if every file listed in the Composition Playlist is listed in the related Packing List. Note: A valid DCP may have unresolved Composition Playlist asset references.

Trackfile duration matches reel duration

Checks if every Composition Playlist assets "Duration" entry added to the related "EntryPoint" entry is smaller or equal to the related "IntrinsicDuration" entry.

Asset hash matches PKL

Checks if every Compositon Playlists assets 'Hash' entry is equal to the related Packing List Hash entry.

Reel duration is valid

Checks if the duration of the asset entrys are equal to the reels duration. Reel duration is either the duration of the "MainPicture" if existing or the duration of the shortest asset in the reel.

Tracks are homogeneous

Checks if all Composition Playlists reels contains the same collection of asset types.

Reel has a picture

Checks if all Composition Playlist reels contain a 'MainPicture' or 'MainStereoscopicPicture' element.

Reel has a sound

Checks if all Composition Playlist reels contain a 'MainSound' element.

Key usage is valid

Checks if the Composition Playlist assets cryptographic keys are unique.

Tracks encryption is homogeneous

Checks if all Composition Playlists assets are encrypted if one of them is encrypted.

CPL Signature is valid

Checks if the signature is valid if the Composition Playlist file is signed.

Asset files exists

Checks if every file listed in the Composition Playlist exists in the DCP volume at the path specified in the assetmap.

A.1.4. Track Files

Trackfiles are valid

Checks if the mxf track file complies to certain definitons of the DCI specification for OP-Atom MXF trackfiles.

Encrypted trackfiles are valid

Checks if the mxf track file complies to definitons of the DCI specification for encrypted OP-Atom MXF trackfiles.

CPL languages codes are valid

Checks if the Composition Playlists language code entrys are syntactical correct and known language codes.

Track file durations are valid

Checks if the mxf track file is at least one second in duration.

Image tracks frame boundaries are valid

Checks if the mxf image track file frame boundaries are valid.

Soundtrack frameboundaries are valid

Checks if the mxf audio track file frame boundaries are valid.

Subtitle subpictures are valid

Checks if the subtilte .png subpicture resources are valid.

Subtitle font is valid

Checks if the subtitle font resources are valid TrueType or OpenType files.

Subtitle font file sizes are valid

Checks if the subtitle font resources sizes excede a maximum size of 640kb for Interop DCP volumes.

Subtitle XMLs are valid

Checks if the subtitle definition xml file structure is valid by checking it against the related xml schema.

A.2. IMP Validation

The following checks can be performed:

A.2.1. Metadata

Assetmap exists

Checks if the assetmap file exists in the root of the IMP volume and if it is correctly named.

Volume index exists

Checks if the volumeindex file exists in the root of the volume and if it is correctly named.

Assetmap is valid

Checks if the assetmap file structure is valid by checking it against the related xml schema.

Volume index is valid

Checks if the volumeindex file structure is valid by checking it against the related xml schema.

Assetmap contains PKL

Checks if the assetmap file contains at least one Packing List (PKL) entry.

Assetmap files are existing

Checks if every file listed in the assetmap exists in the IMP volume at the path specified in the assetmap.

Files exists in assetmap

Checks if each file in the IMP volume is listed in the assetmap.

Assetmap chunk length is valid

Checks if the referenced files sizes match the related assetmap files 'Length' entry.

A.2.2. Packetlists (PKL)

PKL exists

Checks if the Packing List files listed in the assetmap exists in the IMP volume at the path specified in the assetmap.

PKL is valid

Checks if the Packing List file structure is valid by checking it against the related xml schema.

Assets exists in assetmap

Checks if every file listed in the Packing List is listed in the assetmap.

Asset sizes are valid

Checks if the Packing Lists 'Asset' elements 'Size' entry matches the size of the related file in the IMP volume.

A.2.3. Composition Playlists (CPL)

CPL exists

Composition Playlists must exist in the root of the volume and must be named correctly.

CPL is valid

Checks if the Composition Playlist files listed in the assetmap exist in the IMP volume at the path specified in the assetmap.

CPL is valid XML

Checks if the Composition Playlist files' structure is valid by checking it against the related xml schema.

Durations are valid

Checks if every Composition Playlist resources 'IntrinsicDuration' entry matches the duration of the respective mxf track file.

CPL assets exist in assetmap

Checks if every file listed in the Composition Playlist is listed in the assetmap. Note: A valid IMP may have unresolved Composition Playlist resource references.

CPL assets exist in PKL

Checks if every file listed in the Composition Playlist is listed in the related Packing List. Note: A valid IMP may have unresolved Composition Playlist resource references.

Asset hash matches PKL

Checks if every Compositon Playlists resources 'Hash' entry is equal to the related Packing List Hash entry.

Segment duration is valid

Checks if the duration of the asset entrys are equal to the segments duration. Segment duration is either the duration of the 'MainImageSequence' if existing or the duration of the shortest asset in the segment.

Segment has a picture

Checks if all Composition Playlist reels contain a 'MainImageSequence' element.

Segment has a sound

Checks if all Composition Playlist reels contain a 'MainAudioSequence' element.

Asset files exists

Checks if every file listed in the Composition Playlist exists in the DCP volume at the path specified in the assetmap.

Appendix B: Settings Page

By pressing the Tab key or selecting the corresponding menu item from the drop-down menu, you can open the Settings Page where you can modify and adjust the project settings.

To scroll the page, drag the screen upwards or downwards while holding the right click.

B.1. Custom Default Settings

For frequently used project settings, you can save actual project settings as a default, or save as a project template that store all project-level settings. For details, see the Save Project as Project Template chapter.

To save new default settings based on the current project, click on Save Preferences as Default at the bottom of the panel. However, keep in mind that from that moment all new projects will initialize with these settings. Therefore, we recommend saving a project template so the operator can decide which template to use when creating each new project.

To remove custom settings that have been saved as default, click on Remove Custom Default Preferences at the bottom of the panel. This is for the cases when these settings are easily assumed by the support team.

Setting Page Description

Camera

Settings for Raw Camera Support

Avid

Settings for AVID Software

Base

Main project settings

Conform

Settings for conform workflows

Encode

Settings for Encoders

DCI

Settings for DCP mastering

IMF

Settings for IMF package generation

GUI

Settings for Transkoder Graphical User Interface

Video

Settings for video output

Audio

Settings for audio

Start-Up

Settings for workstations specific settings (saved in the Startup.xml)

Config

Settings related to hardware configuration specific optimizations

Grade

Setting for Grading and Color Processing

Metadata

Settings for the configuration of Metadata

QC

Settings for Quality Control

Subtitle

Settings for titles and captioning

User

Settings for workstations specific settings (saved in the Startup.xml)

Cloud Upload

Settings for cloud based web services such as COPRA, Netflix, Pix, and Remote Server

Cloud Storage

Settings for cloud storage based services such as AWS, Google Cloud Storage, and Swift

Support

Project and Software versions

Dolby Vision

Settings for Dolby Vision workflows

Safe Area

Settings of safe area and safe area guides

B.2. Camera Settings

The Camera settings are the following:

B.2.1. ARRI Camera Settings

Setting Description Note

ADA Apply CDL

Controls if the ADA debayering process should automatically apply CDL correction.

ADA Pixel Aspect Correction

Use ADA automatic pixel aspect ration correction to resize image

If enabled, this will automatically desqueeze anamorphic footage when using the ARRI developed debayer.

Alternate Crop

Use alternate cropping method for ARRIRAW files

If enabled, this will crop files according to the crop settings in the ARRIRAW file metadata.

Arri Metadata Bridge

Use ARRI Metadata Bridge when rendering EXR

Decode Resolution

ARRIRAW decode resolution

This allows the selection of resolution to use for debayering into viewable RGB images. + Resolution and overall quality increases from HD to CFD-5 for GPU accelerated Colorfront developed debayering, and from ADA-2-SW to ADA-5-SW for ARRI developed debayering. + ADA type debayering is significantly slower because it is done in software and is not GPU accelerated.

Force Black and White Processing

Enable or disable Black and White ARRIRAW processing

Should be enabled in instances where the Black and White processing is not correctly set in the metadata.

Full Open Gate

Use all pixels from the sensors, ignore metadata

Ignore ND Filter

For old projects to use the regular white balancing gains

Should be enabled in instances where the Ignore ND Filter processing is not correctly set in the metadata.

Saturation in Look

How to interpret the saturation in the header. CDL is not compatible with Arri Look

In the ARRIRAW pipeline, saturation processing is handled by default in the ARRIRAW node. This is a slower but higher quality process, but in some cases where the footage is to be exported for use in CDL based applications, this may be changed to use CDL instead.

B.2.2. Sony Camera Settings

Setting Description Note

Decoding

Debayering method to use

Decide between either the Colorfront developed debayering or the Sony developed one.

MXF Fast Read

Enable/Disable Sony F65 fast read

With certain storage configurations, where real-time playback is not working, enable this option to optimize MXF RAW file read patterns for faster performance.

No Crop

Crop Image to HD / QuadHD or not

If this is enabled, the image will not be cropped.

Resolution

Size of the debayered image

The option determines the debayer resolution. 8K is only supported for F65 cameras.

B.2.3. Panasonic Camera Settings

Setting Description Note

P2Format App

Path to the P2FormatApp executable

P2Format App Options

Options set to the P2FormatApp executable

Panasonic 444 Decode Type

Panasonic 444 decode quality

Decoding to YUV42210bit is faster to process but results in a lower image quality than RGB16bit.

V210 Decode Resolution

YUV42210bit H264 decode resolution

Half setting is faster to process but results in a lower image quality than Full.

B.2.4. Red Camera Settings

Setting Description Note

Color Space

RED decode color space

This setting is for selecting which color space to use for primary colors when debayering RED camera footage. The Auto option will work best in most cases, since it will use the setting from the RED camera footage metadata.

DEB

Use Dragon Enhanced Black option

Enabling this option will result in better blacks, but also in significantly slower processing speed.

Decode Bit Depth

Bith depth of decoded images

10bit is faster to process but results in a lower image quality than 16 bit.

Decode Mode

RED decode resolution

The Auto option will use the highest quality processing possible with the hardware present in the Transkoder work station. The options from Full to 1/16 range from slowest processing and maximum image quality, to fastest processing and lowest image quality and describe the debayered resolution in relation to the original footage resolution.

Default Metadata

Default image processing setting

This setting determines the default metadata to use, is from the RED camera footage or from the RMD Sidecar. If no RMD Sidecar metadata is present, Transkoder will revert to using the RED camera footage metadata.

Default TC

Default image processing setting

This setting determines which Timecode to use by default.

Disabled Rocket ID

Exclude a Rocket Card with this ID from the decode cards.

File Read Timeout

Timeout for RED image access in ms

In cases where RED render jobs fail as a result of File Read Timeouts, increasing the length of time to wait before a time out error, may help prevent these errors.

Gamma Curve

Gamma curve to apply to the image

This setting is for selecting which color space to use for tone colors when debayering RED camera footage. The Auto option will work best in most cases, since it will use the setting from the RED camera footage metadata.

HDRx Mode

Force HDRx Mode for imported RED movies

This setting determines which detection mode to use for detecting multi-exposure HDR footage in RED camera footage. Automatic: this should be used in general. On-Set Dailies auto-detects if there are two differently exposed tracks and interprets it as a HDRx shot Alternate Frames: this should be used when parsing RED media with a single track, with hi and low exposure images alternating. (looks flashing otherwise). As much as we know this was only used in older RED cameras, we were very surprised to see such a clip from you. None: OSD interprets the RED media as "normal" media even if it is HDRx. As only the normally exposed image is used no ghosting should be visible. However, Alternate frame HDRx media will look wrong (strong flashing) with this setting.

HDRx Offset

Offset for Alternative HDRx Mode

This setting determines the offset between the high and low exposure images and works for most cases, but allows for fine tuning if needed. It works only when HDRxPlus enabled.

HDRx Plus

Alternative mode for HDRx

This setting enables a different mode for the processing of HDRx images. On: The two images differently exposed are combined such that it produces a single high dynamic range image. Off: The two images differently exposed are simply mixed. Use the HDRx Bias to balance the mix based on which image has more details.

Helium Full

Using 8k resolution for Helium

IPP2 Stage

Primary Raw Development Stage

IPP2 Primary Raw Development Stage is supported by specifying Primary as the value for this field. Only Tint, Kelvin, Brightness, ISO, Exposure settings available in this stage. This mode uses RED Wide Gamut RGB color space and Log3G10 gamma curve, thus when IPP2 is enabled all the other color space options are hidden.

Ignore User CDL

Ignore the CDL values from the incoming camera metadata and start with identity

Ignore User Curve

Ignore the user curve metadata from the RMD/header

This setting is used in cases where the Ignore User Curve flag found in RED camera footage metadata or RMD Sidecare metadata should be ignored.

Lock Color Settings

If enabled pipeline creation will not change the Curve and Color setting

Enable this setting to prevent the automatic configuration of the Color Space and Gamma Curve settings when creating a new RED camera footage pipeline.

Use GPU

Use GPU or CPU for Red Debayer

If enabled, this setting will use GPU acceleration for debayering RED camera footage, and if disabled, it will use the CPU instead which is significantly slower.

Use RED CDL

If enabled, the CDL tool will adjust the RED CDL metadata

If this setting is enabled, the RED debayering API will apply the CDL processing, instead of the CDL node on the Node page pipeline. In this case, the CDL node will only act as a settings interface for the RED debayering API for CDL processing. This is less interactive with fewer configuration options than if this option was disabled and the default, Colorfront developed, CDL node processing was used instead.

Use Rocket

Enable/disable RED Rocket acceleration

Enable to use the RED Rocket expansion card for accelerated processing of RED camera footage. When working on a project with Red Rocket acceleration enabled, without the card present the processing will fall back to GPU mode. This improves multiworkstation Red projects with different hardware configurations. If you have a revision 2 Red Rocket card, turn this setting on for 4k output, otherwise turn it off.

Use Rocket for Render

Enable/disable Red Rocket acceleration for Background Rendering

If the option is enabled, Transkoder will use the RED Rocket expansion card for the backend rendering, and if disabled, Transkoder will used the RED Rocket expansion card for the processing of RED camera footage for real-time use in the frontend GUI.

B.2.5. Quicktime Camera Settings

Setting Description Note

Crop Clean Aperture

Crop the image based on the Quicktime Clean Aperture parameter

Enabling this will use the metadata information in the QuickTime file to crop the image.

QT 4K Decode Resolution

YUV8bit decode resolution

2K is faster to process but results in a lower image quality than 4K. Use 2K if realtime playback is not working with the 4K setting.

QT H264 Decoder

QT H264 decoder

In the rare cases when autodection of the decoder fails, switch between the two options if there is any problem with the image decoding.

QT Prores Decode Alpha

Decode the alpha or not from 444 formats

Disable to ignore apha channel information, thus speeding up decoding, but lowering image quality.

QT Prores Decode Type

ProRes Decode Quality

Faster processing with lower image quality when moving from RGB16bit to YUV4228bitHalf. Each move to the right, will increase this tendency.

B.2.6. Phantom Camera Settings

Setting Description Note

Phantom Hour Compensation

Compensating the Local and UTC timecodes with hours

This setting defines the number of hours to compensate the Local or UTC timecodes of the Phantom file metadata.

Phantom Main Timecode

The main timecode of the Phantom footage

This setting determines which of the four possible Timecodes to use from the Phantom metadata.

Phantom Tone Curve

Turn off the Tone Curve in the CINE file if there is one

This setting determines if the Tone Curve is processed from the CINE file or not provided one exists.

B.2.7. Indiecam Camera Settings

Setting Description Note

Force White Balance

Use white balance coefficients

Forces the image processing to use the white balance coefficients for the primaries used. This is to deal with missing/bad metadata flagging and normally should not be used.

Pattern

Type of bayer pattern

Indicates the Bayer pattern type in the source files.

B.2.8. CinemaDNG Camera Settings

Setting Description Note

Exposure Compatibility

Alternative processing of exposue to match implementation in other products

Only enable this setting when trying to match a potentially faulty output in other products from the same DNG source.

Force Half Processing

If DNG image width is greater thean 3072 this option forces half resolution for debayering

This option, if enabled, is used to speed up decoding but lowers image quality.

B.3. Avid Settings

The Avid settings are the following:

Setting Description Note

Avid Half FPS TC in ALE

Convert the 60fps TC to 30fps for older Media Composers

This setting is for compatibility with older AVID Media Composer applications that cannot fully process 60fps TC.

Avid MOBID

Generate common MOBIDs for Avid MXFs

If rendering multiple resolution results for AVID Interplay, this setting determines how to generate common MOBIDs. Master + Source is the most commonly used option.

Interplay Automatic Upload

Enable automatic upload of generated Avid mxf files to Interplay

This setting uses an AVID API to automatically upload AVID mxf result renders to AVID Interplay

Interplay Direct Write

Write directly to the Shared Folder before Interplay checkin

If this setting is enabled, Transkoder will write AVID mxf result renders directly to the AVID Interplay shared storage even before they have been checked in to Interplay.

Interplay WS Endpoint

URL of the Interplay WS

This setting defines the URL or IP Address of the AVID Interplay Work Space such as http://12.13.14.15/services/Assets.

Interplay URI

URI of the Interplay Asset

This URI is in the interplay://bin/catalog/folder format.

Interplay Username

Interplay Username

This is the user name of the user whose credentials will be used on the AVID Interplay system.

Interplay Password

Interplay user’s password

This is the password of the user whose credentials will be used on the AVID Interplay system.

Interplay Shared Storage Folder

Interplay’s shared storage

This is usually a mapped network drive or folder on the local Transkoder workstation.

Interplay Shared Folder Prefix

Subfolder on the Interplay’s shared storage

This can be any name desired for the project sub-folder on the AVID Interplay shared storage and is used to group and identify result renders copied from Transkoder.

Interplay Metadata

Extra metadata for Interplay Clips, declined as key:value pairs. Example: USER:attr1=valueof attr1,USER:attr2=value of attr2

Avid Old Compatibility Mode

Use older Avid SDK to generate DNxHD with older codec IDs

B.4. Base Settings

The Base settings are the following:

Setting Description Note

Frame Rate

Project frame rate

Main playback frame rate setting also driving default encode FPS.

Timecode

Set drop or non-drop timecode

Drop-frame vs Non-Dropframe Timecode, global to the project.

Start Timecode

The start timecode of the Timeline

Default timecode where video events dragged to the timeline will be starting.

Image Width

Preferred output image width

Project resolution driving the Reframe node part of the default pipeline.

Image Height

Preferred output image height

Project resolution driving the Reframe node part of the default pipeline.

Stereo

Stereoscopic 3D project

Stereo / Mono project.

Project Base Path

Home path for the project

Several project folders will be created within: thumbnails, metadata files and this is where the default render target points to.

Additional Media Paths

Additional media paths

If the project media is loaded from under the project base path, it is registered with a relative path. Having multiple Additional media paths settings the user can flexibly copy media from one master location to the other, as long as the folder structure is preserved below the project home folder.

Render Path

Location to render to

This drives the meaning of the $RenderPath$ keyword in the path definition.

Project Data Names

Metadata fields to identify shots

Relevant for Dailies: basic metadata fields the operator can inspect and edit in the basic metadata inspector (E).

Project Match Indices

Indices of metadata fields that identify audio, first metadata is 0

Indices of the metadata items specified above that match audio metadata: Scene/Take or Scene/Slate/Take typically.

B.4.1. Advanced Base Settings

Setting Description Note

Imported Still Length

Default length of imported stills in seconds. 0 means single frame

Container

The name of the input media in the GUI

Affects the Load Page screen UI

Timeline Sorting

Timeline sorting options

Order of clips in dailies "card" timelines.

Project Data Upper Case

Entered Scene, Slate imformation forced to be upper case

Playback Type

Special playback methods like Double Speed

Playing back every or every single frame. May be useful for HFR shows.

Transfer Date Type

Date format of the transfer date (e.g. the date folder) Additional characters can be added to the end

Volumes to Ignore

Volumes to hide on the load page

Folders to Ignore

Folders to hide on the load page

File Names to Ignore

Comma-separated list of files to ignore at new footage parse. Wild card characters like * and ? are possible to use. For example, *.mov, *.txt, A001_C002_1234.chk, video???.mov

Burn-In Logo Dir

Folder containing logos for burn-ins

These logo files can be a accessed as a custom burn-in: $logo:mylogo.png$ where mylogo.png is the logo copied to this folder. Supported formats include PNG and TIFF.

Content Descriptors

Home for sidecar files, such as bitrate, PSNR, maxFALL, maxCLL, etc.

Project Media Structure

Describes the project structure for images files

Relevant for dailies workflows.

Project Audio Structure

Describes the project structure for audio files

Relevant for dailies workflows.

Project Look Structure

Describes the project structure for imported look files

Relevant for dailies workflows.

Project Organization

Metadata field for organization media, usually the date

Relevant for dailies workflows.

Metadata Path

Home of the project metadata. (SystemPojectFolder stands for C:/ProgramData/…​)

Project Additional Data Names

Metadata fields that are also copied to captured frames

Minimal Shot Length

Shots shorter than the value set here are deleted when executing auto-cut

Fix Node Connection

Forces Result1 to be connected if grade and viewing templates are disconnected

Pipeline for Imported Looks

Specifies which node template to use for stills and slate images. Enter the format name here.

This setting refers to one of the formats within the project; the imported stills in the "import" look store will be initialized with the default pipeline of this format. This is relevant in cases when the operator is unable to drag CDL grades from the still store to the timeline, due to incompatible pipelines.

MP4 Type

MP4 decoder type

Save Thumbnails with EMD

Save thumbnails along with Export Metadata file

Still Export File Name

Naming pattern of the exported EMD file

Default Timeline Mode

Default Timeline mode

By default vertical editing is turnes on. You can disable it using this project setting.

GOP Double Read Optimization

Optimize plaback for long gop media at 2x playback speed

YUV Decode Matrix

Matrix used for YUV to RGB conversion

Thumb Cache Path

Path to local thumbnail cache

B.5. Conform Settings

The Conform settings are the following:

Setting Description Note

EDL Match Timecode

EDL Timecode will be matched to this TC in the database

EDL Match Tape Name

This metadata will be matched to EDL TapeName. If LabRoll is chosen, the source TC is matched

EDL Match Custom Tape Name

Custom metadata field of the shots table to be used for EDL conform

EDL Handle

Shots start or end handle in frames when loading an EDL, -1 means complete shot

EDL Match Relative TC

The SrcTC will be used as a relative TC to the begining of the matching clip

Grade Settings to Load

List of tools to load from the dailies grades when applying with Shift+G

Conform Log Path Prefix

Path prefix for conform Logs

Black Missing Media

Show black instead of Missing Media

Media Search Path

Media path where to search during loading EDL

EXR Metadata Mode

Driving the EXR metadata parsing for faster media loading

B.6. Encode Settings

The Encode settings are the following:

Setting Description Note

Render Shot Boundaries

Which timeline track will define the shot boundary of the rendered files

Skip GAPS During Rendering

Whether to skip the GAPs during rendering

Rendered Media TC

The main timecode in the MXF/ALE

to encode files with a predefined start TC, most typically 00:00:00:00, the operator can set the Rendered Media TC to Manual. In this case all rendered clips should have the timecode of the value specified in the Custom Encode TC field just below.

Custom Encode TC

Custom TC used for encoding if Manual Rendered Media TC is used

Rendered Media Tape Name

The tape metadata in the generated MXF and ALE

ALE Media Metadata

Add all metadata from the file headers to the ALE

ALE CDL

Add ASC CDL values to the ALE file

ALE Audio Order

Order of the audio metadata in the ALE

This option has been introduced to handle channel interleaving broadcast metadata present in the audio WAV file header. When set to InterleaveIndex then the ALE generator exports the audio tracks accordingly. The default option ChannelIndex uses the straight channel order without considering the interleave metadata.

File Sequence Numbering

File numbering when rendering file sequences

File Sequence Custom Start

Custom start file numbering, may be a number, or a metadata $keyword$

Flash Burn-In

Flash burn-ins

Where the flash burn-in - displayed for a few frames only - is shown: Head / Tail / Head and Tail

Burn-In Padding

Padding of the ShotFrameCount and CustomFrameCount burn-in

Date Format

Date format for burn-ins

Possible options are the following: D/M/Y, M/D/Y, Y/M/D, M/D/YY

B.6.1. Copy Settings

Setting Description Note

Archive Audio

Archive the audio file next to the shot

Archive Audio Later

Archive the audio files after everything else archived

Archive Extra Files

Path to specific files to be archieved along with any Card, e.g. render proxies

Archive Extra Files for Date

Path to specific files to be archived when the user selects a Date on the load page and clicks on the Archive Extra Files

Ingest with Verify

Verify the source and destination footage with checksum after ingest is finished

Archive with Verify

Verify the source and destination footage with checksum after archive is finished

Archive with Separate Jobs

…​

Default Hash Method

Hash method to verify copied files

Available hash methods include md5, xxHash, xxHash64BE, and sha1. xxHash64BE (Big Endian) is similar to the xxHash algorithm, but it outputs a hash text in hexadecimal values, which makes it more compatible with other post production products. xxHash64BE is also compatible with Google’s xxHash implementation.

Ingest Path (1-4)

Paths where Copy Central is ingesting from

Archive Path (1-4)

Paths where Copy Central is archiving to

Copy with Direct IO

Copy encoders are using direct I/O or not

Copy MD5 Level

Level of MD5 file from the source

Copy MHL

Create and support MHL checksum

Continue on File Missing Error

Continue the COPY job even if some files are missing

Copy Report File

Copy report log file to destination after archiving. Disabling this for LTO drives will increase performance

LTO Path

List of volumes separated by semicolon that should use LTO optimizations

B.6.2. Advanced Encode Settings

Setting Description Note

ALE Video Format

Video format tag in ALE header

ALE Sound Channels

Sound channels number in ALE

ALE Aux TC

Aux timecode field in ALE

Additional Templates

Location of the custom encoder templates XML

Encoder File

Location of the encode parameter XML

Encoder Post Batch

Batch to be created upon finishing a render

Use Source DPX metadata

Use the original DPX header if rendering from DPX to DPX

MXF Film Gauge

The FilmGuage metadata in the rendered MXF files

MXF Audio

Enable or disable rendering Audio MXF files

AAC Audio Offset

Audio offset in samples for AAC encodingto be in sync with video

MXF Need Film Data

Enable or disable rendering Keycode metadata into AVID MXF files

Chapter Format

Chapter name formatm, e.g. $Scene$/$Take$

Show All Frames

Should Transkoder show all frames during encoding or not

Wait for Postprocess

Pause encoding until post-processing completes

Copy Source Verify

Verify the source footage with checksum after copy

Generate Chapter Track

If disabled, no chaptertrack in the generated QT files

Timecode for Web

Generate TC track for PIX

ProRes Render Path

Temp path for rendering ProRes files locally first, and only copying to final location upon finishing

Postprocess Temp Path

…​

Postprocess Temp Path Enabled

…​

Codex VFS Config

Configuration of the Codex VFS

H264 Post-Mux

H264 encoders mux after encode

FCP Audio

Render audio into the FCP deliverable or not

Avid Sound TC Fix

Avid Sound TC compensation

Forensic Watermarking Strength

Strength of Watermarking

Encode V210 Range

Defining the range when converting to V210 from RGB

AS02 Audio Frame Rate Fix

…​

B.6.3. NexGuard Settings

Setting Description Note

NexGuard Manager Enabled

Enable built-in client to the NexGuard Manager

NexGuard Manager Address

Address of the NexGuard Manager (IP address)

NexGuard Manager User

Username registered on the NexGuard Manager

NexGuard Manager Password

Password of the NexGuard Manager user

NexGuard Mgr Update Interval

Interval in seconds between progress updates sent to manager

NexGuard Manager Timeout

HTTP GET timeout in ms (0 to use curl defaults)

NexGuard Mgr Conn Timeout

Connection timeout in ms (0 to use curl defaults)

NexGuard Transfer Characteristics

Transfer Characteristics for watermarking

B.7. DCI Settings

The DCI settings are the following:

Setting Description Note

Package Asset Creation

By default Transkoder creates Hard Links in DCP package folder. User can force Transkoder to copy the media anyway

VF Asset Selection

Controls the logic driving the asset selecton for VF finalizing

Remote Hardlink Creation

Try to create hardlinks on remote systems

Load with Reel Markers

Load CPL timeline with untrimmed assets and with reel marker track

Hash Validation

Enable hash validation in DCP validation process

Default DCI Type

Define the type of the DCI deliverables

Default DCI Encryption

Encrypted DCI or not

KDM Duration

KDM duration in days

KDM Home

Project folder for incoming KDMs

KDM Delivery Home

Folder where KDMs are auto-generated

KDM Certificates

Folder to collect certificates to generate KDMs for

B.7.1. Advanced DCI Settings

Setting Description Note

Atmos Sync Channel

Overwrite 14th channel with Atmos sync data

Atmos Main Audio Channels

Defatul channel number for Main Audio in case of Atmos DCI packages

DCI Encryption Key Home

Project folder to collect encrypted DCI MXF keys

Write Signature

Controls the signing of the CPL and PKL of the DCP XMLs

Write HMAC

If true HMAC values are to be generated and written into MXF

XYZRGB Conversion

Conversion from DCI XZY to P3 RGB in case of DCP decoding and from P3 RGB to DCI XYZ incase of DCP encoding

Validate J2K Frames

Enable validating JPEG2000 frames or not

J2K Check Stop on First Error

If on, then JPEG2000 validation will stop on first error, avoiding huge error message duplicating

Import After Finalize

Import generated DCP package after finalize or not

Finalize Auto Render

Launch automatic render for missing assets upon Finalization

Search Missing VF Assets

Automatically search file system for missing assets

VF Search Path Levels

Levels to search for missing asset from the current CPL directory that is being imported

This search is recursive meaning it will search all sub folders as well.

Path for Validation Report

If empyt, reports will be generated into default folder. Otherwise to directory set by user

Embed Start TC Into DCIMXF

This will enable start TimeCode embedding into DCI MXF. Use this with caution!

When enabled the timecode information is embedded into the track files. Make sure to adjust the Rendered Media TC field value in the Encode Page to RecordTC. In this case, when loading a CPL with Import CPL With MXF Timecode field value set to On, the track files will be placed in the timeline exactly the same as the timeline they are generated from.

Import CPL with MXF Timecode

Import CPLs based on MXF StartTimecode and place them accordingly on the timeline.

The possible values are Off, and On. When on the track files are added to the timeline according to the embedded timecode.

Half Res Decoding

Decode only half resolution images

B.8. IMF Settings

The IMF settings are the following:

Setting Description Note

Package Asset Creation

By default Transkoder creates Hard Links in IMP package folder. User can force Transkoder to copy the media anyway

VF Asset Selection

Controls the logic driving the asset selecton for VF finalizing.

Remote Hardlink Creation

Try to create hardlinks on remote systems

Hash Validation

Enable hash validation in IMP validation process

Dolby XMLMXF Extraction

The mode of Dobly XMLMXF extraction: Normal - support for full shot dissolve anime, Deep - support for every anim like the middle-shot anime, Legacy - legacy mode

B.8.1. Advanced IMF Settings

Setting Description Note

MaxFALL override

If set to a positive number, MaxFALL in the CPL will be overridden and changed to this value

MaxCLL override

If set to a positive number, MaxCLL in the CPL will be overridden and changed to this value

Hash Type

Hash type for IMF MXF and CPL files

Write Application Version

Fill the Applicationidentification parameter in CPL ExtentionProperties

Write ST2086 Metadata for HLG

Write ST2086 HDR Mastering Display metadata into MXF in case of HLG color

CPL Subtitle Tag Name

The tag name for subtitle sequences in the CPL. The standard 2067-2 supports SubtitlesSequence, but the practice is MainSubtitleSequence

Direct Read

Parse the essence offset at load time and use it for the later read

Ignore Index

Ignore the MXF Index table

Half Res Decoding

Decode only half resolution

Finalize Auto Render

Launch automtice render for missing assets upon Finalization

Import After Finalize

Import generated IMP package after finalize or not

Audio File Postfix

The posfix of audio filenames in the multitrack audio deliverables(like IMF)

Search Missing VF Assets

Automatically search file system for missing assets

VF Search Path Levels

Levels to search for missing asset from the current CPL directory that is being imported

This search is recursive meaning it will search all sub folders as well.

VF Search Path

When non-empty overrides VF Search Path

Path for Validation Report

If empyt reports will be generated into default folder. Otherwise to directory set by user

Encoder Auto Correct

When changing IMF encoder settings, Transkoder auto corrects the profile. This can be switched off.

Warning: in Off state it is possible to create MXF files with non-valid metadata

Meta Cache Dir

Directory where the IMF readcache files are saved

B.9. GUI Settings

The GUI settings are the following:

Setting Description Note

Scale

Scale all GUI elements

Tables Scale

Scale all GUI tables

Style

GUI style

Menu Always Visible

Menu always visible

Analyzer 2nd Head On

Enable Second Head Analyzer

Clean GUI

Clean GUI style

Clean GUI Angel

Node drawing style

Panel Type

Defines which type of panel is used

Default JKL Speed

Default playback speed multiplyer if J or L is pressed for playback

Frame Entry Mode

How to define frames when clicking on the REC TC above the timeline window

Player Navigation Bar

…​

Trackball Sensitivity

Sensitivity of panel trackballs, default setting is 100

Encoder Sensitivity

Sensitivity of rotational encoders, default setting is 100

Jog Sensitivity

Non-linearity of the jog control on the panel, default setting is 100

Color Circle Horizontal

Color circle allignment (default 1)

Insert Embedded Audio

Insert embedded audio to the timeline

Enable Alpha Channel

Insert video alpha channel if possible

Default Dissolve Length

The default dissolve length in seconds

Default Dissolve White

The default dissolve white in percent

Ins Emb Audio with Diff fps

Insert embedded audio to the timeline which the fps dows not match

Auto-Retime

Automatically retime the newly inserted shot to the project framerate

B.9.1. Advanced GUI Settings

Setting Description Note

Pan with Scroll

Allow image and timeline pan with mouse scroll

Control Brightness

Brightness of controls (default 100)

Exit with Esc

Exit the application with Esc

This setting has previously controlled both the timeline-exiting and application exiting. Instead, timeline-exiting is assumed, and this setting only controls the exiting from the application. If turned off, the user will not be able to exit on the main load page by pressing Esc, only through the menu.

Color Circle Angle

Orientation of the color wheel

Color Circle Sensitivity

Sensitivity of the color wheel (default 150)

Color Circle Slider Sensitivity

Sensitivity of color wheel’s slider (default: 200)

Upper Timeline

Display timeline indicator on the top of the screen

Background Intensity

Background intensity (default 0.06)

Show Preset Dates

Show the save date and time in the Preset List on the tool HUD

Start/End Frame Indicator

Display on-screen start and endframe indicator

Swipe Gesture Enabled

Enables navigation between windows by tapping window edges with the mouse pointer

Panel Shot Jump

Defines where the playhead jumps during navigating from shot to shot on the Wave panel

Up/Down Arrows

How the navigation with arrows work

Still Store Names

Turns on/off the names (scene/take metadata) on the still store thumbnails

Show Stereo Tools

Turns on stereo tools in monoscopic projects, so over/under or side-by-side material can be processed

Mouse Drag

Mouse scrub directions

Show Audio Sync

Display the status of the audio sync on the shot

Show Circle Takes

Display the status of the circle take on the shot

Show Graded

Display the status of the graded state on the shot

Show Uploaded

Display the status of the uploaded state on the shot

Show Bin Metadata

Display metadata colors on the bin thumbnails

Thumnail Overlay Transparency

Transparency of the flags displayed on the thumbnails

Tags on Thumbnails

Type of indicators on the thumbnails to show tags

Window Opacity

Opacity of windows

Timeline Trim Hovering

Enable/disable timeline trim hovering

Quick Metadata Position

Defines where the quick metadata is displayed

Thumbnail with Burn-In

Image burn-ins on thumbnails

Metadata Viewer Cell Width

Width of cells in the small metadata viewer (default 280)

Lock Window Positions

Prevent windows from moving (default 0)

Render Table

Rendertable database server

Always Show Template Wizard

If this is enabled, upon loading a new camera format the template wizard window will be shown

Alternate Marker Drawing

Enable alternate Positional Marker drawing

Bin Cut Load Type

Whether to load the timeline events or just the source media

Image Linear Scale

Using linear scaling when zooming into the image on the screen. Nearest neighbor is used when Off

Player Saved Drag Mode

…​

B.10. Video Settings

On the Video settings page, the SDI video output of the video I/O cards can be adjusted by the following settings:

Video Out

Video output resolutions and modes

Video Device

Device to generate video output signal

Progressive Format

Progressive (P) or Progressive Segmented Frame (PSF) mode

HDMI Output Enabled

Use HDMI output - supported on AJA IoXT, Io 4K, Io 4K Plus, Kona3G, and Kona4 devices only

Dual Video Out Mode

Second video out resolution - supported on AJA devices only

This is useful, for example, when the SDR and HDR version of the same content needs to be inspected on the video output.

B.10.1. Advanced Video Output Settings

Clean Output

Show clean video result or video result with GUI on SDI output.

Full to Legal on Second Screen

Apply full to legal conversion to video output on second screen, otherwise it will be full to full range

This feature works only on Mac.
12-Bit Color Depth on SDI

Use RGB 12-bit color depth on SDI output

Video IO Speed

Sets video I/O speed to 1.5 GB or 3 GB

3G Level B Video Signal

Enabling this setting forces Level B signal instead of Level A. The default option means 3 GB Level B for RGB output, while Level A for high-frame-rate YCbCr output.

RGB Over 3G Level A

Special option to set RGB over 3G Level A flag on AJA devices

AJA HDR Signaling

Set SDR, HDR10, or HLG standard for HDMI signal on AJA devices

AJA RGB to YCbCr Matrix

Matrix coefficient selector for RGB to YCbCr conversion on AJA devices

Single-Link UHD

Single link 6G/12G output on Kona 5 when video resolution is UHD or 4K

HDMI HDR10 Metadata Mode

Use custom HDMI light level values (defined by the following two parameters), or use precalculated values from the timeline

Custom HDMI HDR10 MaxCLL

Custom Maximum Content Light Level value for AJA HDR10 HDMI metadata

Custom HDMI HDR10 MaxFALL

Custom Maximum Frame Average Light Level value for AJA HDR10 HDMI metadata

QHD/4K on Two Wires

Two/Four-channel QHD modes on AJA devices

Two Wires SDI Output

Select SDI 1/2 or SDI 1/3 for output

10-Bit Color Depth on HDMI

Use 10-bit color depth on HDMI output, otherwise 8-bit color depth will be used

HDMI Output Mode

HDMI output mode and range when Dolby Vision HDMI Tunneling is enabled, where Auto means RGB Full, otherwise it is YCbCr Legal

Use 8 Channels on HDMI Audio

Use 8 channels for HDMI audio, otherwise 2 channels will be used

Video with Audio

Enable audio in SDI video output

Reference Signal

Set the reference (synchronization) signal source for AJA devices

Auto

Use external reference signal if detected, otherwise freerun

External

Use external reference signal

Freerun

Use on-board signal generator of AJA device

Input

…​

Halved HFR

Reduces the frame rate in HFR modes by half by dropping every other frame

YCbCr Output Range Mapping

Maps full or legal range node pipeline result to legal range YCbCr video output

RS422 Control

Enable RS422 control on AJA devices

To connect the controller device with Transkoder, a special cable is required. Please contact support for assistance.
YUV mode

Send YCbCr data directly to AJA device. In this case, the RGB to YCbCr conversion is done by the CPU and not the AJA device

Multichannel Stereo

Legacy option to set color space and depth of stereo SDI output

Off

SDI output signal is limited to 8-bit YCbCr only

On

SDI color space and color depth is not limited

AJA Alternative Audio Init

Alternative initialization of audio engine on AJA devices

AJA Board Sort Mode

Use first or last device when having more than one AJA devices

Forced 2Kx1080 Mode

Set AJA 2Kx1080 flag

3G Level A to Level B Conversion

Special option to set Level A to B conversion on AJA devices

AJA VPID

AJA 32-bit VPID options for SDI output

Clear

Send zero as VPID

Auto

VPID generated automatically based on corresponding video settings

Bypass

Bypass option keeps the current VPID setting

Forced AJA Video Standard

Override video standard on AJA devices. Use 1080P to force 1080 progressive, and 1080 to force 1080 PsF or interlace output, otherwise the default Progressive Format setting will be applied.

Embed Audio in All Channels

Embed audio on all SDI wires

Generate LTC

Generate SDI-embedded LTC (Longitudinal Timecode)

LTC Channel

Select SDI audio channel for LTC (1-16)

LTC Volume Multiplier

Set LTC Volume (0-255)

LTC Offset

Set generated LTC Offset in frames

SDI Embedded TC Source

Choose to embed either the record timecode (RecTC, the timecode of the timeline), or the source timecode (SrcTC) of the footage for the video output. The default setting is record timecode. Camera Timecode Source: Select camera timecode source

Calibration

Sets SDR, HDR, or HDRC 3D LUT as calibration for the video output if needed

Resize

Apply automatic image resizing to match the defined video out resolution

Dolby Vision HDMI Tunneling

Enable Dolby Vision HDMI tunneling

B.10.2. Dual Video Out Mode - Secondary Video Output Settings

Progressive Format

Progressive (P) or Progressive Segmented Frame (PSF) mode

HDMI Output Enabled

Use HDMI output - supported on AJA IoXT, Io 4K, Io 4K Plus, Kona3G, and Kona4 devices only

12-Bit Color Depth on SDI

Use RGB 12-bit color depth on SDI output

Video IO Speed

Sets video I/O speed to 1.5 GB or 3 GB

3G Level B Video Signal

Enabling this setting forces Level B signal instead of Level A. The default option means 3 GB Level B for RGB output, while Level A for high-frame-rate YCbCr output.

RGB Over 3G Level A

Special option to set RGB over 3G Level A flag on AJA devices

AJA HDR Signaling

Set SDR, HDR10, or HLG standard for HDMI signal on AJA devices

AJA RGB to YCbCr Matrix

Matrix coefficient selector for RGB to YCbCr conversion on AJA devices

Single-Link UHD

Single link 6G/12G output on Kona 5 when video resolution is UHD or 4K

YCbCr Output Range Mapping

Maps full or legal range node pipeline result to legal range YCbCr video output

HDMI HDR10 Metadata Mode

Use custom HDMI light level values (defined by the following two parameters), or use precalculated values from the timeline

Custom HDMI HDR10 MaxCLL

Custom Maximum Content Light Level value for AJA HDR10 HDMI metadata

Custom HDMI HDR10 MaxFALL

Custom Maximum Frame Average Light Level value for AJA HDR10 HDMI metadata

QHD/4K on Two Wires

Two/Four-channel QHD modes on AJA devices

Two Wires SDI Output

Select SDI 5/6 or SDI 5/7 for output

10-Bit Color Depth on HDMI

Use 10-bit color depth on HDMI output, otherwise 8-bit color depth will be used

HDMI Output Mode

HDMI output mode and range when Dolby Vision HDMI Tunneling is enabled, where Auto means RGB Full, otherwise it is YCbCr Legal

Use 8 Channels on HDMI Audio

Use 8 channels for HDMI audio, otherwise 2 channels will be used

Forced 2Kx1080 Mode

Special option to set AJA 2Kx1080 flag on AJA devices. Auto means disabled if the resolution is HD or UHD (QHD), and if resolution is 2k or 4k, it is enabled. Bypass option keeps the current system setting.

3G Level A to Level B Conversion

Special option to set Level A to B conversion on AJA devices

AJA VPID

AJA 32-bit VPID options for SDI output

Clear

Send zero as VPID

Auto

VPID generated automatically based on corresponding video settings

Bypass

Bypass option keeps the current VPID setting

Forced AJA Video Standard

Override video standard on AJA devices. Use 1080P to force 1080 progressive, and 1080 to force 1080 PsF or interlace output, otherwise the default Progressive Format setting will be applied.

Embed Audio in All Channels

Embed audio on all SDI wires

YUV mode

Send YCbCr data directly to AJA device. In this case, the RGB to YCbCr conversion is done by the CPU and not the AJA device.

B.10.3. Cloud IO Stream Settings

Video Encoder

Choose HEVC or H.264 video encoding

Enable Color Depth 10 Bit

Use 10-bit color depth on video stream [HEVC only]

Enable Audio

Enable audio stream

Address

Destination address

Port

Destination port

Bitrate

Output stream bitrate

Latency Correction

Number of frames to compensate clien buffers

B.11. Audio Settings

The Audio settings are the following:

Setting Description Note

Audio Force Sample Rate

You can force sample rate for all used audio files. It is useful when the sample rate in the header does not match the record rate. The value -1 means use the rate from the header.

Audio Playback Frame Delay

The delay of audio playback in frames. This setting does not affect the rendering.

Soundroll Metadata

Where the soundroll metadata should come from. FolderName refers to the containing subfolder of the audio file

Move Metadata at Sync

Apply metadata from broadcast wave audio file to shots upon syncing

This setting is useful for displaying sound timecode on smart slates, or for syncing when video and audio timecodes match.

Move Select Tag at Sync

Apply circle take flag from audio BWF to shots upon syncing

Preserve Leading Zeros

Preserve the leading zeroes in scene and take metadata coming from the BWF header

Auto-Detect Slate

Split the scene metadata info scene + slate if a spearator character is present

Remove Space from Metadata

Remove spaces from the BWF metadata, such as Scene and Take

B.11.1. Advanced Audio Settings

Setting Description Note

Adjust Playhead Upon Stop

Step back a few frames when stopping with spacebar if the audio waveform is on (during sync). Disable if you’ve got good reflexes

Audio Play Pull

You can pull up and down the audio playback. It does not affect the rendered audio. The values -1, 0, 1 mean pull down, normal, pull up.

Audio TC 60 to 30

Convert the audio TC to 30 fps from 60 fps

Audio TC Convert

Forces audio timecode conversion on burn-ins and in the ALE

Audio Timecode Offset

Frame offset for the compensation of delay between audio and video timecode to help further timecode-based syncing throughout the project

Collapsed Audio Channel

Audio track to be displayed when track collapsed

Detect LTC Timecode

Auto-detects the running LTC embedded in the audio samples

Detect LTC Sensitivity

LTC detection algorithm sensitivity

Mixdown from Project

Using the Mixdowns of the specific project. The default is the actual "self" project

Playback Mixdowns Auto-Update

Auto update the playback mixdowns based on the selected audio track

Playback Volume

…​

Auto-Pull

…​

Mixdown Channels

Number of channels displayed in audio mixdown window (default 24)

Group Multifile Audio

Attemp to join multi-file single channel audio into single mult-channel audio track (Default is On)

Mult-file = Multi-mono

Do Not Use Audio fps Sync

Do not use the Audio fps for TC sync

B.12. Start-Up Settings

The Start-Up settings are the following:

Setting Description Note

App Support Path

…​

CP200 Panel IP Address Prefix

The IP prefix (first 3 digits) of the network of the CP200 panel. Default setting is 10.0.0.

CP200 Panel IDs

Comma-separated list of connected CP200 panel IDs, e.g.: 1,1327

Console Enabled

Enable the console window

Database Server

Network address of the database server. The default value is localhost.

Home

Install home

Master Project Path

Home folder of the Projects without database

Permissions Check

Enable if you want to run folder permissions check at start

Project Name

Project name

Submit to Local Database

Submit to local database server

Project Renamed to

…​

B.13. Config Settings

The Config settings are the following:

Setting Description Note

Resolution

…​

Thread Num Override

This will override the loading threadnum, if not -1

Use Display GPU for J2K Decoding

Use the display GPU for J2K decoding

NAS Mode

Use NAS optimized settings

GUI Image Delay Threshold

Maximum milliseconds to wait for the image to become available for the GUI

B.14. Grade Settings

The Grade settings are the following:

Setting Description Note

Master Nodes

Nodes to be dragged and dropped when in Master role

Color Grading Nodes

Creative color grading nodes, copy-pasted in Grade role

Stereography Nodes

Nodes the Stereographer role adjusts

CDL Tool

If multiple grading tools are used, force the system to import/export CDL from/to this tool. e.g. RGBlin1

CDL Reference Image Type

THe type of the CDL image: original, ungraded, or graded or the graded with viewing LUT

Auto-Apply Looks

Enables the application of imported looks upone loading a new timeline

Auto-Apply EMD

Enables the application of grade imformation from EMD files

Apply Metadata at Auto Apply

Enables the application of metadata of imported looks upon loading a new timeline

B.14.1. Advanced Grade Settings

Setting Description Note

RRT Type

RRT Version to use in RRTODT node

ACEScc Mode

Formula used for ACEScc curve

Code Value per Printer Light

Code value (in 0-1024 range) per printer light

Dolby Vision Offset Weight

Weight parameter for Dolby Vision Offset value (Lift)

Animation Keyframes

Controls how animation keyframes are created. Group indicates that grading/transform/stereoscopic nodes are animated together

LUT Dir

Location of the folder containing LUTs

Export Lut Type

…​

EMD Name in Look Store

The text on the thumbnail of the imported EMD files from Live

Preset File Location

Location of the saved looks

Look Reference Metadata Name

Metadata name for automatic LUT application (Ctrl+Shift+I)

iPad Path

Result iPad path

iPad Tool Name

This tool stores the iPad grade setup

LUTs Legal2Legal

3D Luts will be saved as legal-to-legal instead of full-to-full range

auto Save Thumbnails to DB

Upon saving, the thumbnails for the clips are saved in the database server

CDL Printer Lights Enable

Enable printer lights controls in hud and color circle

Lift Gain Gamma Mode Enabled

Use CDL nodes in lift/gain/gamma mode

Convert V210to Legal

Convert V210 image sources to legal range instead of full range

Save High Res Still

When OFF it forces the exported stills to be downScaled to fit into 2048x1080

This applies when exporting stills through the File menu.

Still File Extension

Type of the saved still image file: jpeg, png, tiff, or dpx

The default setting is png, and when saving a still, the format selected in this setting will be used.

Still File Name

Filename pattern of the saved still ($DateTime$ and $AppName$ usable only here)

Still File Path

Filepath pattern of the saved still

Use 2094-10 Metadata

Whether to use metadat coming from the files or use the SDRRemap tool parameters

B.15. Metadata Settings

The Metadata settings are the following:

Setting Description Note

EMD Transfer

Controls what is imported when importing EMDs from the data/emd folder

EDL Tape Name Length

Truncation of the tape name in the exported CDL-EDL

File Seq TC from Frame Number

Enable for calculating the main Timecode from the framenumbers

Camera Letter in Camroll

The index of the character in the CamRoll string indicating the camera identifier (A/B etc). 1 means first character

Small Metadata View

Metadata to be displayed in the small Metadata Window

Metadata Overlay

Metadata to be displayed in the metadata overlay

B.15.1. Advanced Metadata Settings

Setting Description Note

Audio Channel Names on Timeline

Channel names separated by comma

Movie Metadata Cache

Enable movie metadata cache for faster loading

Ingest Hours After Midnight

For some extra hours after midnight media is copied into the last days folder

Parse Pattern

Define a pattern to extract metadata from file path and name. Example: /$CamRoll_$##/capture/$Scene$_$Take$.mov

Use it if you want to copy data from full path, including file name to metadata, such as scene, slate, take, etc. It can only be applied to new labrolls, cards, tapes and magazines (i.e. when status of the labroll/tape/card is new in the project load page).

QTake Link

Describes how to link QTake entries to shots. For example: shots.Take=qtake.Take and shots.Scene=qtake.Scene

Lens Data Link

Describes how to link lens data entries to shots

For example: lensdata.timecode>=shots.TCin and lensdata.timecode<=shots.TCout and lensdata.roll=shots.LabRoll

ARRI DCS Link

Describes how to link ARRI DCS entries from .cus files to shots

For example: arridcs.timecode>=shots.TCin and arridcs.timecode<=shots.TCout and arridcs.transferdate=shots.TransferDate

EDL Link

Describes how to link input EDL entries to shots. Example: edls.TapeName=shots.LabRoll and edls.SrcTCIn=shots.TCin

Flex Link

Describes how to link Flex events to shots

For example: flex.LabRoll=shots.LabRoll and flex.VideoInpointTC=shots.TCin

Metadata Matching LUT name

List of metadata in order of priority to match LUT names when auto-apply LUTs (Shift+G)

Metadata Matching LUT node

The name of the LUT node to adjust when auto-applying LUTs (Shift+G)

DPX Timecode HFS

Whether the DPX TC is HFS or not

Bin General Search

Bin search allows general search in all metadata

Shots Table Header Items

Metadata to be displayed in the shots table window (Alt+T)

Clip specific metadata from the Shift+U Window may be also used with the "MainMedia:MediaMetadata:" prefix. Example: MainMedia:MediaMetadata:CameraType

CSV Header Items

Metadata to be exported in the look CSV

QC Table Header Items

Metadata to be displayed in the QC table window on a dailies / bin timeline (Ctrl+Q)

QC Table Cut Header Items

Metadata to be displayed in the QC table window on a CPL / Cut timeline (Ctrl+Q)

Read Clip Metadata

Enable clip metadata parse

Calc TC When Missing MP4

Timecode will be calculated from file creation if the timecode is missing from MP4

AS02 Audio Frame Time

Duration of 1 frame of AS02 audio

B.16. QC Settings

The QC settings are the following:

Setting Description Note

QC Operator

QC operator name to be displayed in the QC report

QC Comment Timecode

Type of timecode associated with the QC notes

QC Comments

List of QC comments for quick-buttons

HDR Color Space

Controls the color space of the Analyzer tool input

Maximum Brightness for HDR

Valid values are between 100.0 and 10000.0

Additional Marker for HDR

Valid values are between 100.0 and 10000.0, -1 to turn off

Minimum Brightness for CIExy

Minimum brightness of pixels to show up on HDR Analyzer

Gamut Tolerance

Percentage of the most saturated pixels to ignore

Brightness Tolerance

Percentage of the brightest pixels to ignore

Brightness Frames Tolerance

Percentage of the brightest frames to ignore

Light Level Threshold

Threshold for cut detection

Solid Frame Tolerance

Black Frame Pixel Value

PDF Report

Organization of the PDF report by container (magazine/card/labroll) or Transkoder assigned VTRolls

PDF Report Sorting

Sorting of clips in the PDF report

PDF Report Takes

Takes to appear in the PDF report

PDF Report Columns

Columns in the PDF report

Control Knob Scrubbing Audio

Scrubbing audio when jogging with the control knob

Camera Base Gray

Brightness of gray card in nits

Camera White Stops over Gray

Camera stops for white clipping warning over base gray

Skin Tone x

Skin tone CIE x coordinate

Skin Tone y

Skin tone CIE y coordinate

Audio Level Warning

Warning level for audio, should be between -30 and -8

Auto QC Tolerance

Tolerance value for audio QC

B.17. Subtitle Settings

The Subtitle settings are the following:

Import Start Timecode

Defines the imported subtitle media container start point.

IMSC Timecode Base

This setting defines the IMSC subtitle timecode base, for In/Out time calculations. The default setting is Media.

SMPTE

use real floating point fps timecode base

Media

use rounded fps timecode base

If the imported titles are not in sync with the video content, try to switch this option to SMPTE then reimport the title media to the project and reinsert it to the cut.
Regenerate Subtitles

Defines the method of generating the required subtitle files when creating a package.

Off

Transkoder uses the original XML file for packaging without any modification.

On

Transkoder reads the subtitles and creates a new subtitle file when packaging.

TimeExpression

This setting makes no changes to the structure of the subtitle XML, only updates the timing of the subtitles. This is very important in case of using IMSC1 subtitles for IMF.

Regenerate PNG Names for IMSC

If diff then Transkoder will use original PNG file names when generating IMSC subtitle files

Subtitle Font Source

Here, you can define the font to use for rendering titles.

Incoming

Use a separate font file defined in the title file itself. This option is useful if you want to check if the incoming compressed font file includes all the characters needed for the cut.

System Font

Use the default embedded font of Transkoder, which is an open font and includes every character of all languages.

Custom Font

Provide the path of a desired font in the Custom Font File input field.

Custom Font File

Here, you can give the path of the font file to use. This works only if the Subtitle Font Source is set to Custom Font.

Subtitle Path

Home path for the subtitle metadata and PNGs.

Upper Timeline

Show subtitle positions on upper timeline.

Language

Defines the language to load.

Subtitle Selection

Defines where the subtitles selection (title/language) is saved.

Preferred Image Height

For IMF IMSC format, this parameter is used to count the appropriate font size if the title file does not contain any information about the size of the titles.

B.17.1. Advanced Subtitle Settings

Subtitle Color Conversion

The target color space for titles.

Allow Retime on Import

Allow retime when importing a new subtitle if the source frame rate is different from the target frame rate

HDR Subtitle Color Space

Sets the color space of HDR titles.

HDR Subtitle Brightness

Sets the brightness of HDR titles.

Bidirectional Text Level

Specifies the default direction for the rendered text both on the T1 and the CC1 track.

Auto

Choosing this option, Transkoder uses the text direction given in the subtitle file itself. If it does not contain any information on the text direction then the software decides by the language of the subtitle file.

LTR

Left to right

NLTR

Left to right with the punctuation on the left

RTL

Right to left

NRTL

Right to left with the punctuation on the right

Off

No change. The characters and the punctuation will be displayed exactly in the same order as written in the subtitle file.

DCI Subtitle MXF Encryption

Controls if the DCI SMPTE subtitle MXFs are encrypted or not. Default setting is Auto.

Interop DC Subtitle Version

Interoperable DCI Subtitle XML version

SMPTE DCI Subtitle Version

Sets the SMPTE DCI Subtitle XML namespace version

DCI Safe Caption Export

Exports DCI Closed Caption files according to ISDFC best practice document. This means ignoring LoadFont attributes.

Regenerate DCI Closed Captions

Defines the method of generating the required closed caption files when creating a package.

Off

Transkoder uses the original XML file for packaging without any modification.

On

Transkoder reads the subtitles and creates a new subtitle file when creating a package.

SCC Use Absolute TC Values

If turned Off, Transkoder counts control bytes into start timecode, therefore the titles will be displayed slightly later than defined in the source file.

Avoid Overlapping Subtitles

Enable this option to avoid overlapping subtitles and adjust timecodes accordingly when exporting subtitles.

B.18. User Settings

The User settings are the following:

Setting Description Note

Log Level Console

Logging severity level in console

Log Level File

Logging severity level in log file

Bin Metadata Labels

Metadata to be displayed in the bin window (List mode)

Bin Audio Metadata Labels

Metadata to be displayed in the bin window (List mode)

Bin CPL Metadata Labels

Metadata to be displayed in the bin window (List mode)

Role

Role base permission

Use Clone Version

Prefer the clone (prox) version when loading a timeline. It works only after the clone versin is generated from Transkoder

Video Logo Opacity

Opacity of the Logo

Video Logo Opacity2

Opacity of the Logo for dual mode, secondary output

Show Console

Show/Hide Console window

MRU Languages Num

Size of the list of most recently used languages for media tracks

Client

For future use

Dailies Bin Sorting

Enables Dailies bin sorting mode (affects Scene/Take order)

Bin Icon Labels

Show loaded timeline name in player

Show Loaded Name

Show loaded timeline name in player

Show Alteration Times

Show alteration times of projects and timelines on load-page

Show Timelines Ordered

Show timelines ordered naturally or ascending

Timelines and projects may be sorted alphabetically or by time last modified in ascending or descending order.

Default Transkoder Project Path

…​

Clone Active

…​

B.19. Cloud Upload Settings

The Cloud Upload settings are the following:

B.19.1. Remote Server Cloud Upload Settings

Setting Description Note

Auth Password

Password for the user

Auth User

User of the remote server

Compression

Compression of the http communication to gain speed

Database Name

Name of the project on the remote server

Database Server

IP address of the remote database server

File Server

IP address of the remote server

File Server Alias

Friendly name of the remote server

File Server Folder

Folder on the remote server

File Server Password

Password for the user

File Server User

User of the remote server

HTTPS

Security of the http communication

Targe Host

IP address of the node processing uploads

Upload Type

Select upload type

B.19.2. COPRA Cloud Upload Settings

Setting Description Note

Colorist

Dailies Colorist

COPRA Log File

Switch to enable creating a log file while generating and uploading

COPRA Project

COPRA project name if different than Colorfront project name

COPRA Push Notification

Send push notifications if server supports it

COPRA Release

Clip autorelease on COPRA server

DoP

Director of photography

Editing

Film Editor

Encode Job to Task Mapping

Mapping of EncodeJobID (first column in the deliverables screen) to COPRA4 Task Name. 1:MisterX,2:MisterY. ATTENTION: These JobIDs get shuffled around as you add/remove new lines to each result’s encoding table

FTP Editor Login

Editor Login for Auto Upload to COPRA Server

FTP Editor Password

Editor Password for Auto Upload to COPRA Server

FTP Server

COPRA server to upload to, e.g. COPRA DE, COPRA US, custom (can be set below)

FTP Server Custom

IP or hostname of custom COPRA server

FTP Upload

Upload via FTP to server after rendering finished

Location

Locations where the Film was shot

Material

Name the different formats the film is shot with, like RED Epic, ARRI Alexa, Film 35mm, Phantom