The objective of this article is to provide a quick reference guide with recommended Best Practices for creating Dolby Vision content. It is highly recommended that users undergo the necessary training before embarking on a Dolby Vision project.

 

Learn to create in Dolby Vision

In partnership with The Dolby Institute there are a range of options for learning about how to create spectacular content in Dolby Vision. Many of these options are available for free on a self-guided basis, or you can sign up for instructor-led training where you can be guided through in-depth training curriculums and have questions answered by the expert instructors as you progress. 



The Dolby Vision content creation process involves the following steps:

1. Create the HDR Master
2. Create Dolby Vision Analysis (L1) Metadata
3. Check content mapping and apply creative Trims as required
4. Create Deliverables (output from color corrector)
5. Quality Control

Creating the HDR master

Creating the HDR master involves using images of the highest quality, coloring on an approved Dolby Vision enabled color corrector and monitoring on a professional, reference grade HDR monitor. Here are some of the most important considerations to bear in mind during this process:                    
 

Monitoring

It is highly recommended to use professional, reference grade displays during the Dolby Vision content creation process. 
For recommendations on monitoring for Dolby Vision content creation, refer to the HDR Display Considerations and Color Grading & Mastering suite setup articles.
 

Source Images

Use the highest quality images/video as the source and starting point for the Dolby Vision HDR Master. It is important to ensure that the import process into the color corrector does not limit or compress the dynamic range and color gamut of the source media. Also, remove or modify any LUTs or transforms designed for SDR workflows that may be limiting dynamic range and color.
For more information on recommended file formats and best practices around image import, follow the Dolby Vision Recommended Source Formats.
 
 

Color Grading

Use a Dolby Vision certified color correction or mastering system for Dolby Vision metadata creation. There are many acceptable color pipelines and workflows that can be used, but in all cases the output must be configured for ST.2084 (PQ), and either P3D65 or Rec.2020, for monitoring, final render and delivery.
The entire range of tools in the color corrector can be utilized to create the best HDR images, according to the creative brief and requirement. Be careful not to clip highlights in the HDR master as these cannot be recovered in the mapped/derived Rec.709 SDR version. The Dolby Vision content mapping algorithms are designed to maintain the look and intent of the HDR master.

You can refer to the following articles and tutorial video for specific instructions on how to setup your project and create Dolby Vision content using Blackmagic Resolve and Filmlight Baselight:

How to Setup a Dolby Vision Project (tutorial video)
Quick Start Guide Dolby Vision: DaVinci Resolve Studio
Quick Start Guide Dolby Vision: Baselight


A list of approved partner products supporting Dolby Vision can be found here:
Dolby Vision compatible Color Grading & Packaging Systems
 

Setting Up the Output Color Space

There are many acceptable color pipelines and workflows that may be used when creating content in Dolby Vision, but the output color space must be set to either P3D65 or Rec.2020 color gamut for monitoring and delivery. Consult the studio’s delivery specifications for information on which color gamut to use (and any special instructions for Rec.2020 delivery – see below).
 
Guidance on Output Color Space: Some studios request P3 deliveries while others may request for the delivery of their content in Rec.2020. Dolby recommends using Reference Grade Mastering Displays that can accurately reproduce at least 99% of the P3 color gamut. However, since most monitors today cannot display 100% of Rec.2020, it is not recommended to create content in unrestricted Rec2020. 
There are a few recommended options for delivering a Rec.2020 Dolby Vision master, when requested:

Option 1: Work in P3

• Set the output of the color pipeline (color management) in the color corrector to P3D65.
• Setup the HDR Monitor to monitor in P3D65.
• Set the Mastering Display to P3D65 in the Dolby Vision settings.
• Before rendering out the final master, change the output of the color pipeline (color management) in the color corrector to Rec.2020, limited to P3D65. This will apply a color conversion to the project/timeline from P3D65 to Rec.2020, by preserving the P3 color primaries inside a Rec.2020 container.
• You must also set the color encoding primaries inside the exported Dolby Vision XML to Rec.2020 on export or use Metafier to change the color encoding on an already exported XML.

Option 2: Work in Rec2020 (limited to P3)

• Setup the color pipeline (color management) in the color corrector to output Rec.2020 limited to P3. This will constrain the output color space to P3 and prevent the colorist from accidentally using colors outside P3.
• Setup the HDR display to monitor in Rec.2020.
• After grading and before running the Dolby Vision analysis, remember to set the Mastering Display to P3D65 in the Dolby Vision settings since the output color space is still P3.
• After grading, the final master can be rendered out in Rec.2020. The rendered output will have pixels that are constrained to P3 inside a Rec.2020 container.
• The color encoding primaries inside the exported Dolby Vision XML must be Rec.2020 to match the rendered image/video files. In the current versions of Resolve and Baselight you can choose the color encoding primaries while exporting the XML. You can also use Metafier to set the correct color encoding primaries.

For more information on this topic, refer to the following articles:

How to Setup a Dolby Vision Project (tutorial video)
How to Deliver Dolby Vision in Rec.2020 (tutorial video)
Quick Start Guide Dolby Vision: DaVinci Resolve Studio
Quick Start Guide Dolby Vision: Baselight
 

Once the HDR master image has been created and finalized, the colorist can begin the Dolby Vision process.
Different color correction or mastering systems may vary subtly with regard to how these steps are performed and there may be differences in the user interface from one system to another. It is recommended to refer to the color corrector’s documentation for product specific settings and instructions.
You can refer to the following Dolby Knowledge Base articles for specific instructions on how to setup your Dolby Vision project and create Dolby Vision content using Blackmagic Resolve and Filmlight Baselight:

Quick Start Guide Dolby Vision: DaVinci Resolve Studio
Quick Start Guide Dolby Vision: Baselight
How to Setup a Dolby Vision Project (tutorial video)

There are a few steps that MUST be completed before the analysis and trims can be performed. These are typically found in the project settings on the color correction or mastering system that is being used.
They are:
 

Setting the Dolby Vision Algorithm Version

Users must set the Dolby Vision Content Mapping (CM) version before running the analysis. CM algorithm version 2.9 (CMv2.9) has been in use since the introduction of Dolby Vision. CM algorithm version 4.0 (CMv4.0) which was introduced in the fall of 2018, is now widely used by content creators around the world and is accepted by all studios and clients.
CMv4.0 incorporates many improvements to the algorithms and mapping, provides better control and enables faster turnaround of projects. CMv4.0 automatically generates backward compatible CMv2.9 metadata.
It is important to select the Dolby Vision version beforehand as this cannot be changed midway through the project. 

Here are two articles that provide more detail for the different Dolby Vision versions/metadata:

Dolby Vision Versions - CMv2.9 vs. CMv4.0
Dolby Vision Metadata Levels
 

Selecting a Dolby Vision Mastering Display (Level0)

Next, you must set the Mastering Display in the Dolby Vision settings in the color correction or mastering system. The Mastering Display selected must match the intended output color volume.
 

Selecting the Analysis Tuning

Users can now select from six options for the Analysis Tuning that will affect the strength of highlight retention and mapping. In tools that have implemented the new Analysis Tuning, a new selector will appear labeled “Analysis Tuning” with six options:

1. Legacy CM4 L1
2. Most Highlight Detail / Most Mapping
3. More Highlight Detail / More Mapping
4. Balanced
5. Less Highlight Detail / Less Mapping
6. Least Highlight Detail / Least Mapping

For more info on Analysis Tuning see the Dolby Vision Analysis Tuning in CMv.4.0 - FAQs

Note:
Dolby recommends that users leave the Analysis Tuning set to Balanced as a starting point for new projects. 

Setting the Active Image Area (Level5)

• The aspect ratio should be set to constrain the analysis area to the active image area and avoid analyzing the black levels of the Letterbox or Pillarbox region.
• If the project contains multiple aspect ratios, set the global aspect ratio to the most widely used ratio for the project. Some color correctors may allow you to set a separate, per-shot aspect ratio for shots that differ from the global aspect ratio. In such cases the aspect ratio can be set separately for shots that are different from the global aspect ratio.
• The aspect ratio of the project is described as L5 metadata in the exported XML. L5 is described as two values namely Canvas Aspect Ratio (full image raster) and Image Aspect Ratio (active image area).

Creating Dolby Vision Analysis (L1) Metadata

 

Important points to note when creating Dolby Vision Metadata

Dissolves and Transitions

Text and Graphics

Using iCMU or eCMU

Setting the iCMU Target (Display)

Creating Trim Metadata (Level8)

1. You can use the mid tone offset (L3) control to modify the effect of the primary tone curve and adjust the black/shadow areas of the image. Once a good starting point is obtained, the trims can used to refine and finesse the image.
2. You can use Gamma to elevate the blacks/shadows as required and use the Gain trim control to restore any highlights that may have been elevated or clipped.
3. You can also choose to go back and modify your original HDR content to lift the blacks slightly and then re-analyze and trim the image as required.
4. You can also copy and paste analysis and trims from another similar shot to address such issues.

Setting Additional Target Trims

• For a 4000-nit Dolby Vision HDR Master, the colorist may add optional target trims at 1000 nits and/or 600 nits.  
• For a 1000-nit Dolby Vision HDR Master, the colorist may add an optional target trim at 600 nits to check the mapping at the peak luminance level of many leading consumer HDR televisions.

Delivery Specifications

Quality Control

1. Check the Dolby Vision Metadata
2. Check the file metadata
3. Run a full visual QC to check the mapping using the image/video essence combined with the accompanying metadata.

QC - Metadata

Check the file metadata

A great tool to check technical and tag data of video and audio files is MediaInfo. This useful tool is available via the Apple App Store or directly from the MediaInfo website.
The specifications of the rendered master must conform to those requested by the client or studio that commissioned the project. Ensure that all technical and tag data match your delivery requirements. 

Visual QC of Dolby Vision Masters