Dolby Vision is a set of technologies developed by Dolby Laboratories for high dynamic range (HDR) video.^[1][2][3] It covers content creation, distribution, and playback.^[1][4][5][6] It includes dynamic metadata that define the aspect ratio and adjust the picture based on a display's capabilities on a per-shot or even per-frame basis, optimizing the presentation.

This article may be too technical for most readers to understand. (September 2023)

Quick Facts Inception, Website ...

Dolby Vision

Inception	2014
Website	https://www.dolby.com/technologies/dolby-vision/

Close

Dolby Vision was introduced in 2014,^[1][7] making it the first available HDR format. HDR10+ is a competitor HDR format that also uses dynamic metadata.^[8]

Dolby Vision IQ is an update designed to optimize Dolby Vision content according to the ambient light.^[9]

Dolby Cinema uses Dolby Vision too, though because of the use of 2.6 gamma and thus 48 nits in SDR cinemas, the 108 nits used in Dolby Cinema is already HDR.^[10]

Dolby Vision allows for a maximum resolution of 8K, up to 12-bit color depth, maximum peak brightness of 10,000 nits.^[11] However, according to the Dolby Vision white paper, as of 2018 professional reference monitors, such as the Dolby Vision HDR reference monitor, are currently limited to 4,000 nits of peak brightness.^[12]

Dolby Vision includes the PQ transfer function, a wide-gamut color space (ITU-R Rec. BT.2020 in YC_BC_R or IPTPQc2), up to 8K resolution, and for some profiles (FEL) up to 12-bit. It can encode mastering display colorimetry information using static metadata (SMPTE ST 2086) and also provide dynamic metadata (SMPTE ST 2094-10, Dolby format) for each scene or frame.^[13]

This dynamic metadata allows adjusting of the image on a scene-by-scene or even frame-by-frame basis.^[14] These adjustments, called "trims" in Dolby's support documents, include parameters such as Lift, Gamma, Gain, Saturation, Chroma Weight, and so on. Each set of trims are specific to their respective target displays defined by nits and colorspace. Image aspect ratios can also be defined on a per-shot basis.^[15]

The Dolby Vision format is capable of representing videos with a peak brightness up to 10,000 cd/m² and a color gamut up to Rec. 2020.^[16] Current displays cannot reproduce the full Dolby Vision brightness and gamut capability. There are no brightness and color gamut capability requirements for consumer displays. When the consumer display has lower color volume than the mastering display, the content is adjusted to the consumer display capability based on the dynamic metadata.

Dolby Vision mastering displays require:^[17]

• EOTF: PQ
• Peak brightness: at least 1,000 cd/m²
• Black level: at most 0.005 cd/m²
• Contrast ratio: at least 200,000:1
• Color gamut: at least 99% of P3

Dolby Vision is a proprietary solution by Dolby.^[26]

In 2021, compatible color grading systems can create Dolby Vision automatic metadata with no additional cost for content creators.^[26] A $2,500 annual license is required to activate the trims allowing content creators to manually adjust the video.^[26] OEM and manufacturer of a grading, mastering, editorial, or other professional application or device need to apply for a license.^[26]

Dolby SVP of Business Giles Baker has stated that the royalty cost for Dolby Vision is less than $3 per TV.^[27][28][29]

Libplacebo (as part of VideoLAN project) implemented support for Dolby Vision profile 5's IPTPQc2 color space, MMR and reshaping, yet no support of dynamic metadata and NLQ yet.^[30]

MPC Video Renderer implemented partial support for Dolby Vision (Profiles 5, 8.1, 8.4 are supported).^[31]