W3C home > Mailing lists > Public > public-css-archive@w3.org > January 2017

Re: [csswg-drafts] [compositing] SDR and HDR compositing

From: Pierre-Anthony Lemieux via GitHub <sysbot+gh@w3.org>
Date: Thu, 26 Jan 2017 20:24:18 +0000
To: public-css-archive@w3.org
Message-ID: <issue_comment.created-275502971-1485462257-sysbot+gh@w3.org>
> the problem with that is that HDR luminance is an absolute scale,
> while SDR is a relative one (relative to a white diffuser).

sRGB defines a reference display with a luminance level of 80 cd/m^2 
(see Section 4.1 at IEC 61966-2-1), so I would not characterize sRGB 
as relative to a white diffuser.

Both sRGB and Rec 2100/PQ (HDR) content are effectively mastered 
against an absolute scale. They are also both experienced, by the 
end-users, on a wide range of displays with different luminance 

> If we map SDR to HDR, 

@frivoal It would be good to list the use cases where this might 
happen in order to best answer the question of what peak luminance 
level should be used when compositing SDR and HDR content a common 
plane. Did you have specific ones in mind? In the case of 
subtitles/captions, it is necessary to allow the author to specify 
that peak luminance level to ensure the right experience. In the 
long-run, subtitle/captions should ideally be specified using HDR 
pixels natively (instead of being specified in sRGB and composited 
using a gain value).

GitHub Notification of comment by palemieux
Please view or discuss this issue at 
using your GitHub account
Received on Thursday, 26 January 2017 20:24:24 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 5 July 2022 06:41:07 UTC