Re: Specifying an API to query the display color volume

On Tue, 12 Sep 2023 03:21:36 +0000
Lars Borg <borg@adobe.com> wrote:

> How about this:
> Start with standardized values, from maybe 40 to 10,000.
> (Such as 48, 80, 100, 160, 203, 235, 300, 500, 600, 720, 1000)
> If some standard values are close together, fake example 200 and 203, drop one.
> Fill in large gaps (> 40%? increment) with intermediate values
> 
> These values would be max luminance for small patches.
> Some displays can return this value over HDMI, but the value must be
> mapped to the limited set to prevent fingerprinting.

Hi,

are you perhaps referring to EDID CTA-861-H: HDR Static Metadata Data Block?

I'd just like to point out that this block in question defines
"desired" luminance levels. I don't think it is any kind of promise of
emitted luminance levels.

My HP Pavilion 27" quantum dot monitor was advertised with 400 cd/m²
luminance, but the EDID says desired max luminance is 600 cd/m², for
example, and with desired max frame-average luminance of 350 cd/m².

In other words, it does not look like these numbers would be usable to
correlate with actual viewing environments, assuming you had
information about the actual viewing environment.

Then again, that's just one example that could as well be explained by
EDID being incorrect, as EDID generally is notorious for.

> 
> No ambient info, as:
> 
>   *   this would require a measurement device.
>   *   Standardized mastering display color volume does not include it

SMPTE ST 2086 mastering display information does not include it, but
BT.2100 does give guidelines about it in "Reference viewing environment
for critical viewing of HDR programme material". I guess assuming that
is the best guess.

The PQ system uses a display-referred signal, so one could say the
signal needs to be displayed as encoded literally. But displays have
different capabilities, not to mention different viewing environments.
The PQ system can use ST 2086 metadata to describe the display it was
mastered for, and BT.2100 recommends a viewing environment, so these
should define the intended appearance which could be mapped to the
display at hand in the actual viewing environment.

However, I'm confused by claims that PQ monitors would display the
input signal nit-for-nit up to capability and lack end user
adjustments. That would mean any adaptation to the actual viewing
environment would need to happen at the video signal source. My monitor
indeed disables brightness and contrast adjustments in PQ HDR mode.


Thanks,
pq