Observations re: enumerated values to describe the display color volume

Good morning/evening,

I had taken an action item to propose enumerated values to describe
the capabilities of the display color volume [1] -- this would be in
addition to a fully parameterized description.

Below are my observations after spending quality time looking at the issue:

- based on the reflector thread, I am not convinced that we can find
consensus on a limited set of display luminances values (having a
large number of enumerated values defeat the purpose of having
enumerated values in the first place, i.e., fewer errors and simpler
applications)

- I am not convinced that the display capabilities are complete
without specifying the viewing environment (color volume mapping for a
400 nit display is going to be different depending on the ambient
light level and temperature)

- the existing dynamic-range [2] and color-gamut [3] media queries
seem to already provide a gross parameterization of the display
capabilities

I suggest we consider the following questions:

- can parameterizing the viewing environment be avoided? If so, how?
- is providing a long list of enumerated values useful since the
target applications probably
- for an initial version of the API, are the dynamic-range and
color-gamut media queries sufficient?
- should we consider alternatives to exposing detailed display
capabilities, e.g. the "output image description" used by Wayland [4]?

Best,

-- Pierre

[1] https://github.com/w3c/ColorWeb-CG/pull/106
[2] https://drafts.csswg.org/mediaqueries-5/#dynamic-range
[3] https://drafts.csswg.org/mediaqueries-5/#descdef-media-color-gamut
[4] https://gitlab.freedesktop.org/pq/color-and-hdr/-/blob/bb44957a29c49232c3f94530e92617f6e8fc9d36/doc/color-management-model.md

Received on Wednesday, 4 October 2023 04:34:23 UTC