Re: [csswg-drafts] [css-color-4] Channel clipping breaks author expectations, especially when using 'perceptually uniform' spaces (#9449)

> Suppose we create the spaces `oklch-srgb`, `oklch-p3`, and `oklch-rec2020`. They are _exactly_ the `oklch` space, except that they are always gamut mapped to the gamut indicated in their name.

To make sure I understand what is being proposed here, I want to flesh this out a bit.

Starting with the CSS keyword `red`, we have `oklch(63% 0.26 29)`, which is in the srgb gamut. This color could also be represented by `oklch-srgb(63% 0.26 29)`, `oklch-p3(63% 0.26 29)` and `oklch-rec2020(63% 0.26 29)`. In other words the ranges are identical between spaces, and we aren't attempting to adjust the ranges to fit into the specified gamut, correct?

Then, a user increases the lightness by 5% to 68%, giving `oklch(68% 0.26 29)`. This is outside of srgb but inside p3 and rec2020. If a user specifies `oklch-srgb(68% 0.26 29)`, a gamut mapped color would be displayed. If a user specifies `oklch-p3(68% 0.26 29)`, the color would not be gamut mapped.

If I'm understanding this correctly, we would still need to determine how colors that are out of the specified gamut should be mapped to that gamut. Could we move the idea of `oklch-srgb`, etc,  to a separate issue, to help keep this issue focused?

-- 
GitHub Notification of comment by jamesnw
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/9449#issuecomment-1998326713 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Thursday, 14 March 2024 20:02:12 UTC