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

> I just implemented showing ΔE2000 for each mapped color, as well as an icon for the colors with the 1st, 2nd, 3rd smaller ΔE.
Surprisingly, it looks like for all of these Scale LH is still the winner, indicating that while the CSS gamut mapping algorithm matches our expectations better, it is actually farther from how the color is supposed to look.

 ΔE2000, because it is well established and also because it does not depend on Oklab, is a good choice here for a sanity check. But gamut mapping will always produce a significant  ΔE, if it didn't then the color was already almost in gamut anyway. The main component of that will be a reduction in chroma, which is why the individual deltas in Hue, Chroma and Lightness are crucial in that visualization you did; it makes it easy to do hue sweeps at various lightness and chroma levels, see visually the result, and objectively verify the closeness.

-- 
GitHub Notification of comment by svgeesus
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/9449#issuecomment-1944224909 using your GitHub account


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

Received on Wednesday, 14 February 2024 16:52:54 UTC