12:19 pq: EDID base block has gamma value. CTA-861-H says that for the default RGB colorimetry (i.e. nothing else indicated) use BT.709 Item 1.2, which is essentially the home of the footnote saying displays use BT.1886 which says gamma=2.4. Who to believe and when?
12:21 pq: or is this again the secret contract to achieve system gamma 1.1 if base block says 2.2?
12:28 pq: FWIW, linuxhw/EDID.git has 96k samples with gamma=2.2, and less than 2.5k samples with other values.
12:30 emersion: FWIW, edid-decode has a warning when gamma ≠ 2.2
12:30 emersion: when "Default (sRGB) color space is primary color space" is printed
12:30 emersion: it checks gamma = 2.2 ("sRGB is signaled, but the gamma != 2.2.")
12:31 emersion: and also chromaticities match sRGB ("sRGB is signaled, but the chromaticities do not match.")
12:32 pq: oh, naturally yes, because sRGB is approximately gamma 2.2. Saying otherwise would be inconsistent indeed.
12:33 pq: but default RGB colorimetry can also be not sRGB AFAIU
12:33 emersion: i see
12:34 pq: then gamma does not need to be 2.2 either, I guess
12:36 pq: there is an explicit bit in base EDID saying that the default RGB colorimetry is sRGB, and if the bit is set, then primaries and gamma *must* be according to sRGB.
12:37 pq: ..."If display is sRGB compliant"
12:38 pq: whatever sRGB compliant means
12:39 emersion: "does not blow up the whole building when sRGB data is sent"
12:40 pq: I'm fairly sure not blowing up depend on timings and not colorimetry.
12:40 pq: While reading all this, I've come to the conclusion that the base block primaries might not be be physical primaries.
12:41 pq: base block primaries apply only when using the default RGB colorimetry *with an IT video mode*
12:45 pq: I wonder if one could find out the practical primaries of a consumer HDR display by looking at primary gradients; when the gradient goes flat as saturation increases, the cut-off point is the primary.
12:54 drakulix[m]: Hmm, the state of the next wayland meeting doodle is unfortunate. d_ed and bl4ckb0ne have no overlap. Are any of you fine with skipping this one or should we delay the topics till next week?
12:59 d_ed[m]: next week is a big kde conference
12:59 d_ed[m]: as in, from the 15th for a week
12:59 d_ed[m]: but you can meet without me, Vlad is on the same page as me for most things
13:00 bl4ckb0ne: im fine reading the notes
13:02 drakulix[m]: Thanks you two. I'll roll a dice then. 😂
13:02 pq: "Monochrome displays shall indicate the appropriate white point x, y coordinates regardless of the actual color" - curious, wonder what that means. White point is arbitrary? Or matches the monochrome color?
14:08 wlb: weston Merge request !1304 opened by Simon Ser (emersion) clients/simple-dmabuf-egl: do not block until rendering completes https://gitlab.freedesktop.org/wayland/weston/-/merge_requests/1304
22:53 Ishq: Hello, Is it possible for you to know what screen you are hovering over? Something like xwininfo for wayland? It seems this might have to be compisitor specific?
22:54 Ishq: (Sorry, by screen, I mean application window)