https://bugs.kde.org/show_bug.cgi?id=527151

            Bug ID: 527151
           Summary: Night Light tints white content teal/green instead of
                    warm yellow with "Prefer color accuracy" (6.8.0
                    regression, EDID built-in profile)
    Classification: Plasma
           Product: kwin
      Version First 6.8.0
       Reported In:
          Platform: CachyOS
                OS: Linux
            Status: REPORTED
          Severity: normal
          Priority: NOR
         Component: night color
          Assignee: [email protected]
          Reporter: [email protected]
  Target Milestone: ---

DESCRIPTION
After upgrading  kwin 6.7.5 -> 6.8.0 (pacman log, 2026-10-15: kwin 6.7.5-3.1 ->
6.8.0-1; no other color-stack package changed), Night Light (Mode=Constant ,
3000K) turns white backgrounds a teal/greenish color instead of warm yellow.

STEPS TO REPRODUCE
1. Wayland session, SDR, display color profile "Built-in", Color accuracy
"Prefer color accuracy", no ICC profile, HDR off
2. Enable Night Light at 3000K
3. Open anything white (e.g. a blank page)

OBSERVED RESULT
Whites shift teal/green rather than warm.

EXPECTED RESULT
Warm yellowish shift, as in 6.7.5.

SOFTWARE/OS VERSIONS
Operating System: CachyOS Linux
KDE Plasma Version: 6.8.0
KDE Frameworks Version: 6.31.0
Qt Version: 6.12.0

ADDITIONAL INFORMATION
Switching Display Configuration -> Color accuracy to "Prefer performance" on
all outputs restores the correct yellow tint: the legacy per-channel multiplier
path is fine, the color-managed path computes a wrong white-point adaptation.
Setting the color profile to "None"  instead of "Built-in" changes nothing.

GPU: NVIDIA RTX 3080 Ti, driver 615.78.08, OpenGL ES compositing.

Three outputs, all with "Color profile source: EDID" and no ICC profile
(kscreen-doctor):
* HDMI-A-1 BenQ 2560x1440: R 0.660/0.331 G 0.300/0.632 B 0.148/0.055 (slightly
wide gamut)
* HDMI-A-2 Samsung 1920x1080: ~ sRGB primaries
* DP-2 LG 2560x1440 10bpc: R 0.686/0.309 G 0.265/0.668 B 0.150/0.058 (wide
gamut)

Possibly related: bug 527129 (6.8.0 color pipeline regression direct scanout
path) and MR !10042 (6.8 color pipeline shader changes).

-- 
You are receiving this mail because:
You are watching all bug changes.

Reply via email to