search

OpenICC / xcalib

Load 'vcgt'-tag of ICC profiles to X-server and MS-Windows. Works on calibration stage, which can be a precondition for display ICC color conversions.
Other
67 stars 12 forks source link

xcalib -invert -alter doesn't work, but increases gamma #9

Open juergendrachta opened 5 years ago

juergendrachta commented 5 years ago

On Arch-Linux with AMD APU (amd-xorg drivers), xcalib v. 0.10, the inversion via "xcalib -i -a" but only increases the gamma heavily.

This bug was also reported by others: https://bugzilla.redhat.com/show_bug.cgi?id=1662620 https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=913849 https://github.com/zoltanp/xrandr-invert-colors/issues/20

Driver-Info:

lspci -k | grep -EA3 'VGA|3D|Display'
05:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Picasso (rev c1)
    Subsystem: Lenovo Picasso
    Kernel driver in use: amdgpu
    Kernel modules: amdgpu
beku commented 4 years ago

Thanks for reporting. This issue appears to be a special case. Unfortunately I have no means to test on such a card. Can you verify the ramps are correctly written?

juergendrachta commented 4 years ago

Thanks for the comment! Hmm... do you want to see the full output of running xcalib with the "-p" flag (it gives quite a bunch of numbers i cannot say if it's worth copying it :-) )

But I found some interesting thing, when i ran xcalib with "-v":

(base) [...@... ~]$ xcalib -i -a -v
XRandR output:          eDP
Red Brightness: 0.000000   Contrast: 99.903870  Max: 0.999039  Min: 0.000000
Green Brightness: 0.000000   Contrast: 99.903870  Max: 0.999039  Min: 0.000000
Blue Brightness: 0.000000   Contrast: 99.903870  Max: 0.999039  Min: 0.000000
X-LUT size:         1024
(base) [...@... ~]$ xcalib -i -a -v
XRandR output:          eDP
Red Brightness: 99.903870   Contrast: -103925.398438  Max: 0.000000  Min: 0.999039
Green Brightness: 99.903870   Contrast: -103925.398438  Max: 0.000000  Min: 0.999039
Blue Brightness: 99.903870   Contrast: -103925.398438  Max: 0.000000  Min: 0.999039
X-LUT size:         1024

It seems that the number of the contrast aren't ok (at least I think so by comparing them to the xcalib output of a machine, where the color inversion works properly)!