[argyllcms] Re: Mystery monitor profile installed itself
- From: Elle Stone <l.elle.stone@xxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Fri, 4 May 2012 08:41:15 -0400
On 5/4/12, Richard Hughes <hughsient@xxxxxxxxx> wrote:
> Yup, this has been done by gnome-color-manager for about 2 releases
> now (1 year+), and colord-kde for a few months.
> The so called "auto-edid" profile doesn't have any VCGT data, and so
> if it's being used by default then then VCGT is cleared to a unity
> gamma ramp as if there was no profile set at all.
Hmm, the odd thing is, the weird vcgt is being loaded even when the
auto-edid profile is being used.
> Are you loading a
> VCGT before gnome starts somehow?
No, I'm not *intentionally* loading a VCGT before gnome or kde or
icewm start up. On my main computer I'm running OpenSuse 12.1 pretty
much straight kde, but with the icewm and a handful of gnome/gtk apps.
I've searched diligently but have failed to find what configuration
file or program might be loading a vcgt. The only part of colord that
is installed is libcolord. ixicc and xcalib are not installed. All
Oyranos-related programs have been uninstalled and I've tried to
locate and destroy all configuration files that Oyranos might have
created.
On a whim I checked my laptop with "dispwin -s lut.cal". My laptop is
an old Aspire with a completely different video card and screen
display than my main computer. The laptop is running Debian Sid,
minimal kde with icewm. I've never profiled it. libcolord and colord
aren't installed. Neither is any Oyranos stuff or xicc or xcalib. And
the laptop has the exact same lut being loaded at startup as my main
computer.
I pulled the mystery vcgt RGB_I and RGB_R columns into a spreadsheet
to see if I could see a pattern. I multiplied all the values by 255,
which made the highest I value equal to 255 and the highest R,G,B
values equal to 254.008 (rounding to three places). The difference
between the two columns increases steadily from 0 for the first entry,
to just under 1 for the last entry. It is essentially linear, but
looks like there is some kind of accumulating rounding error:
I*255 R*255 delta
0.000 0.000 0.000
1.000 0.996 0.004
2.000 1.992 0.008
3.000 2.988 0.012
4.000 3.984 0.016
5.000 4.981 0.019
6.000 5.977 0.023
7.000 6.973 0.027
. . .
253.001 252.017 0.984
254.000 253.011 0.989
255.000 254.008 0.992
What program might be generating, and what configuration file is set
to load this odd vcgt on two different computers?
Elle, feeling mystified
--
Elle Stone
http://ninedegreesbelow.com
Other related posts:
