[Openicc] Minimal X Color Management

[Kai-Uwe Behrmann]

edmund ronald <edmundronald at gmail.com> schrieb:
>On Tue, Feb 21, 2012 at 10:46 PM, Chris Murphy
><lists at colorremedies.com>wrote:
>
>>
>>
>> On Feb 20, 2012, at 12:00 PM, Chris Lilley wrote:
>>
>> > My monitor (An Asus PA-246Q wide gamut 24" display) offers native,
>sRGB,
>> > AdobeRGB and some other modes. The EDID seems to be the same
>regardless
>> > of which mode is chosen.
>>
>> On Feb 21, 2012, at 6:02 AM, Kai-Uwe Behrmann wrote:
>> > That's a bug. And writing the manufacturer to fix that would be IMO
>> > appropriate.
>>
>> Are you sure it's considered a bug, and not a flawed "works as
>intended"
>> design?
>>
>> EDID is a string of data. I'm unconvinced, off hand, that this string
>is
>> dynamic. For resolution information, it doesn't present the currently
>set
>> resolution but rather all supported resolutions.
>>
>> I know of no case where display firmware is updated by the
>manufacturer,
>> so it would seem most likely once behavior is released to the public:
>bug
>> or not, we maybe stuck with a lot of devices with said behavior.
>>
>>
>You'd have to tell the system to re-read the EDID anyway, except if
>there
>is some special driver installed, via standard hardware there is no
>signal
>that I know off to indicate it has changed.

My desk monitor, a HP one, does inform the OS about colour space emulation changes. It triggers a EDID rescan. On osX that works flawless and the correct profile is loaded or generated. The OS can detect the emulation change by parsing the colorimetric part in EDID. That part is as well used to generate ICC profiles on the fly, when no other profile is avalable. libXcm can do the parsing under Linux/osX.


kind regards
Kai-Uwe