[Openicc] VMs and display correcotions

[Chris Murphy]

On Mar 9, 2011, at 4:58 AM, Stefan Döhla wrote:

> Am 09.03.2011 13:33, schrieb Kai-Uwe Behrmann:
>> Am 09.03.11, 11:11 -0000 schrieb Richard Hughes:
>>> On 9 March 2011 11:45, Kai-Uwe Behrmann <ku.b at gmx.de> wrote:
>>>> To what extend shall a host generate EDIDs? display size, extensions,
>>>> colorimetry ...?
>>> 
>>> The host already generates a virtual EDID for most VMs. At the moment,
>>> I'm pretty sure all the fields are hardcoded, but it might sense to
>>> mirror some values from the active output that the VM is being
>>> displayed on.
>> 
>> Ah. In case the host EDID's colorimetry gets passed to the host EDID's colorimetry, than it should as well be ensured that the VM window is itself not colour managed by the host or set the guest EDID back to sRGB.
> 
> If LUT changes happened, the display's EDID is no more valid. Hence, a new one would have to be generated by the VM - or the same profile be simply used within the VM (knowing that it's currently not possible to do LUT changes there again) ...

EDID color space info are chromaticity coordinates. So the primaries themselves are fixed no matter what you do to the LUTs. However, the LUTs do change luminance, which affects the XYZs for the primaries used to build the display profile. So yes, small issue if the LUT alters the white point. If it's just altering TRC, then I don't see it as a problem so long as that TRC is defined by either gamma 2.2 or the sRGB function.


Chris Murphy