[argyllcms] Re: dispwin vs xcalib loading LUT

  • From: Gerhard Fuernkranz <nospam456@xxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Tue, 02 Jan 2007 01:22:14 +0100

Stefan Döhla wrote:
>>> Therefore I took the system gamma into account, which is 2.2 on Win32
>>> and 2.222 on Unixes
>> Hi Stefan, why 2.222 on Unix?
>> According to which standard/specification?
> I assumed this when reading Charles Poynton's GammaFAQ and some other
> UNIX resources some time ago: As gamma usually 1/0.45

IMO this magic number "0.45" is rather meant to be  0.454545... (i.e.
1.0 / 2.2), but written with only two decimal digits precision.

(But anyway, the difference is not very significant)

> and 0.45 are
> assumed (1/0.45 == 2.22222). Maybe that's the reason for this "system
> gamma" as I named it, that one must take into account and then the
> 2.20 of Win32 should be 2.222 as well.
> We'd have to doublecheck with the other profile loaders again - but
> I'm sure they give slightly different results and we can never find
> out whether it's 2.20 or 2.22 on Win32 or Linux ...

I'm wondering, which representative vcgt loaders for Unix do actually
exist (besides xcalib and dispwin), which could be used for comparison?

Is there actually a difference between vcgt loaders on Mac and Windows
(i.e. is the gamma value of the VideoCardGammaFormula treated relative
to a "system gamma" of 2.2 on Windows, but relative to 1.8 on Mac)?

Btw, do you (or somebody else) happen to know how the the "Gamma"
parameter in the "Monitor" section of xorg.conf is related to this
stuff? Does this parameter just set the initial graphic card LUTS (i.e.
the state immediately after the start of the X server), which can be
later overwritten by xcalib or dispwin, or is this parameter applied _in
addition_ to the LUTS which are loaded with xcalib or dispwin?


Other related posts: