Re: [Lcms-user] small question about transicc conversion
An ICC-based CMM for color management
Brought to you by:
mm2
From: Michal S. <so...@zi...> - 2010-10-21 19:05:31
|
On 10-10-21 12:46, mar...@li... wrote: > Hi Michal, > > Please check the GIT code, as the bugs you report may > be already solved. The unicode works fine to me now and > the V4 sRGB profile works also fine, given the limited > precision it has. > > Regards > Marti > Checked the git - unicode display is indeed fine, sorry for the noise then. As for conversion using the same profiles (-t1 -v3 -i prof -o prof): - with no-bpc srgb v2 I get: R? 0 G? 0 B? 0 R=0.5409 G=0.5409 B=0.5409 [PCS] Lab=(10.9189,0.0044,-0.0040) XYZ=(1.2051,1.2497,1.0311) still somewhat high (and rounding to 1), compared to other values we get using this profile - with srgb v4 I get pretty much the same values as in the official version; from somewhat troubling results: Enter values, 'q' to quit R? 3 G? 3 B? 3 R=3.4337 G=2.6753 B=2.9570 [PCS] Lab=(11.5511,0.0234,0.0078) XYZ=(1.2926,1.3398,1.1047) Enter values, 'q' to quit R? 4 G? 4 B? 4 R=3.4337 G=3.8682 B=4.1738 [PCS] Lab=(11.7586,0.0078,-0.0039) XYZ=(1.3215,1.3703,1.1306) Notice, that R here for both 3 and 4 yields the same value after the roundtrip. Well, just for your information. regards Michal |