[argyllcms] Perceptual intent gives too bright blacks
- From: Gerhard Fuernkranz <nospam456@xxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Tue, 12 Jan 2010 22:29:27 +0100
Hello Graeme,
somehow I'm not perfectly happy with the blacks produced by the
perceptual intent of recent 1.1 RCx versions. For my taste the resulting
blacks are too bright.
The device links below are created with the command
collink -v -qh -G -ip -cmt -dmt sRGB.icm zeus8.icc linkXXX.icc
for versions 1.0.3, 1.1 RC1, RC2 and RC4.
(zeus8.icc is a display profile for my notebook LCD)
Results:
$ icclu -s255 link103.icc
0 0 0
0.000000 0.000000 0.000000 [RGB] -> Lut -> 0.000000 6.684445 0.000000 [RGB]
$ icclu -s255 linkRC1.icc
0 0 0
0.000000 0.000000 0.000000 [RGB] -> Lut -> 5.801711 8.055397 5.294810 [RGB]
$ icclu -s255 linkRC2.icc
0 0 0
0.000000 0.000000 0.000000 [RGB] -> Lut -> 16.764869 19.218541 18.009442
[RGB]
$ icclu -s255 linkRC4.icc
0 0 0
0.000000 0.000000 0.000000 [RGB] -> Lut -> 18.664281 20.751646 18.966948
[RGB]
Corresponding CIE numbers:
$ icclu -s255 -ir -ff zeus8.icc
0 0 0
0.000000 0.000000 0.000000 [RGB] -> Lut -> 1.166733 0.368478 -1.854839
[Lab] ==> device BP
0.000000 6.684445 0.000000
0.000000 6.684445 0.000000 [RGB] -> Lut -> 1.463407 -0.566909 -1.087113
[Lab] ==> 1.0.3 black (still close to device BP)
18.664281 20.751646 18.966948
18.664281 20.751646 18.966948 [RGB] -> Lut -> 4.189088 -0.483794
-0.728603 [Lab] ==> 1.1 RC4 black (noticeably brighter than device BP)
Regards,
Gerhard
Other related posts:
