[argyllcms] Re: Colprof feature request: amount of black point hue correction control

Graeme Gill wrote:
> If you specified a 300% ink limit in targen, then it gets propagated
> to coloprof which by default applies the usual 10% reduction rule of
> thumb to ensure that the profile is within the measured gamut,
> resulting in a 290% limit.

Thanks Graeme for the explanation. I should have know that, but since
this is RGB and not CMYK this idea did not come into my mind. Yes, the
.ti3 actually contains TOTAL_INK_LIMIT 300. Myabe I had specified a 300%
limit in targen - whysoever. I don't remember, it's an old, archived
.ti3 file. I still cannot explain the effect below, though.

Best Regards,
Gerhard


=> remove TOTAL_INK_LIMIT 300 from the .ti3 file, then
$ colprof -v -B xxx

$collink -v -qh -ip -cmt -dpp sRGB.icm xxx.icc link.icc
$ icclu -ff link.icc
0 0 0
0.000000 0.000000 0.000000 [RGB] -> Lut -> 0.176970 0.130748 0.132303 [RGB]
$ icclu -ff -ia xxx.icc
0.176970 0.130748 0.132303
0.176970 0.130748 0.132303 [RGB] -> Lut -> 14.366123 2.867852 -1.582322
[Lab]
=> black comes out approx. 5 L* units too bright

$collink -v -qh -l300 -ip -cmt -dpp sRGB.icm xxx.icc link.icc
$ icclu -ff link.icc
0 0 0
0.000000 0.000000 0.000000 [RGB] -> Lut -> 0.001296 0.001800 0.002326 [RGB]
=> now with -l300, the numbers are OK, pretty close to RGB=0,0,0.
$ icclu -ff -ia xxx.icc
0.001296 0.001800 0.002326
0.001296 0.001800 0.002326 [RGB] -> Lut -> 9.092308 0.136239 -1.520532 [Lab]

Other related posts: