[argyllcms] Re: fwa compensation unexpected results

Alastair M. Robinson wrote:

Maybe the gamut mapping strategy for the white point (and vicinity)
should be optimized differently (than the rest of the profile), even
for absolute colorimetric.

We had a conversation here a while back about Absolute Colorimetric gamut clipping at the *dark* end of the scale - and it was suggested then that the results would be improved if L were weighted a little more heavily in the clipping. It sounds like it's a non-trivial modification, unfortunately. 

There's already a "tweak" to take care of this, if you want to
do it manually, and if you are using device links (which is what
we always did for serious proofing). Generally, it has to be manual,
because the instruments aren't accurate enough - we're talking fractions
of a delta E. It's to use the -w [J,a,b] option to icclink, to fiddle
the white point match.

Graeme Gill.

Other related posts: