[argyllcms] Re: Optimizing Printer Profiles
- From: Florian Höch <lists+argyllcms@xxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Mon, 10 Jul 2017 19:03:37 +0200
Am 10.07.2017 um 17:52 schrieb Ben Goren:
colprof -v -D"Printer A" -qm -S sRGB.icm -cmt -dpp PrinterA however, most
modern inkjet printers have a gamut range that is larger than sRGB and many
of us use AdobeRGB1998 instead.
You'd use Graeme's example if your working space is sRGB.
As long as actual image gamuts are close to whichever source colorspace
is used, that's fine. If the actual image gamuts are smaller (or larger)
than the given source colorspace, you may end up compressing more than
needed (in case image gamut is smaller than source space) or throwing
image information away, and in those cases giving the actual image gamut
(using -s source.gam) can improve the situation.
Choice of working space is not easy. Bruce Lindbloom's BetaRGB remains the
best, but it's showing its age, especially for the newest classes of
printers. Now would be a great time for somebody to do some solid research
into a GammaRGB...but that's almost certainly an even worse name....
A while ago, I derived a colorspace based on the Rec. 2020 red and blue
primaries, but an altered green primary (and D50 reference white) as an
alternative to ProPhoto (I consider the latter to be somewhat
inefficient, because it encodes quite a number of non-real/imaginary
colors, so it's a less suitable choice as source colorspace for colprof
or collink). It encompasses almost completely PhotoGamutRGB_avg6c, the
largest color space derived from actual print measurements that I've
come across, but has the benefit that it's a well-behaved matrix-based
space. I've dubbed my attemp "PhotoPrintRGB". In case someone is
interested, I can make the profile available standalone (for
unrestricted use however one sees fit).
Florian.
Other related posts:
