[argyllcms] Re: Green tint on transparency in macOS
- From: Florian Höch <lists+argyllcms@xxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Thu, 20 Jul 2017 17:02:19 +0200
Am 20.07.2017 um 16:47 schrieb KU:
The profile has a PCS of *XYZ, which is difficult for LUT profiles. Try
better CIE*Lab as PCS for LUT profiles.
Not sure what the basis for that statement is. In practice, it's
probably the other way around: L*a*b* PCS encoding is limited to -128
+127 in the a* b* direction, making it unsuitable for wide gamut
displays whee the green primary usually exceeds the encoding range.
Also, you always have to encode the whole (within encoding limits)
L*a*b* gamut (a large portion of which will be unused) in the B2A
table(s), reducing the efficiency of the cLUT and potentially increasing
the chance for quantization artifacts (XYZ cLUT can use a matrix to
limit the encoded XYZ space, and this is the way DisplayCAL generates
B2A tables for this type of profile).
Florian.
Other related posts:
