[argyllcms] Re: cctiff output colorspace and the consistency of appearance
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Fri, 29 Apr 2022 11:10:10 +1000
Wire ~ wrote:
Hi,
Per Lindbloom's RGB Working Space Information:
http://www.brucelindbloom.com/index.html?WorkingSpaceInfo.html
Common Colorspace Coding efficiency:
- Lab space: *35*% (most of Lab space are not colors)
- ProPhotoRGB : *87*% (most of are colors, 0.9 of Lab)
- Adobe RGB: *100*% (all codes are colors, 0.5 of Lab)
- sRGB: *100*% (all codes are colors, 0.35 of Lab)
I'm guessing that's gamut efficiency. So it doesn't
really cover quantization efficiency. The aim of L*a*b*
is to be perceptually uniform, so its quantization
efficiency is reasonably close to perfect (i.e. 0.5 dE at 8 bit).
RGB spaces can be as good with the right
non-linearity. So you can argue about which is better
in which part of the colorspace, but my understanding
is that what L*a*b* loses in poor gamut efficiency,
it makes up for in more perceptually efficient encoding,
so they end up about even.
The best examination of these topics
I've found for SDR spaces is:
"An Analysis of Selected Computer Interchange Color Spaces"
James M. Kasson, Wil Plouffe
ACM Transactions on Graphics, Vol. 11, No. 4, October 1992
(The article following that one is pretty interesting too.)
Cheers,
Graeme Gill.
Other related posts: