User-agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.7.2) Gecko/20040804 Netscape/7.2 (ax)
Todd Shirley wrote:
This may be an obvious question, but I have to ask it just so I'm
clear: All this talk of gamut-mapping and gamut compression and color
transforms really only refers to the perceptual rendering intent,
correct?
Perceptual and saturation.
> What, if any, advantages do DLPs offer to colorimetric
rendering? Clearly out-of gamut colors have to be dealt with in some
fashion, but do DLPs re-map these better than the 2 profile conversion
method, or is it more a matter of different vendors' "secret sauce"?
There should be less secret sauce. The advantages that I see in
generating device links directly for colorimetric rendering,
IF the process only uses the A2B information (ie. it inverts
the output profile on the fly) are:
A much smoother, more accurate result.
Choice of black generation parameters.
Option of black through mapping.
I thought matrix based RGB profiles don't have ANY look-up tables, just
sort of a mathematical transform based on the colorants and repro-
curves, which can only give the colorimetric transform. So in that
sense there are identical "tables" (in that there is only the single
transform, which is called on no matter what intent you call for), but
really I actually don't understand how this works, so any illumination
would be greatly appreciated! How do device-links "optimize" RGB-to-
CMYK transforms, considering the above?
I think you're confused. Jan-Peter Homann's statement was about
matrix profiles in a conventional CMM situation. In generating
device links, there's nothing special or different about
matrix profiles, the fact that they represent colorimetric
information is a good thing.
