Re: ProfileMaker Yellow problems
Re: ProfileMaker Yellow problems
- Subject: Re: ProfileMaker Yellow problems
- From: neil snape <email@hidden>
- Date: Sat, 08 May 2004 09:53:58 +0200
on 08/05/2004 7:08, citizenray wrote :
>
This issue in not unique to ProfileMaker. I've seen at least three other
>
packages build the same "greenish-yellow" in more than one shop with
>
different makes and models of spectrophotometers.
Yes this is true as if the spectral readings see the yellow not being bright
enough and or hue angle quite far off most packages truncate the notion of
intelligent mapping. According to the math this should fetch accurate
adapted results yet most peoples vision says it's not.
This doesn't happen all the time, and probably not often to the degree that
it's totally unacceptable. When it does happen it presents the problem of
what to do with the yellow that doesn't jive with what your eyes say.
I see the questionable yellow mapping essentially on papers that are not
very bright , especially when the yellow colorant is "thick and saturated
but hue angle towards the orange side. I see this on spectros from both
Gretag and X-Rite , yet can't say for other. Over the years all the ICC
profile applications have encountered this, and the same developers have
tried to predict error correction or just correction for some of the
factors like UV fluorescence etc. Most of the time this brings the resulting
profile in line and will have the optimal characteristics compared to
editing by hand the subtle discrepancies your eyes see.
>
Pantone folks will be the first to explain that one can not always trust the
>
math (and resulting Lab values to make the colors) from which such
>
transforms are made. Better color matches are achieved via some human
>
interpretation and intervention and that takes time which demands money.
The CIELab model is probably correct , it would rather be the derived values
of the hardware that have not seen what your eyes are telling you.
>
Peter Figen's solution of editing profiles is fitting and probably the best
>
quick-fix of the problem for the time being. Unfortunately, he has to take
>
the time (and therefore money) to make it work as well as we expect it
>
should out-of-the-box.
I edit profiles when I see the yellow problem too but be aware that editing
profiles breaks the profile to some degree. It won't show up on a
target/chart but it denatures the character and eventually will have some
failure somewhere. Now, that said, for the 95 times you prefer the solid
yellow and combinations thereof, the 5 times the profile creates some small
anomalies you can see which would be more important.
>
It's interesting that I've seen the issue with "all things RGB imaging" from
>
photo paper, Pictrography, and even LVT. Peter says he sees it with CMYK and
>
inkjet also. Andrew says he doesn't notice the problem.
I don't see this on all media , all the time. Just on certain media >see
above.
Neil Snape nsnape @ noos.fr neil_snape @ mac.com
http://mapage.noos.fr/nsnape
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.