[argyllcms] Re: fwa compensation unexpected results
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Sat, 06 May 2006 00:26:16 +1000
Roberto Michelena wrote:
As it should. If it's all working, the papers should now match (although,
maybe not perfectly, since adding the yellow will pull the L down.
If the L goes down too much, it's not a good match, and your reference
is out of gamut as far as the proofing paper is concerned.)
This is quite true... not easy to find proofing papers with L>=95, and
when so, they tend to be either very expensive or ridden with
brighteners.
Maybe the gamut mapping strategy for the white point (and vicinity)
should be optimized differently (than the rest of the profile), even
for absolute colorimetric.
Yes, if your target is very bright, this can be a challenge.
We struggled to find a paper to match that used in the Cromalin
analog proofing. Often papers with high levels of FWA, have
poor basic characteristics that the FWA is meant to cover up.
Standard cheap office copy paper is a perfect example.
I usually view by the window, but I have to admit Lima's sky is a
permanent gray. We call it "rat's belly gray" and only in summer does
it recede a little to show the blue sky.
Normal glass filters out a lot of UV, so this could well be
a low UV viewing environment.
So I should start by measuring that illuminant? unfortunately I think
an Eye-One won't cut it, right?
Let's say I measure with my Eye-One now and get the custom spectra of
my daylight.
- should I measure over various days (cloudy, sunny) and daytimes and
average it? or is it a standard saying "measure at noon"? I guess it's
always indirect measurement, right? (through a window)
I think some Eye-Ones will do this, but I'm only going by what I've read.
From personal experience, I've used a Spectroscan, and a Spectrolino
for this sort of purpose. I wouldn't worry about the circumstances exactly,
pick something "typical", and then go from there.
- how would I "complete" the spectra towards lower wavelenghts? copy
from D50?
That's the tricky bit. The best first approximation would be to simply
repeat the last reading (380 nm ?) down to 300nm. Be prepared
to vary this a bit though. In your situation, if the proof
was still too yellow, reduce the extrapolated illuminant levels.
If it's too blue, increase them a bit. The main advantage
is that once it's right, it should work OK with other
combinations of media, without having to fiddle (in theory).
Regarding the -I flag in spec2cie... when dealing with common
spectros (Eye-One, DTP-70, etc) then I should always specify "A"
illuminant?
If you were intending to view the prints under standard tungsten
lighting, then 'A' would be appropriate. I think quartz halogen is
a little hotter though.
It seems FWA compensation in Argyll would only work if you have access
to spectral data for both your reference and your proof?
Correct. If you're reference was fairly free of FWA, and you were
viewing in actual D50, you might get away without it, but as soon
as you set your viewing illuminant to non-D50, you need spectral
information for the reference anyway, or there will be large color
mismatches, irrespective of whether FWA compensation is used.
Graeme Gill.
Other related posts:
