Re: What color do SWOP neutral builds create?
Re: What color do SWOP neutral builds create?
- Subject: Re: What color do SWOP neutral builds create?
- From: Roger Breton <email@hidden>
- Date: Fri, 25 Jun 2004 09:36:09 -0400
Randy,
Sorry for my late reply.
>
In an ideal world, what color will I get if I send SWOP neutral
>
builds to a "perfect" SWOP output device?
First, where do you get "SWOP neutral build" from? 9th Edition SWOP Book on
page 18? OK.
>
Photoshop tells me that - when I load SWOP coated as my CMYK space-
You mean Photoshop v6+ and US WebCoated SWOP v2.
>
I build a few swatches in photoshop
>
>
25, 16, 16 (cmy) shows a Lab value of 79, -2, -3
>
50, 39, 39 shows a Lab value of 60, -2, -2
>
75, 63, 63 shows a Lab value of 43, -3, -2
That's true when ConversionOptions are set to RelCol in ColorSettings. When
you choose AbsCol things are a litte different, more inline with your (and
everyone's) expectations.
>
Is photoshop saying that my (predicted) printed output will be those
>
Lab values? Those are quite bluish-greenish for neutrals. I had
>
sort of expected something closer to L, 0, 0 for output
That last property seems to be true for RGB color spaces more than for
presses, in RelCol. RGB "working spaces" like AdobeRGB or sRGB or
ColorMatchRGB are designed to have that property work. But on a press of a
proofing system, that property may not be berifiable. Why? In my new, it
depends on how the original press colorimetric data was measured. There are
two agreed-upon techniques. Iin technique A, the ink patches are measured in
absolute terms, relative to what's call the "perfect diffuser", the default
more for all spectrophotometers, then the mixing formula for coming up with
acceptable neutrals by the profiling package, strictly speaking, is only
applicable to absolute colorimetry. In tehnique B, the data is measured
relative to the paper, some people like to do that (for reasons of their
own). Then the mixing formula for coming up with acceptable neutrals by the
profiling package, applies at least to relative colorimetry and possibly to
absolute colorimetry. I never tested the latter.
What this means is that, in technique A, in order to meet ICC
specifications, all colors need to be rid of the substrate particular shade
so that C=M=Y=K=0 should yield L=100 a=b=0. Having done that, in the case of
SWOP media which has a slightly yellowish tint, what you get when you look
at the published neutral receipes, in relative colorimetry, is some kind of
complimentary color appearance. So that, if the paper was yellow to begin
with, you end up with its opposite, blue. I believe that the expectation is
that since you are going to be printing on this yellowish shade of paper,
the bluish-green shade you get, in RelCol, will be neutralized by the paper
color at the time of printing.
That's my take. I'm open to criticisms, amplifications, corrections...
>
If I use Sheetfed Coated as my CMYK space,
Which is based, as we all know, on a Matchprint at 175lpi on some unknow
basis (commercial, publication? -- you can find more if you look at the
media white point tag, unless Adobe has tampered with that information,
too?).
>
25, 16, 16 (cmy) shows a Lab value of 74,-2,-2
>
50, 39, 39 shows a Lab value of 54, 0, 0
>
75, 63, 63 shows a Lab value of 37, 0, 1
Exact. Try the same using AbsCol and you should get the same figures. Could
be a function of Adobe "window-dressing" or "re-arranging" the figures. I
can't say for sure. If I run the SWOP published neutral on a custom-build
ColorArt prpfiole, at 175 lpi, in RelCol BPC off, I get :
25, 16, 16 (cmy) shows a Lab value of 78,-1,-1
50, 39, 39 shows a Lab value of 59, -2, 2
75, 63, 63 shows a Lab value of 42, -2, 3
In AbsCol, I get :
25, 16, 16 (cmy) shows a Lab value of 72,-1,-1
50, 39, 39 shows a Lab value of 54, -1, 1
75, 63, 63 shows a Lab value of 38, -2, 3
Those values are more in line with MY expectations.
It depends on the instrument I used to read the target patches. Such as, was
it fitted with a UV filter? How fluorescent was the paper the ColorArt was
laminated on? Those variables would all affect the outcome. Notice also that
as the formula corresponds to a ligher and lighter shade of CMYK the paper
starts to show up more and more through the ink and therefore influences the
final color. If the paper is very blue then the corresponding mix will
become bluer and bluer. If the paper is very yellow, as in the now defunct
Champion TextWeb #5, the corresponding mix becomes yellower and yellower.
>
I have dialed in a Polaproofer to print those L values, but 0,0 (dead
>
neutral) at the quarter tones, with good success, but recently have a
>
client complaining that the proofs are 'too red' compared to another
>
shop.
It's an endless starting over ("recommencement", in french). Why do we
always have to agonize over proofs: because that's the only language that is
tangible before we get to the press. So, I believe that all the time spent
arguing over proofs at the prepress stage, finding out how your printer's
proofs are made over your proofs, is time well spent. What it does,
invariably, is bring up to the surface all kinds of unspoken (and
unspeakable) assumptions one every one's involved parts about what the final
printed colo will be. Incidently, in a four color process brochure I did for
a client, I was arguying with a printer about a set of proofs I supplied vs
the set of proofs he was coming up with, from my supplied CMYK data. It took
a few interations but it turned out that he was simulating a ColorArt on his
9600 whereas I was simulating some press profile. In the end, I ordered a
set of ColorArt for my profiling package and I made a profile from it. When
I compared my ColorArt simulation on my inkjet printer to ColorArt
simulation of of their printer on their 9600, the truth rose to the surface
and it was clear to see where the discrepancy was coming from. It was time
well spent.
>
So, would a shop with a proofer dialed in to "SWOP" be used to
>
greenish -blue neutrals?
I don't think so. I don't see a Kokak Approval or Fuji FinalProof will yield
"greenish-blue neutrals" for SWOP dialed-in values. Although they should in
RelCol if they were to print on real SWOP paper. But, because of the pact
that they are produced on a paper closer to real commercial offset paper,
it's possible that SWOP's published neutral will come out neutral after all.
>
Randy Norian
Roger Breton | Laval, Canada | email@hidden
http://pages.infinit.net/graxx
_______________________________________________
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.