Re: CMYK files
Re: CMYK files
- Subject: Re: CMYK files
- From: Andrew Rodney <email@hidden>
- Date: Sat, 17 Feb 2001 15:25:21 -0700
on 2/17/01 3:01 PM, Lee Blevins at email@hidden wrote:
>
The way we look at the source is by determining what method the cmyk was
>
proofed at. For example, if the cmyk was output and proofed on
>
Matchprint, then we assume the source of the cmyk to be Matchprint.
My first question would be, how do you know it was proofed on Matchprint if
it's untagged (unless of course someone told you this).
2nd, you'd be amazed at how varied "Matchprint" is. I've profiled a lot of
shops running Matchprint and using ProfileMaker Pro's measure tool, I can
load the spectral data files and look at the differences in deltaE. I've
seen differences in average of 10 deltaE. That's a lot.
I see where you are coming from and I think it's quite reasonable
considering that with untagged files, we all get to play detective and your
approach seems to be logical and useful when we are presented with those
damn untagged files.
>
Here again, my suggestion would be to determine how the files have been
>
seen before. Were they printed on a press or were they show as Iris
>
proofs or what? However they were printed is the origin of their
>
profile.
Again, it's a guessing game. If the guess is close to reality, you'll be in
good shape. There are a lot of ranges to how accurate this all works out
based on how good our guesses are. I believe an Iris is as variable as just
about any other CMYK device so without a profile that actually describes the
conversion, it's once again a guessing game.
>
It's a safe bet to use a common contract proofing system as your origin
>
since any disupute of the color in cmyk is usually resolved by
>
outputting and proofing on one of these systems.
The word that concerns me here is "common" contract proofing system of which
I have never seen. I guess if you are willing to accept a pretty large
statistical and numeric difference between just one type of contract
proofing system you have to be willing to accept a much larger difference
among ALL such systems. If this were possible we'd only need one output
profile to describe all printing processes to press since the contract
proofer is supposed to accurately produce output that one can match on said
presses. So we might as well just load the SWOP default conversions from
Photoshop for all conversions and all untagged CMYK files. But I don't see
that being a very effective or accurate approach.
Andrew Rodney