Re: Quark CMS
Re: Quark CMS
- Subject: Re: Quark CMS
- From: Henrik Holmegaard <email@hidden>
- Date: Fri, 19 Jan 2001 09:28:14 +0100
The first
time I did the test I used CMYK EPS files that I
wanted to convert to a different CMYK space. Sounds
like there might be 2 problems with that idea. If I
have an embedded profile in the image for CMYK and I
want to convert to another CMYK color space, Can I
drop the embedded image into Quark and expect it to
convert if I use the CMS?
EPS(F) stands for 'Encapsulated PostScript Format'.
With this format color management must either happen before the EPS
is created or once it is inside the RIP.
QXP doesn't touch the EPS because it isn't supposed to.
Adobe has specified how to put ICC profiles into EPS as well as Color
Space Arrays.
If you place ICC profiles into the EPS using Linocolor 6 or
Photosphop 5/6, QXP will place them, and when it translates its
internal data structures into PostScript, the encapsulated PostScript
just passes into the PostScript stream untouched.
This way you can have an ICC color server snoop for the ICC profiles
in the PostScript job. But in reality this is a legacy workflow, you
should be using PDF 1.3 / 1.4 with ICCBased color spaces per object
in the file.
Or would I have to use the
"composite simulates separations" which I guess you
folks are stating doesn't work, right? So is the
answer 'you can't do a CMYK to CMYK directly in
Quark?'
You can, but it's hardwired to relative colorimetric (: the last time
I looked -:)).
The second time I did the test I used RGB TIF files to
convert to CMYK. Now I see different results as
expected.
TIFF can be interpreted at any point in the color managed workflow,
whether in the scanning application, the retouching application, the
layout application, etc.
This is why you see a difference with TIFF and not with EPS.
Does anyone know what CMM uses Quark uses if
I have ColorSync set to Automatic?
Stay away from this option as it gives inconsistent results.
ColorSync is trying not to play policeman and in the process leaves
the street in a riot -:).
When I did the
second test with RGB TIF files I did a P2P in
photoshop from RGB to CMYK and then took the same
image in it's RGB state w/ the embedded RGB profile
and told Quark to use the CMYK profile on print. So
they don't match well as expected.
Fatal Pshop 5.0 - 5.5 bug documented time and again on the List. Get
Pshop 6.0 and it works the way it should - along with half a dozen
other fatal bugs now fixed.