Re: Creating a SWOP proof with an Epson
Re: Creating a SWOP proof with an Epson
- Subject: Re: Creating a SWOP proof with an Epson
- From: Terry Wyse <email@hidden>
- Date: Mon, 17 Jul 2006 09:10:52 -0400
On Jul 17, 2006, at 7:01 AM, Martin Orpen wrote:
On 17 Jul 2006, at 2:30 am, Roger Breton wrote:
Terry's point is valid, and by not responding directly to his
point you are
just dragging the issue further. Terry and myself and I'm sure a
host of
others on this list all well understand Fogra. We're not debating
Fogra's
tolerances but we think you are citing their numbers in the wrong
context.
All we're saying is that, Fogra or not, the proof has to be within
a close
range of the Fogra or SWOP or whathever reference dataset. How
well the
press is able match that is another issue.
If you "well understand Fogra" then you'd well understand the
figures that I quoted.
The proofs are "verified" to be good if the average dE is less than
4 and the maximum is less than 10 when the FOGRA test strip is read.
I won't pretend that I "well understand Fogra" because Fogra
guidelines are largely ignored or perhaps irrelevant over here in the
USA (if you want to argue THAT point then I would point you to the
"Printing Across Borders" mail list - there's a good little
discussion going over there at the moment!).
If I understand you correctly, the figures you quote are tolerances
to a subset (test strip) of the Fogra characterization data (which
data? I understand that there are many different data sets) and not a
measure of tolerance of the proofing system in day-to-day production.
To illustrate this point, I was just at a client where I compared
their proofer today to what it was 6 months ago. The average dE
compared to the last calibration was about 1.5 dE but yet the proof
had drifted an unacceptable amount in that time, especially in gray
balance.
Personally we prefer to run with greater accuracy here and have
never sent a proof to a client where the maximum exceeded 3 dE (I
don't want a client to be able to *see* a difference).
When you say the maximum never exceeds 3 dE, you mean no SINGLE patch
from the Fogra test strip ever exceeds 3 dE? That's pretty good if
that's the case (hats off!). When I was quoting "1 dE", that's an
average taken over an entire ECI2002 or IT8.7/4 testchart. But within
that average, you'll find that perhaps a dozen or so of the most
difficult patches could fall in the 3-5 dE range. But many of these
"outliers" are values that wouldn't generally exist in "real"
separations. Values like 100m100y100k could be 3 dE or greater but
you would likely never encounter a value like with the possible
exception of a 100k overprint of 100m100y. A value like that is akin
to some of the "imaginary" colors in Lab. Can you specify a value of
100y100k? Sure, but why would you? ;-)
If you can maintain less than 1 dE over the whole test strip then I
take my hat off to you.
It's a fact that maintaining a tight dE over a small test strip (<50
patches or so) is more difficult and, of course, raises the average
dE by quite a bit. Add to that the fact the test strip is likely to
purposely contain a subset of the more difficult patches from a
larger testchart. So, yes, <1 dE over a small test strip compared to
something like a Fogra data set could be challenging.
I don't see the point - especially if the proof is unlikely to even
meet another spectro, let alone the identical model to yours -
which is the only way you are going to get a contract proofing
system to work with tolerances less than 1 dE.
More to the point I was trying to make (and apparently unclear about
it), if I set up a proofing system to an acceptable visual match
agreed upon by my client and, after measuring the dE of the proof
compared to the source (press) data, it happened to be 2 dE average,
5 dE max, then I would want to MAINTAIN that average to <1 dE in day-
to-day production. If my system was unable to maintain that kind of a
tolerance, then I'd be shopping for a new proofing system.
Regards,
Terry Wyse
_____________________________
WyseConsul
Color Management Consulting
G7 Certified Expert
email@hidden
704.843.0858
http://www.wyseconsul.com
http://www.colormanagementgroup.com
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden