Re: PS CMYK Conversions
Re: PS CMYK Conversions
- Subject: Re: PS CMYK Conversions
- From: Terence Wyse <email@hidden>
- Date: Fri, 15 Aug 2008 20:43:12 -0400
All great points Hanno, which is why, after thinking it through more
carefully, I should've recommended workflow #2 right from the get-go
instead of confusing matters with possible workflow #1. Starting with
a LAM image/proof from the very beginning, you're not setting yourself
and the customer up for disappoint down the road.
Regards,
Terry Wyse
On Aug 15, 2008, at 6:06 PM, Hanno Hoffstadt wrote:
Hi Terry, Bob,
let me add that your workflow #2 is along the lines that the ECI
working
group "coating" (which should read "varnish and lamination" :-) is
about
to recommend.
Workflow #1 has the disadvantage that you prepare and proof your
images
for a larger gamut that you can later achieve after matte lamination.
By converting from UNLAM to LAM, you are trying to compensate; and
yes,
there will be some improvements with regard to hue and gradation.
But heavy gamut mapping is involved for all saturated and dark colors
that your UNLAM separation might contain. Just imagine, the dark end
of the gamut has to go from some L* ~ 15 to L* > 30.
And if your client has seen and bought the UNLAM proofs, he will be
disappointed with the washed-out LAM result.
Workflow #2: first a recap, then additional points...
As Terry said, make your separation for the LAM profile (best if done
directly from RGB data of course), which gives you LAM CMYK data.
Make a LAM proof for the client and do color corrections and changes
until the client is happy. You are always staying inside the
realizable
gamut this way.
When done, send these LAM data together with an UNLAM proof from the
same CMYK numbers to the printer. (Terry, your guess is correct: the
proof will look more saturated, but also lighter, since lamination
will later cause additional dot gain.)
The pressman adjusts to the UNLAM proof, producing an UNLAM print.
Laminating the UNLAM print will yield a similar appearance as the
LAM proof has foretold.
Key points:
1) The LAM and UNLAM profiles must be a matching pair!
The underlying LAM print characterization data must correspond to a
lamination of the same UNLAM print target.
2) The LAM and UNLAM proofs should really be made on the same system!
If you add the possible proofing uncertainties of two systems, you are
likely to break the match again.
3) Obviously, it is very important to mark the LAM and UNLAM proofs
accordingly.
4) Varnish and lamination effects depend on the screen. The effects
are smaller for fine FM and AM screens than for 150-180 lpi AM
screens.
5) Be careful with PDF/X: an UNLAM output intent in a PDF/X file will
cause unwanted conversions if LAM profiles are embedded.
I think you need to explicitly specify the LAM output intent here
(the same profile that has been embedded, to avoid transformations).
6) It may make sense to provide the printer with both UNLAM and LAM
proofs - if he can do inline varnishing, for example, the pressman
can adjust the final colors.
7) (Arcane) you can laminate the proof for a better visual result,
*but* your proofer profile needs to be made from a laminated proofer
target. - It is wrong to assume that simply laminating an UNLAM proof
will have the same effect as laminating an UNLAM print.
Finally, an announcement:
I'm (still) working on a synthetic characterization data set based on
FOGRA39 which simulates matte and glossy varnish/lamination. This is
to address key point 1: to provide matching companion profiles to e.g.
ISOcoated v2. Anyone interested in testing the profiles is welcome to
contact me off-line.
If there are any comments or suggestions, please let me know.
Best regards,
Hanno
--
Dr. Hanno Hoffstadt
GMG GmbH & Co. KG
Kirchstraße 5
D-89081 Ulm
Germany
phone: +49(0)731 38900957
mail to: email@hidden
------------------------------------------------------------------------
GMG GmbH & Co. KG. Registered Office: Tübingen.
Registered in Tübingen, Germany at Stuttgart Local Court, No. HRA
381463
General Partner: GMG Verwaltung GmbH.
Managing Directors: Joerg Weihing, Robert Weihing.
Registered in Tübingen, Germany at Stuttgart Local Court, No. HRB
382505
Banking affiliation: Volksbank Filder eG, Account No: 65 000 005,
BLZ 611 616 96, IBAN: DE65 6116 1696 0065 0000 05,
SWIFT: GENODES1NHB, VAT-ID No: DE 230 302 334
On 15 Aug 2008, Terence Wyse wrote:
On Aug 15, 2008, at 1:41 PM, Robert Rock wrote:
I'm a bit confused with the converting/assigning back and forth:
Let's say that after color correcting we convert to the UNLAM
profile and
proof one last time. Client says OKAY.
Just to be clear, you DON'T show the client the UNLAM proof, you
show them the LAM proof because that represents the final results.
You proof the LAM image using the UNLAM profile as a guide (only)
for your pressroom to hit. Once this gets lamintated, it should
them look like the LAM proof. The UNLAM proof should not go to the
customer since it represents what the image looks like PRIOR to
lamination and may look kind of whacky (oversaturated and possibly
too light, just as a guess).
[...]
_______________________________________________
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