Re: Total Ink Limit (TAC)
Re: Total Ink Limit (TAC)
- Subject: Re: Total Ink Limit (TAC)
- From: Graeme Gill <email@hidden>
- Date: Tue, 13 May 2008 11:22:33 +1000
email@hidden wrote:
Graeme, correct if I'm wrong but I think you are saying that the best way to go would
be to make a newsprint version from the original, source, RGB files. If you don't have
those at hand you're trying to do something ICC never intended (to my mind at least).
I'm afraid I tend to regard any other plan as confused thinking. If you are
targeting a particular device, then target that device. All these
products with "proofing" and "preview" options confuse what is
essentially a set of simple conversions, each composed of
a pair of device profiles.
You have a photograph or source image in a particular (RGB?) colorspace.
You are targeting a newspaper press. The gamuts are probably
going to be quite different and you are after a pleasing result,
so perceptual intent is indicated.
If the newspaper profile is setup with specific gamut mapping
for the source, you can simply do a:
RGB source space profile -> PCS -> (perceptual) Newsprint CMYK space
transform.
If your Newsprint profile has more general gamut mapping, then perhaps you
will need to adjust the RGB source in Photoshop to make the gamut
fit and get the result you want.
You can examine the output ("proof") it using devices other than
the actual press pretty simply:
Newsprint CMYK space (colorimetric) -> PCS -> (colorimetric) proof device colorspace
The proof device may be a monitor or an inkjet printer. You may
choose relative colormetric to proof the gamut, or absolute colorimetric
to proof the white point shift as well. The above sort of transform
make one device (the proofer) emulate the colorspace of the source
(target device newsprint).
Now if the first transformation is performed twice, once to
create the output that is sent to the press, and the second
time within the proofer inkjet RIP, then there is no
guarantee that it is actually the same transformation,
and that what you are proofing is the same as what's being
sent to the press. If the correspondence is close enough,
it may be a shortcut if multiple rounds of manual adjustment
are needed, but for the final proof, the proofing RIP
conversion to the target device really shouldn't be used.
Adding to this there are other problems. Normally, I would say that contemplating a
conversion from a 340% TAC file to a 220% profile only exposes a limited understanding
of the behaviour of printing inks, substrates and printing processes. A (properly
constructed) newspaper profile will have a completely different GCR/UCR and Black
generation specification based on the low viscosity inks and highly absorbent uncoated
stock not to mention the relatively high speed of the press. How do you reverse
engineer this? To my mimd the short answer is... you don't. OR you shouldn't. Without
GCR or any black generation it is possible to recreate the RGB values for the
separation setup (a la the Hutcheson Crosfield drum scanner profiling method) from
which you could then create a proper newsprint profile with appropriate GCR and Black
generation.
Right. A good newsprint profile will not only represent the behavior
of the device accurately (A2B table), but will also transform PCS values
appropriately into that space (B2A table), taking care of TAC, black
generation, and the nature of how the inks mix.
Graeme Gill.
_______________________________________________
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