Re: Creating ICCs with pigmented inks
Re: Creating ICCs with pigmented inks
- Subject: Re: Creating ICCs with pigmented inks
- From: Graeme Gill <email@hidden>
- Date: Wed, 03 Jul 2002 17:00:00 +1000
Chris Murphy wrote:
>
I meant to write illuminant dependent - as in the color you get depends
>
on the spectral reflectance of the inks. And some inks are a whole lot
>
more dependent on the illuminant than others. And by illuminant I mean
>
spectral power distribution. Everything has some level of illuminant
>
dependence. I think the term used when comparing the shifting of a color
>
sample under various light sources is color constancy.
Sorry, this doesn't make any sense. There is no non-linearity
involved. The light you see is the product of the reflectance
of the ink and the intensity of the illuminant. One ink
isn't more dependent on the illuminant, they are all completely
independent of the illuminant.
The only way to sensibly talk about this is to talk about
whether one combination of inks is a better spectral match
to the target than the other, making the tri-stimulus match
more or less independent of the illuminant.
>
The original Epson pigment inks don't match up well spectrally to press
>
inks at all.
This is hardly a surprise. The aims, physics and economies of inkjet
printers are rather different to presses. This will continue
to be the case as manufacturers strive for wide gamuts etc.,.
and is the reason why simple CMYK curves are not sufficient
to make an inkjet emulate a press.
>
> Unless you are very lucky, or choose your
>
>inks to match just one target, 4 inks is not enough to create a general
>
>purpose spectral matching based color reproduction system.
>
>
While it might be nice to have such a thing, that's usually not the goal.
>
The goal is to get a colorimetric match, which is by definition is
>
illuminant dependent.
On the one hand you agree that only a tri-stimulus (aka metameric)
match is possible, but on the other hand you complain
that the spectral match is not good enough.
>
>Maybe on some profiling systems, but the better ones let
>
>you choose the illuminant you will actually be using :-)
>
>
Yeah but there is only what - one of these apps available? Profile City's
>
app is the only one I can think of that will compute something other than
>
D50 LAB from spectral measurements.
But this is exactly where manufacturers such as the one I work for
add value. They aren't selling apps, they sell solutions. You
don't have to understand how an Epson 10000CF can produce good
(no, excellent) proofs, nor do you need the tools we use to
achieve this, you just have to judge the results. We do the hard work.
Now if you are doing it yourself, on the cheap, then yes,
you have to use the tools that are available to you,
and you might not be able to achieve comparable results.
>
I've yet to see any pigment based prints containing anything other than
>
saturated colors have good color constancy (in various lighting
>
conditions). Now if you want to prove me wrong, send me an email and I'll
>
give you my mailing address. I'd like to see some proofs illustrating it.
Love to, but I don't have a marketing budget at my disposal :-(
>
> Obviously
>
>there are some who have done it badly. Pity the ink has been
>
>blamed for it.
>
>
Graeme, if it weren't the ink, why did Epson totally redesign it? Why
>
does it perform SOOO MUCH more like the dye based inks? I don't see a
>
paper change occuring to solve the problem. And I don't see a retroactive
>
driver update from Epson to fix the problem with the previous pigment
>
inkset based printers to try to solve the problem merely with black
>
generation. If I recall correctly they DID do this with the 5500, which
>
did very noticably improve the situation but did not solve the problem
>
and added noticable black dots in highlights and 1/4 tones (and hence
>
skin tones).
Since we deal with the printer on our terms, I can't really answer
for Epsons drivers, color tables, other RIPs etc.
>
>The one
>
>think I don't like so much about it is the amount of
>
>bronzing it has. The Ultrachrome is even worse in this
>
>regard.
>
>
Not sure what you mean by bronzing.
Bronzing is where the specular highlights (of the paper, not
the image!) are colored by the ink. This is an effect seen in metal
surfaces, and often the specular highlights have a bronze tint. On
"normal" glossy surfaces, the specular highlights are the color of
the illuminant. Seeing that standard viewing conditions are meant
to be non-specular, and standard measurements are 45/0 or 0/45,
this effect can often give the impression of gross color errors. It
is also often difficult to arrange the viewing conditions to be without
any specular effects. It is an effect that detracts from the general
usability of the proof. Press prints may also suffer from this effect,
depending on the nature of the ink.
>
I haven't heard anyone canning pigment based inks. I've heard people for
>
months and months complaining about a particular pigment based inkset
>
which happens to be used in Epson 2000P, 7500, 9500 and the 10000 (with
>
archival ink option).
And given the set of drivers and color tables, then maybe they
have reason to complain. This doesn't mean it can't work
well, just that it perhaps isn't as easy as people would like
(which is a fair criticism).
Graeme Gill.
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.