Re: CMYK separations on Adobe Photoshop

From: Katharine Thayer ^lt;>
Date: 11/03/04-03:25:37 AM Z
Message-id: <>

Keith, I got into this discussion when Joe said that only the CMYK file
contains correct color information and that color separations from the
RGB file will give a distorted color balance in a tricolor print, and
advised that people should always convert to CMYK before printing
separations. This is not consistent with my own experience, and I said
so. I suspect Joe still doesn't believe that I can print true color
(when I want to) from RGB separations, and I haven't found a way to
prove it for him, so that's just going to have to remain unresolved.

It seems to me that there are two issues here: the first is which file
contains the accurate information; to me the RGB file contains the
accurate color information and the CMYK (and remember, here we are
talking about the default CMYK space) alters that information. It is of
course possible to generate a CMYK space with no K that does not alter
the color information in the RGB file, and gives you three color
separations that should essentially be the same as the three color
printers you get from inverting the RGB channels. But the default color
space isn't anything like this; it alters the color information
considerably and gives you color printers that look very different from
the printers that you get from the unadulterated RGB channels.

The second issue has much more relevance to gum printing than the first,
and that is whether the difference in the color channels and the
separations between RGB and CMYK actually translates to a difference in
the final gum print, in other words whether CMYK results in a more
accurate color balance than RGB, or vice versa. In case I haven't made
myself completely clear on this point, let me say very clearly that I
have no opinion on this issue, as I have no information that would make
me lean one way or another on the relative color balance generated by
the two. I do know from experience that the statement that RGB cannot
produce true color is not an accurate statement, but that is as far as I
can go on the basis of my own experience. I've never printed from CMYK
separations and never wanted to, and seeing them makes me even less
interested, but I certainly do not and have never challenged anyone's
experience that says that CMYK separations can also give an accurate
color balance. I keep saying that I don't think there's a right or
wrong, just different ways of approaching the same target, please
believe me. Each to his own.

Keith Gerling wrote:
> I've been in and out of state campaigning, so I have n't followed this
> thread. But as a person who uses both CMYK and RGB separations, I have to
> say I find your observations curious. Specifically, your statement "because
> it makes no sense to me in terms of photography". Were you using GCR or
> UCR? It would make a big difference in how "sensible" it would look. I'm
> sorry if this has already been discussed.
> I would never suggest you change your procedure, but I have to say that my
> perception is entirely different. I "see" in CMYK, and the RGB-derived
> negatives always look strange to me. If I don't label them, sometimes I
> can't even figure out which is which. CMYK seems more naturally because it
> was designed to do exactly what we are doing: to enable a process to apply
> layers of pigment in separate steps. I think it has everything to do with
> gum printing. IMO. I use RGB and CMYK and they both work fine.
> Keith
> -----Original Message-----
> From: Katharine Thayer []
> Sent: Wednesday, November 03, 2004 11:12 AM
> To:
> Subject: Re: CMYK separations on Adobe Photoshop
> Kate M wrote:
> >
> > I found the comparison of the two sets of negs interesting.
> Thanks Kate,
> I wish I'd thought to do this years ago, it might have saved me some
> grief.
> I don't know that there is anything I have said on this list that has
> elicited more vehement opposition than the simple statement that CMYK
> (meaning here specifically the default CMYK space in Photoshop): (1)
> distorts the color information in the RGB file, (2) is about preparing a
> file for commercial printing with particular inks and particular papers,
> and (3) has nothing to do with gum printing. IMO.
> Up til now, I've just demonstrated the distortion by selecting one pixel
> in the color image and showing what happens to the color information in
> that pixel as you switch to CMYK, by giving the color numbers from the
> info palette. (Keith has also shown this, with various CMYK setup
> options.) But it wasn't until I printed out these CMYK separations and
> had them in front of me that I realized how well the separations
> themselves tell the tale. I was writing at the computer, with the
> magenta CMYK separation sitting on top of the little pile of separations
> on the desk in front of me. and it kept distracting me from what I was
> writing. I kept thinking, what IS that? So finally I stopped and picked
> the thing up and took a good look at it. I still don't know what it is,
> because it makes no sense to me in terms of photography, but I do know
> that I wouldn't care to try to print it, or the yellow printer either,
> because I don't see how they could possibly give me the results I want.
> This is not to say it's impossible, only to say I don't see from the
> negative how it possibly could, and I prefer to use a method that makes
> sense to me at every step.
> The separations from RGB files, I can look at each of the negatives and
> know immediately what it means in terms of the original color image and
> also what it means in terms of what I'm intending to do with the final
> gum print. But these CMYK separations are just... well, like I say, they
> make no sense to me at all. But they do show very well the extent to
> which the color information in the RGB file is altered in the switch to
> the default CMYK.
> Katharine
Received on Wed Nov 3 11:21:53 2004

This archive was generated by hypermail 2.1.8 : 12/08/04-10:51:32 AM Z CST