Re: CMYK separations on Adobe Photoshop

From: Joe Smigiel ^lt;>
Date: 10/31/04-03:49:58 PM Z
Message-id: <>

>>> 10/31/04 5:20 AM >>>
Joe Smigiel wrote:
> >>> 10/29/04 11:48 PM >>>
> Sorry, but...
> How do you get the Cyan channel from a RGB file without going to CMYK?
> Isn't this the question?
> How do you get the Cyan colors?
> Giovanni
> I'm just printing the RGB red channel negative using a cyan pigment
> the image example, Linel Hortensia Blue). The green channel is
> with magenta and the blue channel gets printed with yellow pigment.
> I've said and the image example shows, the colors are different in the
> final image than when you do a CMYK conversion and make separation
> negatives from those files. If you want a more accurate
> you should do the RGB to CMYK conversion.

I'm sorry, I'm afraid I must disagree with this advice. It supposes
that the CMYK file has accurate information in it and the RGB file is a
distortion of that "accurate" information. This is backwards; the
accurate information is in the RGB file; changing it to CMYK distorts
that information to a greater or lesser extent depending on the CMYK
space that's used. As Keith keeps pointing out, you can't talk about
CMYK as if it were one thing, because there is an infinite number of
possible CMYK spaces. If you're talking about the default Photoshop CMYK
space, you are talking about a space that was designed to optimize the
printing of color images in commercial printing inks on coated
commercial printing papers; this space distorts the color information in
the original file to accommodate the characteristics and limitations of
printing inks and papers and has nothing whatever to do with printing
pure pigments in gum on art paper.

Joe's example is interesting but inexplicable to me. It seems to be
offered in part to prove that RGB files print with weird, brownish
distorted colors, but since I have always printed from the RGB file and
have always got very nice, clear and true colors, and since I know of
several other gum printers who also work directly from the RGB file and
also get beautiful and true colors, to my mind the general conclusion
Joe seems to be drawing, that if you print from RGB files you will get
weird and distorted colors, is problematic because there is so much
available evidence to the contrary.
Katharine Thayer


I don't believe I ever said the colors would be "weird, brownish,
distorted colors". I said there would be some distortion and that they
would be less accurate. (And note that I'm not even running for
political office here.) I know of no other actual print test evidence
to the contrary but, I'm eager to see it.

So, would you please cite a few of the several other examples of gum
printers using this method for me? I'd like to see their work. I'm
aware that Christine Anderson, Phil Davis, and Sam Wang also use the RGB
method. Christine and Sam apparently do combination prints using a
cyanotype layer with gum. I'm not sure I would consider their work to
have "accurate" colors although certainly beautiful and interpretive

Do you know of any examples of side-by-side, RGB vs CMYK 4-color gum
image tests which show evidence contrary to my online example? Please
point me to them.


Received on Sun Oct 31 15:48:30 2004

This archive was generated by hypermail 2.1.8 : 11/03/04-10:51:23 AM Z CST