Judy Seigel (jseigel@panix.com)
Mon, 25 Jan 1999 16:46:03 -0500 (EST)
On Mon, 25 Jan 1999 FotoDave@aol.com wrote:
> In a message dated 1/25/99 6:45:23 AM Pacific Standard Time,
> kthayer@pacifier.com writes:
>
> > I've had the feeling that people
> > didn't always understand that when they change the file to CMYK, the CMY
> > values are adjusted in a way that's not straightforward, to keep the
> > density of the pixel the same as black is added. I've had the feeling
> > that some believed that the RGB values were simply converted straight
> > across to CMY and black was simply added, adding density to the file,
> > which is not the case in Photoshop.
> I have a feeling that we were not disagreeing on ONE thing, but we looked
> disagreeing because we were talking about different things! But this thread
> actually shows that it is good when people can clarify more and describe more
> because like right now, I think it is very very clear what particulars and
> details that we are talking about.
Well, I'm not quite as clear as Dave might wish, because I find this
*theoretical* discussion not true to the facts as I have stumbled onto
them.
In the first place, it isn't clear to me why you can supposedly adjust the
colors in RGB and not in CMY -- with or without K. I may be living in a
dream world, but I fancied I was adjusting channel colors separately in
Photoshop CMYK.
True, I couldn't get through that chapter in the manual, but when I
clicked on the channel and then selected contrast/brightness, I could
change all or a part of the photo in that color. (I noticed that "levels"
wouldn't work for one channel, but was able to do what seemed appropriate
with that control.)
But more important, it seems to me that this theoretical discussion has
gotten off the track, seriously, in a much more crucial respect. (As they
say, theoretically the bumblebee can't fly.) First of all, you *can*
program your separation in Photoshop.... As I *without reading the manual
& certified digital luddite* managed to do. If I can do it, so can the
scaliest darkroom lizard.
I did the following:
Printed out 4 sets of color separations of 4 colors each by the Photoshop
default values:
One set for the standard default "Medium Black," meaning that the black
values were for medium.
One set for the standard default "Light Black," meaning -- surprise, that
the black would be light.
[Note that I also could have adjusted the black to any level and changed
the curve of cyan and/or magenta, had I so desired, which I did not, at
that point.]
One set for "No Black," meaning --- ta da!! -- separated for CMY only, no
black.
And one set for UCR, which is Under Color Removal. In other words, the
colors in a dark tone are pretty much removed, leaving the darks to black
alone.
This last, by the way is not actually to save ink, but according to one
source, because commercial printing papers can't take a lot of ink, and
this prevents a buildup of excess ink where all 4 plates might print. Of
course this is something else that doesn't apply to gum, because most
methods of gum can take 10 or 20 coats. But I tried this "algorithm" too.
Now I know you're all sitting on the edge of your seats, wanting to know
the denouement. Well folks, that's the point of this message. Take your
choice. I did my best to print all identically (an artificial situation at
best, since in reality I print NOTHING identically), but feeling the
experiment required it. Basically the emulsion mixes were identical, but
to salvage a "print" from some I had to change exposure times and
development methods as I went along.
To begin with, although I had a black printer for all 4, I didn't print
the black with any of them. They looked so strong, bright, and INTENSE as
they were, that black was beside the point. Got this folks? Even though
in three of them black was theoretically called for, there was no need --
not even a *place* for the black. (As noted yesterday, a different image
shot under different conditions does seem to call for black, but that's
not this story.)
I liked all 4 prints. In fact I was quite tickled with them, so to speak,
although the dots were large (not enough printer memory) and the prints
were too small (only 7 by 9, we need a new printer). They did look
different, but to me the differences in the print had NO relation to the
theoretical differences of the separation method. That is, the medium
black (my favorite, by the way) didn't seem to explain why it differed
from the light black, etc.
Furthermore -- and this calls for capital letters, so here goes:
BY VARYING EXPOSURE, PIGMENT MIX, RATIO OF GUM TO DICHROMATE, AND TIME OF
DEVELOPMENT, I COULD MAKE ANY ONE OF THE "SEPARATION STYLES" PRINT LIKE
ANY ONE OF THE OTHERS..... IF I COULD FIGURE OUT THE DIFFERENCE.
And permit me to put it another way (and picture this in boldface):
THE SYNTAX OF GUM PRINTING IS STRONGER THAN THE SYNTAX OF ALGORITHM. At
least as I gum print, and that's not even counting the very great
differences in "look" among the prints on different paper and with
different sizing methods -- which I did some tests of, too.
I'll add that I began with the mixture of quinacridone red, thalo blue and
hansa yellow that I used a couple of years ago to print Photoshop tutorial
separations a friend sent me. They had to be diluted some for this
exercise -- and I suspect other color modifications could be made as well.
But that's not the point.
The point is that this argument is beside the point. Also that from where
I sit, this operation is much much much easier by Photoshop protocol, that
the notion that that's going to make your print look like a commercial
offset print is *highly* theoretical, ie., mistaken, and finally, that
having taught the RGB business to photo majors for making analog
separations, I will say, and trust me on this -- that's an extra barrier.
Another hurdle. Another impediment.
The idea is to make the method *accessible*. Not to set up a 2nd class
citizenship: digital wonks vs. peasants. Many would like to try color
separated gum, but are daunted by the seeming complexity.
My conclusion is, let Photoshop handle the complexity.
As for Phil Davis on color separation -- I sent a letter to the editor of
the magazine pointing out several errors Davis made about gum. He replied,
not sweetly, "I don't do that to my writers," meaning print a letter
correcting their mistakes. I was surprised to learn that the NY Times is
out of step in that regard, though no publication is perfect. In any
event, although Davis is obviously a polymath, I don't take his word as
engraved directly on stone tablets, especially against my own experience.
Judy
This archive was generated by hypermail 2.0b3 on Sat Nov 06 1999 - 10:06:44