RE: RGB vs CMYK: some experiments

From: Keith Gerling ^lt;[email protected]>
Date: 12/05/04-02:12:32 PM Z
Message-id: <>

Hi Jason -

You are correct. In the context of the question posed by Darryl Baird, my
response was wrong. I should have read it more closely. In addition to
this thread, I'm having two other "discussions" offlist, and one party was
using your example to educate me on the shortcomings of CMYK.

It's sort of funny: you perform a service by posting a nice demonstration.
Another party uses YOUR data to try to make a point that YOU never try to
make, and then I criticize YOU. Sorry! I'll try to restrict my
participation to one argument at a time...

Sort of related to all of this: I found a print that I made for a
commercial photographer a while back. It shows, on one piece of paper, four
examples of one image, with negatives made from four separate spaces. One
is RGB, and the other three are different variations of CMYK. I can post
this if anyone is interested. The problem is, I don't remember what the CMYK
variations are.


-----Original Message-----
From: Jason DeFontes []
Sent: Sunday, December 05, 2004 9:31 AM
Subject: RE: RGB vs CMYK: some experiments

I am certain that "my model" (if you are going to burden me with owning it)
is flawed in numerous ways, but I don't think the fact that I presented my
results on a monitor is one of them. Those among us who do testing with
actual prints also present their results to us on a monitor, and we look at
them and understand them just fine. I have read books that demonstrated the
workings of RGB, and the ideas contained in them were not compromised by
being printed on paper.

I do concede (in fact I already have) that my presentation of the results of
RGB separations was critically flawed by the fact that I used a biased CMYK
working space to combine the RGB separated color data. What I thought at
first was showing me potential inaccuracy in the RGB approach, was actually
demonstrating how the default CMYK working space adjusts the colors to
compensate for the non-linearity of SWOP inks. So for those of you following
the thread, please *do not* interpret my results as in any way demonstrating
a "problem" with the RGB separation approach.

However, what Darryl was referring to specifically in the previous post was
the use of 0% percent black generation to get the CMYK conversion in
Photoshop to produce 3-color CMY separations, without the black. I do
believe my example illustrates the potential for this approach if you wanted
to use a CMKY working space but only print with 3 layers as apposed to 4.
Subject to empirical evidence of course...


> -----Original Message-----
> From: Keith Gerling []
> "So again, what, if anything is wrong with this model of Jason's?"
> It attempts to demonstrate the workings of CMYK, but it does
> this on a video
> monitor. Were Jason (or anybody) to go to his page and print
> the entire
> thing on paper, the results would likely be different.
Received on Sun Dec 5 12:11:15 2004

This archive was generated by hypermail 2.1.8 : 01/03/05-09:29:43 AM Z CST