on 1/13/01 10:07 AM, Frank Paris at [EMAIL PROTECTED] wrote:

> That's not necessarily true at all. The 50 colors could have a much wider
> gamut than the 100 if for example the 100 were different shades of green and
> red but it didn't have any blues, and the set of 50 had shades of all three.

Naturally if the 100 colors were all shades of green and the 50 had a mix of
other colors. The analogy was intended to be quite simplistic. Therefore, if
I have a mix of RGB crayons and have 50 that compose those colors but you
have 100, you'll have a wider gamut.

You're not going to find a Working Space whereby all the colors are in one
area of the CIE chart (all green or biased towards green) but sRGB having a
more equal mix of RGB primaries thus making sRGB wider.

> Of course that's not how boxes of crayons are usually put together, but this
> is not a good analogy to explain what a color gamut is. A CIE chromaticity
> diagram is good for that.

True but it's not easy to describe a CIE chromaticity diagram without a
picture, hence the crayon analogy. I haven't purchased crayons in decades
but I don't recall every getting a box of just green but rather a mix
spanning the color wheel.

> The book *Digital Color Management* defines color
> gamut as "the limits of the array of colors that can be captured by an
> image-capturing device...

A display has a gamut as do output devices. The description above sounds
like it's talking about input (image capturing devices). It's semantics at
this point. The definition above is adequate for this discussion.

> Then on page
> 113 of *Information Visualization* is a color gamut chart that shows the
> monitor gamut far exceeding the gamut of printed inks and the gamut of
> printed inks entirely within the gamut of the monitor

I don't buy that! There are greens and cyan's in a CMYK SWOP like gaumt that
fall outside monitor gamut. A monitor can't display a pure cyan. In the
enclosed gamut map you cans see that the TR001 SWOP profile has areas that
fall outside ColorMatch RGB (a Working Space based on the PressView display
and about as wide a gamut for a display as you'll find). I hope sending this
35k attachment is OK with the list. I used Gretag's ProfileMaker Pro to map
the ICC profiles of these two devices.

In any event, typical display gamuts (and all the RGB Working Spaces) are
triangular in shape and it's quite common for some output devices to have
areas that fall inside of and outside of this shape. Output devices usually
don't have this typical triangular shape. Again, look at the gamut map
enclosed. 

Andrew Rodney 

SWOP vs. ColorMatch

Reply via email to