On May 7, 1:35*pm, Roy Lewallen wrote:
Michael Coslo wrote:
Art Unwin wrote:

I don't know about waves but my understanding is that all colors come
from the mixing *of the three basic colors, or is it four? When you
mix frequencies I
would imagine you could arrive at all possible frequencies. *I think
you should drop the idea of waves with respect to frequency. If you
observe a rainbow how many basic colors are there in the mix!

The visible spectrum does not include "basic colors" It pretty much has
all of them. Well, not Magenta. Is magenta a color?

And when you talk "basic, are you talking Cyan, Magenta, Yellow, or Red,
Green, Blue. It makes a difference.

In a
projector isn't there just three filters required for a movie in
color? One thing you have to get into your mind is the idea of basic
temperature and mass without energy.

That would be a black body radiator at absolute 0.

It's not very simple.

Except for monochromatic light sources like lasers, light of any color
contains multiple spectral lines. Sunlight or, for example, an
incandescent bulb or red hot electric stove element contains a
continuous spectrum, or effectively an infinite number of spectral lines
or "colors". So you can't duplicate these with any finite number of
spectral lines. The interesting thing is that with only three spectral
lines (pure monochromatic colors) you can produce light that *looks*
line nearly any color of light that's really made from many spectral
lines. For example, (transparent) box A can contain an incandescent bulb
whose light contains an infinite number of spectral lines or "colors",
box B can produce light with only three spectral lines, and you won't be
able to perceive the difference by eye if they're the right colors and
brightnesses. This is the trick that makes color TV and color film work.
It would be pretty easy to detect the difference with some simple tests,
though. For example, the light from the two would look like different
colors after passing through various color filters. Or pass the lights
through a prism, and you'd see many more colors in the light from the
incandescent bulb than the three-color source.

But you can't make all perceived colors from any set of primary colors
-- various choices of primary colors give you certain ranges of colors
you can mimic. RGB and CMY of particular wavelengths give wide ranges,
which is why they're common, but no choice can mimic all. I notice that
some color printers have more ink colors, which I assume allows an even
wider range. Creating light by combining colors is a different process
than filtering white light by subtracting colors or letting only certain
colors through. So different primary color sets are required. It's a
fascinating topic, and yet another example of how our eyes deceive us.

Roy Lewallen, W7EL

Further on this topic: one of the interesting things that comes up
when you look into "white" LED or fluorescent lights is that they
pretty much all have "holes" in their spectra. That is, the spectrum
they emit isn't continuous and the same shape as with light from an
incandescent source (including sunlight). The result is that some
things which reflect strongly over a narrow spectral band and much
less outside that band will look funny under such an LED or
fluorescent light. The light reflected by the object under such a
light doesn't have the "right" spectral shape. But it's something
that the lamp manufacturers are paying special attention to these
days, and you can find ratings on many bulbs about how good a job they
do at color rendering.

Cheers,
Tom