Red, green and blue are the primary colours of light ("additive" primaries). Each one emits more light, so you add up the total to get the final colour.
Red, yellow and blue are the primary colours of pigments ("subtractive" primaries). Each one absorbs more light, so you subtract to get the final colour.

benryves is right. If you remember science class where they teach you the wavelengths of light: ROYGBIV (red, orange, yellow, green, blue, indigo, violet), then you'll see why.

A reason why computers often use RGB is because your computer display uses pixels which can only display combinations of red green and blue.

There are two basic ways to get colors: additive and subtractive. The method typically used in art, graphic design, and other "analog" fields is based on RYB (or, more commonly, CMYK).

In CMYK, blending two colors subtracts them: base is pure white, and overlapping a pair of colors generally tends to make them darker. This is where you'll get black if you mush all the colors together. Subtractive models are close representations of what happens when you blend two pigments, such as paints or inks.

In computers, color is generated additively. Rather than mixing two pigments to color each pixel of the screen, monitors mix colors of light. The typical method is RGB. In this model, base is pure black (no light), and overlapping two colors generally tends to make them brighter, since you are combining two light streams.


Strict RYB also exists, but generally only informally in primary schools and the like; it is considered a poor physical model in the physics/optics realm and has far too many practical problems (like poor range) for use in printing.

Quote:Original post by swordfish
A reason why computers often use RGB is because your computer display uses pixels which can only display combinations of red green and blue.

Your causation is backwards [smile] Pixels are designed to emit red, green, and blue because those are the most effective additive primaries - the combination of them additively (i.e. by combining two light streams) yields the broadest and most efficient range of representable colors.

If gold, purple, and chartreuse were the most effective additive primaries, pixels would use those colors instead.

the RYB thing when used and applied in artist paints is actually as such:

There is no such thing as pure Red, pure Yellow or pure Blue when working with paints... I suppose it's theoretically possible to have it be a pure saturation, but that's never the case in practicality. Each one has a portion of a secondary colour.

Red can be either Red-Blue (purple) or Red-Yellow (orange)
Yellow can be either Yellow-Red (orange) or Yellow-Blue (green)
Blue can be either Blue-Red (purple) or Blue-Yellow (green)

As such, when using RYB in paint you actually need two of each primary in order to mix properly... You'll also want white and black because black is hard to mix and wastes a lot of expensive paint, it also never turns out due to pigment impurities especially in cheaper paints... And white because you just can't mix white from coloured pigments. They are each extreme hues which is actually the problem, paint mixing doesn't really take hue into account, you require the white and black to lighten or darken the colour.

As others have said, there is a difference in pigments and light blending when analyzed on a high level, but keep in mind pigments are simply the result of a substance which reflects and absorbs light. While mixing on a practical level is different, it's the same theory on a fundimental level... That is when you look at the actual light being bounced back into your retina.

Wikipedia knows

[Edited by - M2tM on September 19, 2006 8:53:20 AM]