Primary Colors: RGB vs. Red-Yellow-Blue

Janx

Hero
In the vein of Bullgrit's science questions, here's a sciency question:

What's the technical cause of the difference between Red-Green-Blue being the primary colors vs. the "traditional" Red-Yellow-Blue that art uses.

As background: in TVs, LCDs, or just about anything electronic generating color, the 3 colors of Red, Green and Blue are used to generate any other color.

In art, the toilet (remember when You first learned Yellow and Blue make Green!), and even crayons, Green is the product of Yellow and Blue, not the other way around.

For art, it's not a simple chemical reaction (blue paint + yellow paint doesn't bubble up into green paint by way of chemistry). You can get the same effect from urine + tidy bowl cleaner, or any kind of yellow and blue paint (using different chemicals to do so). You can also lay 2 pieces of colored cellophane over each other and see the effect (no chemistry involved).

So, what's behind the difference in the primary colors? Why don't TVs do RYB?
 

log in or register to remove this ad

So, what's behind the difference in the primary colors? Why don't TVs do RYB?

The roots lie in how your eyes receive, and your brain actually perceives, light. You've got a few different types of photoreceptors in your retina - rods for black and white vision, and cones for color.

You have three types of cones, which receive best in Yellow, Green, and Violet. The differences between the levels of each that your brain receives can be used to interpret a wide variety of colors. Say a few photons of blue light enter your eye - you don't actually have receptors for blue. Your cones say that your Green and Violet receptors kicked off pretty strongly, and the Yellow weakly, and your brain decodes that as blue.

*Any* light that kicks off your yellow receptors weakly, and the other two strongly, will look "blue". You may shine pure dim wavelengths of yellow, and strong pure green and violet, and your eye won't know it isn't actually blue. It is the relative levels that matter.

So, any mixing plan that hits the same relative combination will get you the same perception of color.

RYB (what you learned in school with paints) is subtractive mixing. You start with white light, and then take away wavelengths - start with white, take away some wavelengths, and what is left may not actually be photons in the green range of the spectrum, but your eye reacts as if it were.

RGB (what's used in video screens) is additive mixing. You start with no light, and you add levels of various lights until you hit the characteristic pattern of firing cones. Again, you may not be firing actually yellow light at the person, but if the relative pattern of firing neurons is the same, you'd perceive it as if it was yellow light.

RGB and RYB are by no means the only triads you can use. And those triads can't actually reproduce all the colors humans can distinguish. They are chosen because they are convenient and fairly effective at reaching most of what you can perceive, and are convenient for the technologies they're used with.
 
Last edited:


And don't forget CMYK (cyan, magenta, yellow, black) for printing colors.

Bullgrit

which is kind of cyan = purplish blue, magenta = purplish red so it's a variant of RYB.


Good answer. I'd give XP to umbran, but apparently I already approve of what he said sometime earlier.
 

Adding to the CMYK info, in theory you can use only cyan, magenta, and yellow printing inks to create any color, but the reality is the same as if you mix blue, red, and yellow paints yourself: no matter how you balance them, the darkest you can make them is a muddy brown. The black ink (K is for "key", though it's also a useful way to distinguish it from the B in Blue) is essential for getting truly dark and even black tones.

In printing, though one would think that 100% black ink provides a utterly deep black, the reality is that it's only a very deep charcoal color, one that can vary based on the desire to produce good looking photographs and flesh tones and such (ink densities are adjusted on the press at the beginning of the run and throughout to ensure that everything continues to look good). For that reason when preparing something for print you often create what's called a "rich black", a mix of black and the other three inks, to really deepen the color. The precise mixture can be varied in order to produce a "warmer" black (more magenta and yellow) or a "cooler" black (more cyan). While one might propose using 100% of all four colors to create the deepest possible black, on the press the paper gets far too wet when there's so much ink applied, either warping the paper or having the ink transfer to other stacked sheets before it gets a chance to dry.

The actual gamut (color possibilities) of CMYK inks are actually more narrow than it would seem. In particular you can't produce truly beautiful oranges and greens, something thats somewhat evident by the colors of the inks themselves, e.g. when trying to produce a great orange you'd mix magenta and yellow, but due to the somewhat blue nature of the color magenta, it never comes out truly orange. As a result several alternate printing systems have been developed. For a while Pantone's (major manufacturer of press inks) Hexachrome system was somewhat popular, which directly added orange and green, creating a CMYKOG system. Unfortunately it wasn't ever sufficiently popular (largely due to press limitations and Pantone's patent protection) and is no longer in use.

Lastly, what others have posted along with what I added here only barely touches the surface of subtractive and additive color. One primary reason is the fact that the human eye is actually pretty lousy at seeing a very large range of theoretically visible colors (stuff in the "visible light" spectrum, not down into infrared or up into ultraviolet). Another is that really no color model accurately reflects the possible colors that exist -- there's some seriously complex science involved. And finally, I brushed over the complexities of press ink and didn't touch stuff like under color removal, under color addition, gray component replacement, how halftoning dramatically affects what you can actually do, etc.

Crazy stuff!
 

color film photography used CMY (cyan - magenta - yellow) as subtractive filters for creating color prints.

Yes, I am showing my age. I as process some film and prints by hand in a photography class in high school over 15 years ago.
 

And in stage lighting any combination of three colors will get you an off white as long as they aren't part of the same color path on the color wheel. (i.e. pink rose and red will get you a very interesting red color.)

Straight white light from a spotlight is usually to harsh on the human eye and washes out the subjects so combinations of colors are used in stage lighting from different angles to achieve "white" light. Which is why concerts and stage productions have so many different colored lights on while the stage appears "normal". The color wash is only really used to indicate feel (red for anger or violence, blue for sadness or water effect, green for peace and outdoors, etc.)
 

Adding to the CMYK info, in theory you can use only cyan, magenta, and yellow printing inks to create any color

As I noted above, this is not true, even in theory. For both subtractive and additive mixing, the three colors chosen define what is called a "color triangle", and only colors inside the triangle may be created. Adding black does not change the colors that can be reached, but instead allows greater variation in overall saturation of color.
 

My brain is going to BOOM



ouch. too late



There are no easy answers anymore. The more we learn the less ofan easy true answer we can give.
 

There are no easy answers anymore.

That wasn't easy?

The laws of nature have not notably changed over the course of human existence. The answers have always been thus, we just didn't know it in the past. They (at least the correct ones) haven't gotten harder.
 
Last edited:

Remove ads

Top