Topic: the arts and the sciences: the idea of primary coloars is horseshit
Sheila LaBoof started this discussion 9 years ago #59,113
As kids, we are often introduced to the idea of primary colurs - the idea that by mixing three special colaurs, we can make any fuckin collar. The history of this idea is as much in the world of the arts as in the world of the sciences, and is mixed up and misleading and pretty much wrong really. Anyway, we might learn about light and three coluors of light as being additive primaries, contrasting with subtractive primaries of pigments, again with the idea that you can make all coloreds with three special colors, invariably red, green, and blue. We might even learn something about the retina having three kinds of cone cells for colur sense, and those were referred to as red, green, and blue, though at the same time we would see diagrams of sensitivity that showed that each cone reacted to a wide range and not a single colur each, and these ranges overlap. A single wavelenght of light could easily stimualate all three cones types. A very recent text might mention that the peak sensitivities of the cones don't even correspond very well to the typical names red green and blue and that long, middle, and short are better labels for wavelength sensitivities.
These indroductions to the topic lack something crucial: the whole idea of primary colers is horseshit. For one thing, any three suffuciently different colors of light or pigment can make similarly varied palette of colours. Specifically red green and blue or red yellow and blue or whatever is not special, just familiar and do have benefit of representation in the rainbow. And the other thing, no three collars can really do the job of giving every possible percieved coalr possible. There's always something imperfect with any choices people have come up twith. If you read up on the history of modelling of coloaurs and the models' "gamuts" you'll get what that's about. Example, CIELAB versus RGB, CYMK, and so on.
These indroductions to the topic lack something crucial: the whole idea of primary colers is horseshit. For one thing, any three suffuciently different colors of light or pigment can make similarly varied palette of colours. Specifically red green and blue or red yellow and blue or whatever is not special, just familiar and do have benefit of representation in the rainbow. And the other thing, no three collars can really do the job of giving every possible percieved coalr possible. There's always something imperfect with any choices people have come up twith. If you read up on the history of modelling of coloaurs and the models' "gamuts" you'll get what that's about. Example, CIELAB versus RGB, CYMK, and so on.
Anonymous B joined in and replied with this 9 years ago, 10 minutes later[^] [v] #753,325
Colours can all be adequately described by adding an "ish" or "y"
Like greeny blueish, or redish purpley
Like greeny blueish, or redish purpley
The Witch Doctor !6W1Zpe4MCM joined in and replied with this 9 years ago, 12 hours later, 12 hours after the original post[^] [v] #753,420
They are called primary colors, because they are not composed of any two or more other colors.Sheila LaBoof (OP) replied with this 9 years ago, 29 minutes later, 12 hours after the original post[^] [v] #753,425
@previous (The Witch Doctor !6W1Zpe4MCM)
however you can say that about any three specific wavelengths of light if you equate colur with wavelength. However the sense of colur is not a simple reaction to wavelength. It's not one-to-one. We know that the way our vision system processes is not conducive to the idea now. Anyway there's some good reading at http://www.handprint.com/HP/WCL/color5.html
however you can say that about any three specific wavelengths of light if you equate colur with wavelength. However the sense of colur is not a simple reaction to wavelength. It's not one-to-one. We know that the way our vision system processes is not conducive to the idea now. Anyway there's some good reading at http://www.handprint.com/HP/WCL/color5.html