Representing violet with an RGB display

[Warp]

The violet color has a higher frequency than blue. An RGB display can represent at each pixel the frequencies for red, green and blue. Therefore the highest frequency that the display can emit is that of blue. Yet, somehow, RGB displays can represent violet, which has an even higher frequency.

What explains this?

 

[Bandersnatch]

RGB displays simulate violet as a combination of blue and red(so it's actually a kind of purple rather than true violet).
http://en.wikipedia.org/wiki/Violet_(color)#Violet_and_purple

 

[Warp]

RGB displays simulate violet as a combination of blue and red

I know, but it's not what I asked.

 

[DrDu]

The human eye has only three colour receptors for blue, green and red. Violet light stimulates both the red and blue receptors because the red receptor has an absorption maximum both in the red and in the violet region of the spectrum. So the impression generated by a mixture of blue and red light is strictly equivalent to that of violet light.

 

[someGorilla]

I was wondering about the same thing some time ago, or "how can we distinguish violet from blue?"
Apparently the red cones are also sensitive to a narrow band of blueish frequencies. So when you see violet it's exciting your red cones as well.
I don't remember where I read about it and I don't know how reliable that is. But it would explain 1) how you can see violet 2) how you can reproduce violet with blue+red 3) why violet can blend well with both blue and red.

 

[Warp]

The human eye has only three colour receptors for blue, green and red. Violet light stimulates both the red and blue receptors because the red receptor has an absorption maximum both in the red and in the violet region of the spectrum. So the impression generated by a mixture of blue and red light is strictly equivalent to that of violet light.

Red and violet are at almost opposite ends of the visible light spectrum. Do red receptors really have their highest "peaks" at two very different locations in the spectrum?

Looking at the wavelengths, I notice that the shortest wavelength of violet is about exactly half of the longest wavelength of red. In other words, they are about an "octave" apart. Am I guessing correctly that this is not coincidence, but there's some kind of resonance effect in play here?

 

[K^2]

Red and violet are at almost opposite ends of the visible light spectrum. Do red receptors really have their highest "peaks" at two very different locations in the spectrum?

It's not the highest peaks. Peak of red receptor is in red, but it's very broad. So there is enough sensitivity in violet still for both red and blue receptors to be tripped. Green receptor, in contrast, has a very narrow peak, so it receives almost no excitation in violet.

That's why violet looks similar to RGB magenta or purple, but not quite the same. The fact that you cannot reproduce violet has something to do with it being past the range of violet, but it's not the only reason.

You might find this article to be of interest. CIE 1931

 

[DrDu]

Look up "color vision" in wikipedia.
The graphic from the German wikipedia is more informative:
http://de.wikipedia.org/w/index.php?title=Datei:ZapfenAbsorption.png&filetimestamp=20080407230851
Note that the sensitivity curves for the three receptor types have been normalized so that the maxima are equal. However, the number of S receptors is much smaller so that the small secondary maxima of the L are in fact comparable.

 

[Khashishi]

You might also check out http://en.wikipedia.org/wiki/Gamut
Typical monitors only show a small fraction of the full color space visible by the eye.