What is the Definition of Color According to the CIE?

[bobie]

What is really "green"?

I read many articles but still I do not know what are colours like green:
Is it
-a subjective quality that is made up in human brain, or
-the result of sum of yellow and blue, or
-the presence of radiations of two frequencies in the same source, or just
-the quality of a radiation of a single frequency, as wiki says, in the range between 570 and 590 nm?

Thanks for your help

 

[sophiecentaur]

This is a frequent question. Your list of ideas all have some relevance to Colour.
The concept of Colour is entirely subjective and the sensation of colour involves the outputs from the three sets of receptors in your eye and what your brain makes of them. (So your first statement is correct). The receptors are an extremely crude form of spectrometer; the three types have a very broad response to the longer wavelengths (encompassing all the red/orange/ yellow spectral colours), the middle wavelengths ( yellow / green etc. spectral colours) and short wavelengths (greens and blues). There is a massive overlap in the responses so you can see all wavelengths in the visible range with just three receptors. If you could just 'see' narrow bands of wavelengths (pure spectral red, pure spectral green, pure spectral blue) there would be many wavelengths that you couldn't see - pretty useless! Your brain looks at the three signals it gets from the receptors and remembers the relative and absolute levels for any particular combination. It can then recognise when it sees a similar combination and call that (privately) the same 'colour'. Of course, two different people will usually agree (almost) on when two different objects 'look the same' and they use a word to describe this - these are the names of the colours we use. There can be a big spread in people's memory and assessment of colours. 'Matches' between colours of fabrics can be the source of many arguments in the clothes shop and the art of fabric dying is big business. I would say that colour is an essentially private sensation. Many 'colour blind' people get along fine and their disability hardly shows until they take a colour vision test.
It is well known that you can produce a given 'colour' with many different combinations of wavelengths. When you add the right amounts of spectral red light and spectral green light, you can get the same colour sensation as with spectral yellow. Note the difference between that and your statement about green - but what you say is not totally incorrect; it's just that most colour mixing in 'displays' uses additive primaries of R G and B. You are referring to subtractive mixing, using pigments. (Google additive and subtractive colour mixing)

http://photo.net/learn/optics/edscott/vis00020.htm shows how all the colours we see can be represented as co ordinates on a plane diagram. This link shows how three primaries can be added in weighted amounts to produce a 'metemeric' match with any colour within the triangle of the three primaries. (Note the huge number of colours that a typical display cannot produce!)

-the quality of a radiation of a single frequency, as wiki says, in the range between 570 and 590 nm?

I would say it is the 'sensation' rather than the quality. It would be wrong to equate wavelength exclusively with the sensation of colour. Colour involves three variables where wavelength involves just one.

 

[bobie]

I would say it is the 'sensation' rather than the quality. It would be wrong to equate wavelength exclusively with the sensation of colour. Colour involves three variables where wavelength involves just one.

Thanks for your deep analysis and precious links, but ,please tell me: if we have a beam of photons all of 580 nm wavelength, what colour do we see?

 

[Nugatory]

Thanks for your deep analysis and precious links, but ,please tell me: if we have a beam of photons all of 580 nm wavelength, what colour do we see?

A friend of mine sees the same color as a carrot (yes, really - I'm not making that up!) while I see the same color as the leaves of a healthy plant, as will most of the people reading this thread.

Because only one in a few million people have his genetic mutation, we say that he has rare form of color-blindness and the light is "really" green... But we're being sloppy, because the light is really 580 nm and "green" is a label that we've attached to that wavelength.

 

[sophiecentaur]

You will call the colour you see 'green'. It will be similar to (and nearly match with) a lot of other 'greens', like grass, the 'green' part of the rainbow (which is very desaturated by the addition of the blue{isn} sky) and the 'green that you can mix with blue and yellow paints.
The word "really" doesn't come into it - as with most of Science, aamof. Going down that route is more philosophical than scientific.

 

[sophiecentaur]

A friend of mine sees the same color as a carrot (yes, really - I'm not making that up!) while I see the same color as the leaves of a healthy plant, as will most of the people reading this thread.

Because only one in a few million people have his genetic mutation, we say that he has rare form of color-blindness and the light is "really" green... But we're being sloppy, because the light is really 580 nm and "green" is a label that we've attached to that wavelength.

. . . . . and a lot of other green things.

 

[Drakkith]

My dad is red-green colorblind, meaning that he cannot separate wavelengths in that area of the spectrum into colors as well as the average person. Its so bad that he cannot tell the difference between green and red stoplights. They are both the same color to him. This is why color is termed "subjective" and why we use wavelength and frequency in science since they are direct measurements and correspond to what light waves actually do.

 

[AlephZero]

What is "really" green depends on language (or culture) as well as biology. If you spoke a language that only had color words for white, red, and black, you might not even have the concept of "green". Japanese and Mandarin use "blue" and "green" differently from English. For example traffic lights are the objectively the same colors, but are described as blue, not green. And one language describes green as "grass yellow."
http://en.wikipedia.org/wiki/Color_term

 

[D H]

Thanks for your deep analysis and precious links, but ,please tell me: if we have a beam of photons all of 580 nm wavelength, what colour do we see?

I'm surprised no one has caught this yet. You would see yellow, not green. 580 nm is smack dab in the middle of the spectral yellows (570 nm to 590 nm). Green (spectral green) is light between 520 and 570 nm.

@bobie, notice that I said spectral yellow and spectral green. Imagine a screen that is white except for a colored circle in the middle. Imagine that that colored circle emits light of one frequency only, and that that single frequency can be made to slowly vary from the deepest red to the deepest violet. You will see the colors of the rainbow as the frequency gradually changes from red to orange to yellow to green to blue and finally to violet. Those are the spectral colors.

Most people have three kinds of cones in their eyes. The behavior of those cones is dictated by chemistry and physics. Some cones respond strongest to reddish light, others to greenish light, and yet others to blue/violet. The response curves of these three kinds of cones overlap. Both the red and green cones will respond fairly strongly to that 580 nm laser light; the blue cones will not respond much at all. We see the different colors of the rainbow because different frequencies elicit different combinations of responses from those three kinds of cones.

Is that all there is to color? Absolutely not. There are very, very few pure light sources in nature. Almost all of the colored objects we see have a broad spectrum of light coming from them into our eyes. Going back to that colored circle on white background, let's change that central circle so that emits multiple frequencies of light at once. Two key results show up.

1. Metamers.
Tune the frequencies produced just right and you'll get a response from the cones that is indistinguishable from a pure spectral color. The manufacturers of computer screens take full advantage of this effect. Your computer screen does not produce yellow light. So how does it produce this?

If you aren't colorblind you should see that central block as yellow, even though there is no yellow light coming from your screen. The light coming into your eyes is a combination of red and green light that elicits the same response as would spectral yellow. So you see yellow.

2. Non-spectral colors.
We see more colors than just the colors of the rainbow. That white background, for example. There is no such thing as spectral white. White is a non-spectral color. Another even more interesting one is purple. Our eyes do something rather funky with the spectrum. The full spectrum of electromagnetic radiation goes from radio to gamma. Visible light is but a tiny, tiny part of that full spectrum. In terms of the spectrum, there is no such thing as the color wheel. That violet turns into purple, then magenta, and then red: That's something our vision system does that is distinct from frequency.That's just biophysics. There's a lot more to color perception than this low level, biophysical description of how you see color.

 

[sophiecentaur]

I'm surprised no one has caught this yet. You would see yellow, not green. 580 nm is smack dab in the middle of the spectral yellows (570 nm to 590 nm). Green (spectral green) is light between 520 and 570 nm.

@bobie, notice that I said spectral yellow and spectral green. Imagine a screen that is white except for a colored circle in the middle. Imagine that that colored circle emits light of one frequency only, and that that single frequency can be made to slowly vary from the deepest red to the deepest violet. You will see the colors of the rainbow as the frequency gradually changes from red to orange to yellow to green to blue and finally to violet. Those are the spectral colors.

Most people have three kinds of cones in their eyes. The behavior of those cones is dictated by chemistry and physics. Some cones respond strongest to reddish light, others to greenish light, and yet others to blue/violet. The response curves of these three kinds of cones overlap. Both the red and green cones will respond fairly strongly to that 580 nm laser light; the blue cones will not respond much at all. We see the different colors of the rainbow because different frequencies elicit different combinations of responses from those three kinds of cones.

Is that all there is to color? Absolutely not. There are very, very few pure light sources in nature. Almost all of the colored objects we see have a broad spectrum of light coming from them into our eyes. Going back to that colored circle on white background, let's change that central circle so that emits multiple frequencies of light at once. Two key results show up.

1. Metamers.
Tune the frequencies produced just right and you'll get a response from the cones that is indistinguishable from a pure spectral color. The manufacturers of computer screens take full advantage of this effect. Your computer screen does not produce yellow light. So how does it produce this?

If you aren't colorblind you should see that central block as yellow, even though there is no yellow light coming from your screen. The light coming into your eyes is a combination of red and green light that elicits the same response as would spectral yellow. So you see yellow.

2. Non-spectral colors.
We see more colors than just the colors of the rainbow. That white background, for example. There is no such thing as spectral white. White is a non-spectral color. Another even more interesting one is purple. Our eyes do something rather funky with the spectrum. The full spectrum of electromagnetic radiation goes from radio to gamma. Visible light is but a tiny, tiny part of that full spectrum. In terms of the spectrum, there is no such thing as the color wheel. That violet turns into purple, then magenta, and then red: That's something our vision system does that is distinct from frequency.


That's just biophysics. There's a lot more to color perception than this low level, biophysical description of how you see color.

It's important to note that the colours of the rainbow are not 'spectral' colours. When you look at a rainbow you see spectral colours PLUS all the pale blue of the sky / white of the clouds, behind. The rainbow is not pure. (If we want to be accurate in our discussion of such things).
The yellow you can see on a monitor, in the centre of the square will not be a match for spectral yellow because spectral yellow does not lie on a line between the display primaries. They do a fair job but it's not exact - spectral yellow is an 'illegal colour' and outside the gamut of the primaries.

I would say that there are, in fact, 'no' pure light sources in nature, unless you include some of the gas discharges that you can see in the Aurora. It's only since Newton that anyone has actually seen pure spectral colours and we certainly didn't evolve to take them into account in our assessment of colours.

I would be happy if people would just stop equating colours to wavelength - that would be a good start.

 

[sophiecentaur]

I'm surprised no one has caught this yet. You would see yellow, not green. 580 nm is smack dab in the middle of the spectral yellows (570 nm to 590 nm). Green (spectral green) is light between 520 and 570 nm.

@bobie, notice that I said spectral yellow and spectral green. Imagine a screen that is white except for a colored circle in the middle. Imagine that that colored circle emits light of one frequency only, and that that single frequency can be made to slowly vary from the deepest red to the deepest violet. You will see the colors of the rainbow as the frequency gradually changes from red to orange to yellow to green to blue and finally to violet. Those are the spectral colors.

Most people have three kinds of cones in their eyes. The behavior of those cones is dictated by chemistry and physics. Some cones respond strongest to reddish light, others to greenish light, and yet others to blue/violet. The response curves of these three kinds of cones overlap. Both the red and green cones will respond fairly strongly to that 580 nm laser light; the blue cones will not respond much at all. We see the different colors of the rainbow because different frequencies elicit different combinations of responses from those three kinds of cones.

Is that all there is to color? Absolutely not. There are very, very few pure light sources in nature. Almost all of the colored objects we see have a broad spectrum of light coming from them into our eyes. Going back to that colored circle on white background, let's change that central circle so that emits multiple frequencies of light at once. Two key results show up.

1. Metamers.
Tune the frequencies produced just right and you'll get a response from the cones that is indistinguishable from a pure spectral color. The manufacturers of computer screens take full advantage of this effect. Your computer screen does not produce yellow light. So how does it produce this?

If you aren't colorblind you should see that central block as yellow, even though there is no yellow light coming from your screen. The light coming into your eyes is a combination of red and green light that elicits the same response as would spectral yellow. So you see yellow.

2. Non-spectral colors.
We see more colors than just the colors of the rainbow. That white background, for example. There is no such thing as spectral white. White is a non-spectral color. Another even more interesting one is purple. Our eyes do something rather funky with the spectrum. The full spectrum of electromagnetic radiation goes from radio to gamma. Visible light is but a tiny, tiny part of that full spectrum. In terms of the spectrum, there is no such thing as the color wheel. That violet turns into purple, then magenta, and then red: That's something our vision system does that is distinct from frequency.


That's just biophysics. There's a lot more to color perception than this low level, biophysical description of how you see color.

OH yes - and well spotted about the wavelength / colour thing at the beginning. I didn't bother to check!

 

[D H]

It's important to note that the colours of the rainbow are not 'spectral' colours. When you look at a rainbow you see spectral colours PLUS all the pale blue of the sky / white of the clouds, behind. The rainbow is not pure. (If we want to be accurate in our discussion of such things).

You took what I wrote far too literally. The best way to see the "colors of the rainbow", the spectral colors, is to do what Newton did: Sit in a dark room, let a small beam of light into hit a prism, and see the results on a sheet of white paper.

 

[Nugatory]

OH yes - and well spotted about the wavelength / colour thing at the beginning. I didn't bother to check!

Nor me

 

[bahamagreen]

Yellow is a little different from the other colors; the retina has ten neural functional layers of feature detection processing (there are ten layers of the cortex, too - almost as if the surface of the brain made an extension through the optic nerve to the back of the eye), and the processing layer in the retina that extracts yellow is further in the processing sequence than that used for sourcing the other colors.

The extraction of yellow is a little "more synthetic" than the other colors; may be why it is used so much for food advertising, warning and danger signs, other things meant to "catch our eye".

 

[Adderall]

Some background reading in order for people to tease apart the physical/physiological/mental aspects of color: Stanford Encyclopedia of Philosophy: Color

bobie: whether or not one believe colors exist outside of perception changes how one would answer your question. I like Descartes formulation, "It is clear then that when we say we perceive colors in objects, it is really just the same as saying that we perceived in objects something as to whose nature we are ignorant but which produces in us a very clear and vivid sensation, what we call the sensation of color." I think that any statements about color in the context of physical sciences are really statements about photons, and statements about color in the context of perception are actually statements about the contents of particular mental states.

 

[sophiecentaur]

You took what I wrote far too literally. The best way to see the "colors of the rainbow", the spectral colors, is to do what Newton did: Sit in a dark room, let a small beam of light into hit a prism, and see the results on a sheet of white paper.

I was just reading to back to you as it would be read by the majority of people, who believe that the rainbow consists of pure colours. It is no surprise that an ancient expression like "colours of the rainbow" is mis-applied so often. Perhaps 'Hues of the Rainbow' would be an alternative.
Doing what Newton did, initially, will blow the socks off anyone who is not aware of real spectral colours.

 

[bobie]

Thanks for your wonderful response. I made a mistake in the OP, I meant sort of 540 nm.
My question was not philosophical and did not refer to sensations, btw, I thought that the debate between Newton and Leibniz had been settled
I always read that green is not a basic colour, but, from what you say that is wrong: if a beam of photons of 500 or 550 nm reaches a human eye most people will recognize the frequency or "colour" emitted by any leaf.
If that is right, I do not understand the need for further speculations.

Do you know of any site where I can digit a frequency and get a "colour"?
Thanks for your invaluable help!

 

[sophiecentaur]

Thanks for your wonderful response. I made a mistake in the OP, I meant sort of 540 nm.
My question was not philosophical and did not refer to sensations, btw, I thought that the debate between Newton and Leibniz had been settled
I always read that green is not a basic colour, but, from what you say that is wrong: if a beam of photons of 500 or 550 nm reaches a human eye most people will recognize the frequency or "colour" emitted by any leaf.
If that is right, I do not understand the need further speculations.

Do you know of any site where I can digit a frequency and get its "colour"?
Thanks for your invaluable help!

Firstly, you should explain what you mean by a "basic colour". No colour is any more basic than any other - your eye responds in the same way to all colours, and it's not a measuring instrument. There is no colour that will elicit a single non zero value from just one of your receptors.

There are Primaries, which are colours that are chosen so that you can get a good range of colours by mixing them. Paints and pigments use subtractive mixing and the best primaries are Yellow, Cyan and Magenta (look at the inks in your printer - and look up Subtractive Mixing of Colours). For adding light sources (as with TV displays), the best primaries are basically Red Green and Blue - but none of the primaries used in displays are monochromatic (spectral) because you just can't get enough light out of such sources.

A leaf reflects light of many wavelengths. There is no single wavelength that can represent the perceived colour of a leaf (or any normal non-luminous object for that matter).

You should try to read this Hyperphysics link before you pursue this any further.

I have not managed to find a reference to any calculator that will give CIE co ordinates of the spectra colours but it would not take you long to assemble a look-up table from the figure on the Hyperphysics link. Anything you get would be quite accurate enough for any application you may have. How would you want to use the numbers you obtain?

 

[bobie]

You should try to read this Hyperphysics link before you pursue this any further.
... How would you want to use the numbers you obtain?

Thanks, sophiecentaur, that link answered all my questions, cleared all doubts, that is all I was looking for: "green" is as basic as any other colour.
But I think I understand, also, that human mind makes a sort of average between frequencies so that a frequency of yellow and one of blue are perceived as intermediate " green"
Is that correct?: whilst sound frequencies merge into a sum and reache the ear as a single curve, two light frequencies travel side by side and are blended in the mind.

 

[sophiecentaur]

Thanks, sophiecentaur, that link answered all my questions, cleared all doubts, that is all I was looking for: "green" is as basic as any other colour.
But I think I understand, also, that human mind makes a sort of average between frequencies so that a frequency of yellow and one of blue are perceived as intermediate " green"
Is that correct?: whilst to sound frequencies merge into a sum, two light frequencies ar blended in the mind.

As written earlier, the "human mind" has only three signals to work on - the outputs from the three sensors. You really must look at more web pages about tristimulus colour vision and they you will get a better grasp of the 'mechanics' of it.

But, when you get down to it, our appreciation of the light that enters our eyes and the way we 'categorise' the sensation is much the same as we categorise tastes, music, the feel of things etc. We have a very 'internal' memory / model of these things and we communicate a very restricted version to other people.
It always amazes me how people treat 'Colour' as a special part of experience. Would we ever ask the question about how "really soft" is a cat's fur? Humans are strange, complex things.

 

[Pythagorean]

Well, in neuroscience, we definitely ask why things feel soft or hard (the answer has a lot to with tactile sensors in our skin, of course).

The human mind doesn't only have three signals to work with. I understand your point, and I agree to a point, but as DH was implying, these three isolated sensors are too reductionist for the level we perceive at. At the level of perception, it is large, spatially distributed ensembles of these three sensors.

To illustrate this, simply consider a newspaper. If you look very closely at the newspaper, it is made up of "pure" colored dots, but humans perceive a collage of dots as a "color". It's not just one set of three, it's a spatially distributed map of millions! Furthermore, these perception are not processesd and interpreted in a Markovian sense... memory also matters.

There are several illusions that illustrate these two points. The importance of spatial context is illustrated in this illusion:

and here is an illusion that presents the temporal effects of brain processing (stare at the dot and you should see "imaginary" or "ghost" colors when there's only white shapes. note also that when the shapes change orientation, you "imagine" a different color!):

In the above illusion, our brain is still seeing two entirely different colors but there is only "white" entering our retina in both cases. The discrepancy is based on two different histories (one for each shape) as "remembered" by our visual processing system.

 

[bobie]

our appreciation of the light that enters our eyes and the way we 'categorise' the sensation is much the same as we categorise tastes, music, the feel of things etc. .

Is there an explanation why the two extremes of the visual spectrum look similar, so similar that colours are often presented in a circle?

 

[Pythagorean]

They don't really look similar. When people do that, they use a mixture of the two colors to join them, but you could do that with any two colors that aren't an opponent process of each other (like there's no greenish-red or bluish-yellow colors).

 

[dauto]

Is there an explanation why the two extremes of the visual spectrum look similar, so similar that colours are often presented in a circle?

The colors are presented in a circle because purple is a mixture a red and blue (I've seen some people call it pink, but I think of it as purple). Note that I'm not talking about violet. That purple color isn't in the spectrum. No single light frequency will produce that color, so the spectrum by itself will NOT close into itself. There is no purple in the rainbow.

 

[dauto]

But I think I understand, also, that human mind makes a sort of average between frequencies so that a frequency of yellow and one of blue are perceived as intermediate " green"

It's not as simple as a plain average. Mixing blue with red gives you purple, which is not even in the spectrum at all.

 

[Pythagorean]

Yellow and blue do not make green with respect to human visual processing. Yellow/blue are determined by the same opponent process, so they never mix in perception. Green originates from another opponent process.

http://blog.asmartbear.com/color-wheels.html

http://en.wikipedia.org/wiki/Opponent_process

 

[D H]

But I think I understand, also, that human mind makes a sort of average between frequencies so that a frequency of yellow and one of blue are perceived as intermediate " green"
Is that correct?:

No, that is not correct. Going back to my magical screen, suppose we make it emit spectral yellow and spectral blue in some mix of intensities. (For example, suppose the light is formed by a bunch of intermingled yellow and blue LEDs, packed so close together that you can't see the individual LEDs.) You'll never see green. You'll see a muted yellow, or a muted blue, or perhaps even white if you get the intensities just right.

whilst sound frequencies merge into a sum and reache the ear as a single curve, two light frequencies travel side by side and are blended in the mind.

Our ears and eyes are very different beasts. Our eyes only have three kinds of color sensors, but we have lots of them spread over each retina. Our ears have several thousands of different sound sensors, each attuned to different sets of frequencies, but only one of each per ear. Our ability to see different colors is incredibly poor compared to our ability to hear different sounds. On the other hand, our ability to hear where a sound is coming from is poor compared to our ability to see the source of that sound.

Is there an explanation why the two extremes of the visual spectrum look similar, so similar that colours are often presented in a circle?

That's a trick of our mind. Purples, magentas, etc. are a special kind of non-spectral color. Additively mixing yellow and blue, or red and cyan, results in white. Additively mixing blue and red results in some purplish color.

 

[sophiecentaur]

Well, in neuroscience, we definitely ask why things feel soft or hard (the answer has a lot to with tactile sensors in our skin, of course).

The human mind doesn't only have three signals to work with. I understand your point, and I agree to a point, but as DH was implying, these three isolated sensors are too reductionist for the level we perceive at. At the level of perception, it is large, spatially distributed ensembles of these three sensors.

.

In the context of my post, Neuroscientists do not qualify as "people" (haha). They are, I would hope, much more informed than your average Joe and would never be so superficial. You are definitely right about the fact that our vision works in a much more complex way than the tristimulus model and that context is very important. This just adds to my argument that Colour is treated in an over-familiar and erroneous way by the general public. At least, these days, anyone with access to a computer can look at RGB values of the different colours they see on the screen, which imo is potentially a great help towards understanding some basics.

 

[bobie]

Our eyes only have three kinds of color sensors, but we have lots of them spread over each retina. Our ears have several thousands of different sound sensors, each attuned to different sets of frequencies, but only one of each per ear. Our ability to see different colors is incredibly poor compared to our ability to hear different sounds.

That's a trick of our mind. Purples, magentas, etc. are a special kind of non-spectral color. Additively mixing yellow and blue, or red and cyan, results in white. Additively mixing blue and red results in some purplish color.

Three kinds of sensor doesn't necessarily mean 3 sensors versus several thousands.
I asked after a site where I could increase frequency ad lib , as I would like to check my wild guess that my eyes are able to distinguish thousands of different frequencies.
On the other hands, we distingish thousands of single sound frequencies but we experience sensations: chords and group of notes that are not, so to speak, "spectral"

 

[sophiecentaur]

I would say that your brain is not particularly 'interested' in frequencies. It is just interested in the Colour of the light from a given direction. That colour will consist of a combination of many frequencies. We did not evolve with a need for our eyes to be spectrometers and they are very poor spectrometers, in fact. I cannot think of a survival situation in which the knowledge of the frequency of a single spectral colour would be useful. With the resources available on our retina, we get more important, directional information.

 

[FactChecker]

In physics, green is light in the wavelength range 520 to 570 nm. But the human eye does not have sensors for each frequency. It senses high, medium, and low frequencies to give us all the perceived colors. It is a common misconception that light of any color can be made by combining 3 colors like red, green, and blue, but that is just how we fool the eye. In physics, you cannot combine colors to obtain another color.

 

[sophiecentaur]

In physics, green is light in the wavelength range 520 to 570 nm. But the human eye does not have sensors for each frequency. It senses high, medium, and low frequencies to give us all the perceived colors. It is a common misconception that light of any color can be made by combining 3 colors like red, green, and blue, but that is just how we fool the eye. In physics, you cannot combine colors to obtain another color.

'Colour' is not Physics. Green is not a wavelength. You can definitely not produce spectral green on a TV screen. You have to get these things in perspective and not over simplify. (I am not disagreeing with your post - I am expanding on it.)

 

[bobie]

'Colour' is not Physics. Green is not a wavelength.

That is true, but also "note" ,"music" is not Physics. I don't see why we cannot apply the same standard.
"A 7" has lots of frequencies around 3520, and it is recognized as such, independently of the tuning of the instruments. The difference with visible light is that we distinguish 12 "notes" and only 6 "colours".
Stretching our fantasy, we might talk of orange as "red sharp" or "yellow flat", or say that violet resembles red because it is an "octave above" red
Why not?

 

[sophiecentaur]

Music and colour are human experiences. We can associate them with physical quantities but there is a limit to how relevant it is. Obviously, we use technology to provide our senses with stimulae but there isn't a causal path from perception to stimulus ( it only works in the other direction).
I put my thoughts into threads like this in an attempt to get people to stop making invalid inferences about the relationship between physics and the psychology of senses.

 

[Pythagorean]

Music is a particularly interesting case. The math doesn't add up if you approach a note two different ways unless you distribute the error between all the notes (a technique called equal temperament tuning). If you try to force music theory on the physics (by "just" tuning), you'll end up with that error, called a syntonic comma.

http://en.wikipedia.org/wiki/Syntonic_comma