I've always been curious why the visible light spectrum appears to "wrap around". Why does the color wheel appear continuous? Is it related to the fact that the frequency range is about an octave?
The color wheel doesn’t really wrap around. We just take a cut of the visible spectrum, and overlay the red and the blue ends. We don’t perceive periodicity in light; near infrared is not one octave lower than violet light. The color range we see is just one that happens to be most useful in a nitrogen/oxygen atmosphere under a Class G sun.
So it happens that we can hear several octaves in sound, via pressure waves, where a note an octave higher or lower is defined as twice or half the frequency, and when the "same note" in different octaves are played together the sounds are full and noticeably harmonious.
The range of the electromagnetic spectrum we see is indeed very close to "an octave" if defined as the doubling of frequency, but it makes no particular sense to consider the harmonies of octaves of visual light when there is only one of them.
This is like "tritones" in music, which are two notes with an interval spanning six semitones. If there was such a thing as a "note wheel" with a circumference of an octave (12 semitones), two notes that form a tritone are opposite each other on the wheel.
---
There are "split-complementary colors".
> Split-complementary colors are like complementary colors, except one of the complements is split into two nearby analogous colors.
> This maintains the tension of complementary colors while simultaneously introducing more visual interest with more variety.
..And "analogous colors", three colors that are next to each other on the color wheel, and a tertiary.
A musical equivalent of these might be like different types of chords.
---
Also, there are "triads", like primary and secondary colors.
> Art education materials commonly use red, yellow, and blue as primary colors, sometimes suggesting that they can mix all colors.
> A secondary color is a color made by mixing of two primary colors in a given color space.
A quirk of human vision. Our "red" cone receptors in our retinas are mostly sensitive to redder hues, but also have a bit of a bump down in the violet part of the spectrum. This shows some graphs that sort of explain it: https://physics.stackexchange.com/questions/433119/why-does-...
We are often told that the color of light is determined by its wavelength. So why do violet/purple appear as if they were smooth interpolations between red (high wavelength) and blue (low wavelength)?
It's a weird artifact of how the brain reconstructs wavelength from the measurements taken by the retina.
The retina has three types of color-sensitive cells (cones). Each one carries a light-absorbing protein that absorbs only certain wavelengths of light. Photopsin I has maximal absorption around 530nm (a sort of yellowish green) and some absorbption towards higher wavelengths, up to 700nm (the red end of the spectrum). Photopsin II has maximal absorption around 530nm as well (so it's green too), but unlike Photopsin I, its absorption drops off rapidly, and does not go toward red. Photopsin III, however, peaks in low wavelengths, around blue.
Even though the peak responses of Photopsin I and II are very similar, the brain can use absorption ratios to detect wavelength.
If your Photopsin I is absorbing some light, but your Photopsin II isn't, the brain deduces that your retina must be being illuminated by red light.
If both Photopsins I and II are absorbing a lot of light, the brain deduces that you're seeing some shade of green (this is why we can tell greens apart better than other colors: we have two receptors for it, but only one for the other colors).
If only Photopsin III is absorbing light, but the other two aren't, your brain deduces that your retina is being illuminated by high frequency blue light.
Other color perceptions arise as combinations of different wavelengths. If all your Photopsins are absorbing light, the brain deduces that you're seeing a combination of all visible frequencies, white light (this is why prisms can split white sunlight, and why we have rainbows). Similarly, If only Photopsins I and III are absorbing, you must be seeing a combination of red and blue light, which is what you see as purple. And that's why purple appears to interpolate smoothly between red and blue.
Okay, but where does violet fit in? Violet isn't purple (a combination of different wavelength light sources), but a pure low wavelength light. Why does _violet_ also appear to interpolate between red and blue? Well, Photopsin I has its absorption maximum around green - but it happens to have a smaller absorption peak in low wavelengths, around 400nm as well. So very low wavelength blue light is absorbed by both Photopsin I AND Photopsin III: and your brain deduces incorrectly that your retina is illuminated by some combination of blue and red. And that's why violet looks similar to purple, and the color wheel appears to wrap around as wavelength decreases.
So we have four different kinds of light receptors in the eye: three "cones" that pick up a relatively narrow range, with peaks at the wavelengths we call "red", "green", and "blue", and one kind of "rod" that picks up a wider range of wavelengths that tends to kick in more in low light conditions, and gives you more bright/dark sensations than color.
The color we call "yellow" is what you get when the red and green cones are reporting fairly equal amounts of stimulus. It generally corresponds to the wavelengths between them. Similarly, "cyan" or "light blue" is the green and blue cones reporting fairly similar amounts of light. That color is also between the peaks of those cone's responses.
And then there is "magenta" or "purple". Which is what you get when the red and blue cones are reporting light. This does not correspond to any particular wavelength of light. But inside your brain it still produces a sensation of color, that has similarities to the red and blue experiences.
This is why some people like to say that "purple is not a real color": it does not correspond to a single particular wavelength of light. But really "color" is just something your brain makes up to classify different combinations of excitement of the rods and cones in your eyes anyway; the reality is that our eyes are only sensitive to a tiny fragment of the electromagnetic spectrum, with an uneven distribution - the "red" and "green" cones have a lot of overlap in their sensitivity, while "blue" barely overlaps either. Have a look at the diagrams in Wikipedia's page on "trichromacy": https://en.wikipedia.org/wiki/Trichromacy
This is also why all our displays are based on red, green, and blue lights: you can fool the brain into sensing a particular color by showing the eyes nothing but the three frequencies the cones are most sensitive to, in various amounts.
I find these visualisations give me a deeper feel for the structure of the music and the skill of the composers.
I'm not sure why the freeware MAM MIDI player [1] stopped getting updates but I loved that tool. I had a lot of fun with it and made a video demonstrating its visualisation types. [2]
Not really. 2 crucial differences: first, it’s the degree of the scale (I, ii, III…) that’s given a colour rather than the note name (C, D, E…), meaning the colouring is independent of the key (or octave) it is written in or transposed into. Secondly, the colours used are designed to be similar where their harmonic ‘meaning’ is similar (according to traditional Western music) and vice versa.
I've often wondered if people with Chromesthesia (who associate sounds with colours and/or colours with sounds) fall into sets of people with similar colour/sound mappings.
"..studies to date have reported that synesthetes and non-synesthetes alike associate high pitched sounds with lighter or brighter colors and low pitched sounds with darker colors, indicating that a common mechanism may underlie those associations in normal adult brains"
From what ive gathered (doing interactive light/sound stuff and chatting with folks at CymaSpace) the issue is that its subjective. Synaesthetes are often frustrated with color to pitch mapping that claims it is "inherent" because the root color (this author picked blue because "it seemed settled") is the creators personal synaesthetic correlation. The issue gets complex when you consider what training set makes each synaesthete build the correlations between colors, moods, sounds, etc.
The article its-self is rad in that the circle of fifths is very much a "relative" thing to begin with so we are actually seeing some pretty "flavorful" results- but the search for the "true color of sound" is tough when light is 430-730 trillion hz and audio is 20-20 thousand hz. The factor that you then are dividing or transposing by is arbitrary.
> the circle of fifths is very much a "relative" thing to begin with
I'm not sure what you mean.. Is this a pun?
The circle of fifths is right on the intersection of music and physics; each perfect fifth is (very close to) a 3:2 ratio on the previous frequency, which is probably why pentatonic scales are found across cultures. I don't know how much less "relative" you can get with something so subjective as music.
I was hoping this would be about using different colours in musical notation, as I think that black-and-white sheet music doesn't use all the visual tools available to modern designers.
It might be hard for musicians to decide what note to play just by looking at an associated colour (especially if they are colour-blind), but there shouldn't be any downside to doing something like "syntax highlighting" for musical notation, where the existing symbols are used, but are coloured based on what type of information they are conveying, e.g. tempo, dynamics, style.
I'd put "syntax highlighing" on the accidentals (maybe red and blue depending on whether it's flat and sharp relative to the scale), and leave everything else alone.
You certainly find this kind of thing in numerous beginner books, but I’m skeptical that it’d be of much use for complex scores, or for someone with mastery of sight-reading. We don’t tend to do the same for text, after all. That said, it is easy enough to miss a dynamic marking and plough through a section at the wrong volume. That one might be worthwhile!
Musical intervals map mathematical ratios to emotional perception (i.e. all minor thirds: feel similar, are { lower, (6/5)*lower } frequency pairs).
For a long time I've wondered if there's anything analogous in the world of color (since both are waves), but have been unsatisfied with the non-answers I've found.
If we sensed sound like we sense colour we'd only hear the amount of bass/mid/treble not the individual pitches.
But if there's such a thing as a non-linear medium for light, then transmitting two wavelengths through it should do interesting things depending on the ratio of the wavelengths?
I love this! I have been trying to figure out sensible colouring for my Janko-layout Chromatone keyboard (https://chromatone.jp/chromatone/index.html, no longer available). All its keys are white. The original Janko keyboards used the same black and white as traditional piano keyboards, like this: https://imgur.com/a/OUAsobE
(A C major scale is shown with bold key outlines in all these pictures.)
But with so many colours, there are very similar ones close together.
So I tried looking at 6-colour layouts, and at putting tactile dots on some keys to both disambiguate which note it is (since each colour is used for two notes), and also allow for playing by feel, so you can better know where you are: https://imgur.com/a/sFzEH0G
It's not clear where the best locations for the dots are (they'll be little rubber feet, and I have found two different shapes for better touch information).
The 12-colour pattern in this article, using the circle of 5ths, avoids the problem of similar colours close together: https://imgur.com/a/0RN7dqY
I would still like some dots somewhere on it for tactile playing. With this layout, each major scale gets its own distinctive colour scheme: https://imgur.com/a/6FhrAcK
This makes accidentals really obvious on the piano roll pictures in this article.
30 comments
[ 2.9 ms ] story [ 81.4 ms ] threadThe range of the electromagnetic spectrum we see is indeed very close to "an octave" if defined as the doubling of frequency, but it makes no particular sense to consider the harmonies of octaves of visual light when there is only one of them.
The first thing that came to mind was "complementary colors", which are two colors opposite each other on the color wheel.
> They create the most contrast and therefore greatest visual tension by virtue of how dissimilar they are.
- From https://en.wikipedia.org/wiki/Harmony_(color)
This is like "tritones" in music, which are two notes with an interval spanning six semitones. If there was such a thing as a "note wheel" with a circumference of an octave (12 semitones), two notes that form a tritone are opposite each other on the wheel.
---
There are "split-complementary colors".
> Split-complementary colors are like complementary colors, except one of the complements is split into two nearby analogous colors.
> This maintains the tension of complementary colors while simultaneously introducing more visual interest with more variety.
..And "analogous colors", three colors that are next to each other on the color wheel, and a tertiary.
A musical equivalent of these might be like different types of chords.
---
Also, there are "triads", like primary and secondary colors.
> Art education materials commonly use red, yellow, and blue as primary colors, sometimes suggesting that they can mix all colors.
> A secondary color is a color made by mixing of two primary colors in a given color space.
However, such spectral sensitivity graphs vary a lot (perhaps because people's vision does, but also they'll be measuring things differently): https://duckduckgo.com/?q=retina+cone+frequency+response&atb...
It's a weird artifact of how the brain reconstructs wavelength from the measurements taken by the retina.
The retina has three types of color-sensitive cells (cones). Each one carries a light-absorbing protein that absorbs only certain wavelengths of light. Photopsin I has maximal absorption around 530nm (a sort of yellowish green) and some absorbption towards higher wavelengths, up to 700nm (the red end of the spectrum). Photopsin II has maximal absorption around 530nm as well (so it's green too), but unlike Photopsin I, its absorption drops off rapidly, and does not go toward red. Photopsin III, however, peaks in low wavelengths, around blue.
Even though the peak responses of Photopsin I and II are very similar, the brain can use absorption ratios to detect wavelength.
If your Photopsin I is absorbing some light, but your Photopsin II isn't, the brain deduces that your retina must be being illuminated by red light.
If both Photopsins I and II are absorbing a lot of light, the brain deduces that you're seeing some shade of green (this is why we can tell greens apart better than other colors: we have two receptors for it, but only one for the other colors).
If only Photopsin III is absorbing light, but the other two aren't, your brain deduces that your retina is being illuminated by high frequency blue light.
Other color perceptions arise as combinations of different wavelengths. If all your Photopsins are absorbing light, the brain deduces that you're seeing a combination of all visible frequencies, white light (this is why prisms can split white sunlight, and why we have rainbows). Similarly, If only Photopsins I and III are absorbing, you must be seeing a combination of red and blue light, which is what you see as purple. And that's why purple appears to interpolate smoothly between red and blue.
Okay, but where does violet fit in? Violet isn't purple (a combination of different wavelength light sources), but a pure low wavelength light. Why does _violet_ also appear to interpolate between red and blue? Well, Photopsin I has its absorption maximum around green - but it happens to have a smaller absorption peak in low wavelengths, around 400nm as well. So very low wavelength blue light is absorbed by both Photopsin I AND Photopsin III: and your brain deduces incorrectly that your retina is illuminated by some combination of blue and red. And that's why violet looks similar to purple, and the color wheel appears to wrap around as wavelength decreases.
The color we call "yellow" is what you get when the red and green cones are reporting fairly equal amounts of stimulus. It generally corresponds to the wavelengths between them. Similarly, "cyan" or "light blue" is the green and blue cones reporting fairly similar amounts of light. That color is also between the peaks of those cone's responses.
And then there is "magenta" or "purple". Which is what you get when the red and blue cones are reporting light. This does not correspond to any particular wavelength of light. But inside your brain it still produces a sensation of color, that has similarities to the red and blue experiences.
This is why some people like to say that "purple is not a real color": it does not correspond to a single particular wavelength of light. But really "color" is just something your brain makes up to classify different combinations of excitement of the rods and cones in your eyes anyway; the reality is that our eyes are only sensitive to a tiny fragment of the electromagnetic spectrum, with an uneven distribution - the "red" and "green" cones have a lot of overlap in their sensitivity, while "blue" barely overlaps either. Have a look at the diagrams in Wikipedia's page on "trichromacy": https://en.wikipedia.org/wiki/Trichromacy
This is also why all our displays are based on red, green, and blue lights: you can fool the brain into sensing a particular color by showing the eyes nothing but the three frequencies the cones are most sensitive to, in various amounts.
I'm not sure why the freeware MAM MIDI player [1] stopped getting updates but I loved that tool. I had a lot of fun with it and made a video demonstrating its visualisation types. [2]
[1] http://www.musanim.com/Player/ [2] https://www.youtube.com/watch?v=cT6Pk0RNRSA
"..studies to date have reported that synesthetes and non-synesthetes alike associate high pitched sounds with lighter or brighter colors and low pitched sounds with darker colors, indicating that a common mechanism may underlie those associations in normal adult brains"
https://en.wikipedia.org/wiki/Chromesthesia
The article its-self is rad in that the circle of fifths is very much a "relative" thing to begin with so we are actually seeing some pretty "flavorful" results- but the search for the "true color of sound" is tough when light is 430-730 trillion hz and audio is 20-20 thousand hz. The factor that you then are dividing or transposing by is arbitrary.
I'm not sure what you mean.. Is this a pun?
The circle of fifths is right on the intersection of music and physics; each perfect fifth is (very close to) a 3:2 ratio on the previous frequency, which is probably why pentatonic scales are found across cultures. I don't know how much less "relative" you can get with something so subjective as music.
It might be hard for musicians to decide what note to play just by looking at an associated colour (especially if they are colour-blind), but there shouldn't be any downside to doing something like "syntax highlighting" for musical notation, where the existing symbols are used, but are coloured based on what type of information they are conveying, e.g. tempo, dynamics, style.
If we sensed sound like we sense colour we'd only hear the amount of bass/mid/treble not the individual pitches.
But if there's such a thing as a non-linear medium for light, then transmitting two wavelengths through it should do interesting things depending on the ratio of the wavelengths?
(A C major scale is shown with bold key outlines in all these pictures.)
But some people have used a 3-colour layout like this: https://imgur.com/a/ya07LGU
Still, that leaves quite a bit of ambiguity, as different notes are the same colour, and it's hard to orient yourself.
So I tried a spectrum-organised 12-colour layout: https://imgur.com/a/Is9sQiA
But with so many colours, there are very similar ones close together.
So I tried looking at 6-colour layouts, and at putting tactile dots on some keys to both disambiguate which note it is (since each colour is used for two notes), and also allow for playing by feel, so you can better know where you are: https://imgur.com/a/sFzEH0G
It's not clear where the best locations for the dots are (they'll be little rubber feet, and I have found two different shapes for better touch information).
The 12-colour pattern in this article, using the circle of 5ths, avoids the problem of similar colours close together: https://imgur.com/a/0RN7dqY
I would still like some dots somewhere on it for tactile playing. With this layout, each major scale gets its own distinctive colour scheme: https://imgur.com/a/6FhrAcK
This makes accidentals really obvious on the piano roll pictures in this article.