# Are There REALLY Infinite Colours?

1. Feb 25, 2010

### FeDeX_LaTeX

Are there really infinite colours? I have been thinking about this for some time now (well, 2 hours), and I'm still not sure.

The boundaries of wavelength for the spectrum of visible light are 380 nm and 750 nm (nanometres). Now, you can get lots of different colours because fractional wavelengths are possible -- but does that mean there are infinite?

If there *were* an infinite number of colours, that would surely mean that the wavelength of a wave could get increasingly and increasingly smaller. This yields the question... is there a limit on the smallest possible wavelength? If I remember correctly, the amplitude of a wave is caused by particles moving from the equilibrium line, right? So does that mean that a wavelength, of, say, 380.000000000000000000000001 nm is possible? Wouldn't this cause an attraction or repulsion between the particles? If this WERE the case, how many different wavelengths are there, and, if this question is answered, how many possible colours are there?

Last edited: Feb 25, 2010
2. Feb 25, 2010

### CRGreathouse

Even if there were only two possible wavelengths of light, it would not follow that there are only two colors. Our eyes and brains perceive colors as a mix; pure red + pure green is seen as brown or yellow, depending on relative intensity.

3. Feb 25, 2010

### Gerenuk

There might be a quantum limit on wavelength difference, but for colour it doesn't matter. Colours are a product of the human eye and colour resolution relies on that process. So I assume the 24bit colours are about all the eye can distinguish. It's probably hard to tell an exact number of colours and it might even depend on the individual person.

PS: it's a bad misconception at school, when they show you a wave as a sine wave and then say "this is what light is". The light wave has an amplitude in the electric and magnetic field and no-where deviates from a perfect line in actual space.

4. Feb 25, 2010

### Hypercubers

In my opinion, the key to this question is clearly what you really mean by a colour. If you mean a colour in the more empirical sense of the word, as in you mean that a colour is a certain wavelength of light emitted within the visible spectrum, then there can clearly not be infinite colours, as there exist quantum limitations on energy levels, which is what makes a wave a wave. However if you are refering to a more human sense of the word, as in our perception of colour, then I think there aren't infinite colours either, as the factors which influence are perception are very numerous, but not infinite. However this second argument can be u to argument.

-Alex

5. Feb 25, 2010

### Gerenuk

I'd like to point out that being precise, colour cannot be transformed one to one with wavelength, since for example purple doesn't appear as a single wavelength.
Single wavelength correspond to special colours on the border of
http://en.wikipedia.org/wiki/File:CIExy1931_fixed.svg

6. Feb 25, 2010

### Hypercubers

you are missing the point, he has to define what he means by colour. Our perception of purple isn't defined by one wavelength, I agree. But the theoretical colour (as in case one of my previous post) IS defined by one wavelength, as it would depend on the ability the machine you are measuring colours with (in this case your eye), has to distinguish wavelengths (the different colours) from one another.

7. Feb 25, 2010

### Gerenuk

Machines can only measure "wavelength" or full "spectra".
The human eye measures "colour".

If by "colour" he means wavelength, he would be wrong. But I also didn't know the connection before I learned about the above diagram.

So as soon as you talk about "colour" you are refering to RGB values, which has no one to one relation to wavelength. Colours are richer than single wavelengths, but contain far less information than a full spectrum. So colours are distinct from wavelengths.

I suppose you also know this. So it's rather a concept about word definitions.
Maybe if he reads about colour perception, he will agree with me.

8. Feb 25, 2010

### Hypercubers

color |ˈkələr| ( Brit. colour)
noun
1 the property possessed by an object of producing different sensations on the eye as a result of the way the object reflects or emits light : the lights flickered and changed color.
• one, or any mixture, of the constituents into which light can be separated in a spectrum or rainbow, sometimes including (loosely) black and white : a rich brown color | a range of bright colors.

... the first is my case 1, the second my case two i think, however i agree with you on the things you said about case 2.

9. Feb 25, 2010

### Gerenuk

OK, it's a minor difference. Yet, case 1 does not mention the wavelength explicitely. I just find it important to not associate colour with wavelength (unless you mean rainbow colour). Just because it was the reason why I had a misconception about colour for quite a while and I'd make sure to never teach a colour and wavelength connection :)
But I guess everyone can keep this information as he wishes. Doesn't make a big difference.

10. Feb 25, 2010

### Hypercubers

Nah, it really doesn't, the underlying concept is esentially the same, you are just calling it one thing and I'm calling it another. The physics is the same, the language is the bad part.
You are a physics teacher? wau

11. Feb 25, 2010

### Gerenuk

Not a real physics teacher, but I give private lesson for university freshers. That is much more challenging than being a physics teacher, because as a physics teacher you just present the information and most people understand it. Giving private lessons you are facing people who are having difficulties, so it's essential to provide a crystal-clear consistent picture. And most of the time I have to press my students to forget misconceptions they picked up from badly explained examples.

12. Feb 25, 2010

### JTE

The OP clearly indicates the meaning of colour intended by talking about the wavelength, i.e. 380.000000000000001nm. So to answer the question you could skip the physiological interpretation and just look at the possible infinity of fractions of wavelengths.

One way of looking at this is to consider where the light comes from and the medium (ick). Light is emitted by various sources that do emit energy in quanta, so maybe frequencies could be quantised, however unless time is also quantised (it's not afaik) then all you'd have to do to alter the wavelength is run at a slightly different speed to see a slightly different wavelength (because time alters when your speed changes).

13. Feb 25, 2010

### Glen Bartusch

There is a quantity called atomic fine structure. This is that narrow band of frequencies the electromagnetic ("light") radiator is radiating.
To determine the amount of color there is (your result will not lead to an "infinite amount of colors"), simply determine all known sources of visible light radiators, then observe their fine structure (done thru a device called the spectrometer). Obtain the exact waveelengths of each spectral line, and voila! You have all colors known to man. I assure you, the spectral emission lines you obtain will not be infinite in number, though.

As for the 380.000000000000001 wavelength; it's possible, I mean, if there's a spectral line that's emitted by some light radiator that emits such a wavelength, then so be it.

14. Feb 25, 2010

### Staff: Mentor

Hot objects emit light in a continuous spectrum - emission lines don't apply here.

15. Feb 25, 2010

### rcgldr

On a side note:

For the PC, most video cards since the late 1990's have a few resolutions in 30bit color mode, although few home or office type digital monitors support this (you need a CRT monitor to use these modes). I'm not sure when or which games support 30 bit color mode, other than it was supported by Tomb Raider Angel of Darkness (released in 2003).

Digital cinema components support 36 bit color. Most high end Graphic workstations support 36 bit color. Some include an additional 12 bit "alpha" value and call it 48 bit "RGBA".

16. Feb 25, 2010

### Glen Bartusch

You're referring to blackbody radiation, which is infrared to near-infrared. I believe the author of the thread made it clear that he was inquiring "infinite colors" of the entire visible spectral range.
Hot objects emit light in a continuous spectrum we cannot see. Emission lines do not apply to the hot object.