It's practically meaningless to talk about specific color perception by naming colors, because color names/perception are highly variable between individuals and circumstances. "Purple" is especially problematic, as evidenced in this thread.
The shortest visible wavelengths at ~400nm SHOULD appear deepest blue, stimulating only the S cone in the retina, but 405nm light (which is a common laser wavelength, used for blu-ray) appears sort of a pastel purple. This is because, as magic mentioned upthread, the spectral response of the L cone, which is primarily sensitive to reddish wavelengths, has a small bump near 400nm. L-cone sensitivity is still very low at those wavelengths, but the S cone (sensitive to blueish wavelengths) response also tails off towards those wavelengths, so both responses are significant. It's very hard to precisely quantify the spectral response at such low levels, though, which is why that bump isn't always visible on spectral response plots. Even with a precise response spectrum for the light sensitive molecules themselves, there's a ton of processing overlaid on that chemical response that complicates any attempt to quantify the perceptual results. That processing starts in the retinal ganglia (nerve cells) that aggregate the responses from groups of cones even before those signals leave the eye.
If you want to see what a true deepest blue looks like, you could sit in a room lit only by red LEDs for several minutes and then look at a 405nm LED/laser source. The red light will deplete the L cones (and the M cones somewhat) while leaving the S cones (mostly) unaffected, so when you switch to the 405nm source the response of the S cones will dominate.
On the other hand, most of the time when you see "purple" light you're seeing a combination of red and blue in some ratio, simply because RGB mixing gets you most (not quite all!) of the colors humans can perceive. Since the eye isn't very sensitive to such short wavelengths anyway, using red and blue sources is also much more efficient in terms of perceptual brightness vs radiant power. A CIE Chromaticity diagram (such as attached) is IMHO one of the best ways to visualize perceptual color space as it relates to wavelengths. The X axis represents the ratio of stimulus between L and S cones, and the Y axis the ratio between L and M cones, which cancels out overall brightness from the equation. The colors displayed in the shaded area aren't accurate, since the RGB color space of your monitor can't display all perceivable colors, but they provide a rough idea of how colors map to the space. The curved boundary around the top and sides of the shaded area represents the entire visible spectrum, so anything on that line is a spectral color that can only be produced by a single wavelength. The straight portion of the boundary is the 'line of purples', so anything along that edge can only be produced by a combination of two wavelengths at opposite ends of the visible spectrum. Any color that is inside of those boundaries can only be produced by a combination of two or more wavelengths. There are actually an infinite number of possible spectra that can produce any perceivable color, though, except for the spectral colors which consist of one single pure wavelength (even lasers aren't quite that pure in general, but they can get pretty close for practical purposes), and the "purples" that are combinations of the very longest and very shortest wavelengths we can perceive. Pick any two points in the shaded area (or on its boundary) and a line drawn between them will cross all of the possible colors that can be made from that combination (its gamut). Pick any three points and a triangle drawn between them contains all the colors that combination can produce. Likewise for any number of points. Note that there is no combination of three spectral colors that can enclose all of the perceivable colors, showing that RGB mixing isn't always adequate -- you can expand the gamut by adding more and more spectral colors, but in fact no finite number of component wavelengths can be combined to reproduce the entire perceivable color space. (To work around this, some color spaces use 'imaginary' colors that lie outside of the shaded area on the chromaticity diagram as their primaries.)
If we mean an actual "raw" laser emitter (like a laser diode or tube), then they are (for practical purposes) monochromatic spectral colors.
If we mean a laser product, like a laser lightshow, then it's definitely possible to have multi-wavelength laser beams, which are simply made by overlaying RGB laser beams, modulated as needed, from multiple laser emitters. That's how we get full-color laser shows that include white.
Actually, a lot of tube (gas/ion) lasers are capable of lasing at multiple wavelengths. Before DPSS lasers, which came before direct injection diode lasers, there were "white light" gas lasers, typically using argon to produce blues/cyans/greens, and krypton for yellows/reds, and these were commonly used in laser light shows for that reason. As a result, the laser light show industry adopted typical Ar/Kr wavelengths when standardizing color control channels, so you still see optional deep blue/yellow/cyan in addition to r/g/b channels on some control systems, although for the most part they go unused these days. HeNe lasers can also lase at multiple wavelengths in the red-green range, although it's more unusual to see them lasing at multiple wavelengths at the same time. What you actually get out of the laser will depend on how it's set up, and can be controlled to some extent by how the laser cavity is aligned and what additional components are used in the cavity. For some applications you really only want one wavelength, so it's common for lasers to be set up to suppress all but the target wavelength.
(According to a friend with decades of experience in the early days of laser light shows, they used to have multiple sets of cavity mirrors for their ion lasers that could be swapped out depending on what colors were most important to a particular show. Very slight differences in the shapes of the mirrors would cause have different effects on the path lengths through the cavity, which could shift the distribution of power between the different longitudinal modes. So if you had a show that needed a lot of red, you would swap in the red-heavy mirror set -- of course then you have to do a from-scratch alignment of the cavity, which is tricky enough on a small single-line HeNe, let alone on a big multi-line, water-cooled 4ft glass tube that pulls a few kilowatts from the wall!)
Home
Help
Search
Login
Register
