I wish there was more consumer interest in spectral imaging/display technology. We miss out on many chromatic experiences by persisting in the bland world of RGB (RYGB, RGCB, RGBW, CMYK, HSL take your pick).
If you don’t know what I’m talking about, here’s a quick primer on how our color vision works.
Our eyes have three types of color receptors: “red” receptors which are most sensitive to yellow-orange, longer wavelength light, and also slightly sensitive to blue-violet, short wavelength light; “green” receptors which are most sensitive to a bright lime green, medium wavelength light; and “blue” receptors, most sensitive to, well, blue.* Here’s a chart to illustrate this:
Your perception of color is simply your brain measuring the relative levels of each type of signal. When all three receptors are at their maximums, you see white. If your red receptor is firing strongly but your green and blue receptors are firing weakly, you’ll see red. If your red and green receptors are both firing strongly, you’ll see yellow. And so on…
Since every color we perceive is simply a mixture of three signals, we can use just three colors of light, red, green and blue, to reproduce a huge range of other colors. This is how TVs and monitors work. Images printed on paper work similarly: each color of ink subtracts either red, green, or blue from the white background.
So what’s the problem?
Since we can perceive the same color from different combinations of light wavelengths, we don’t really get a full perception of the “actual” color of things (the wavelengths of light reflected off, transmitted through, or emitted by objects). Without a fundamental redesign of the visual system, we will never perceive color as an accurate representation of reality.**
On top of that, three-color image technology cuts out a huge amount of the already-limited set of possible colors. Here’s a gamut chart illustrating this. Don’t worry about the details; the important point is the size of the triangle relative to the larger area of color.
The triangle here represents the gamut of most reasonable quality displays; that is, the set of colors it’s capable of representing relative to all the colors we can see. The gamut of ink or paint created with only three primary colors is similarly limited, though the points of the triangle are in different places. Dyes and pigments have an advantage over displays; they can be colored with substances that fall outside the usual gamut. Unfortunately, it’s unfeasible at the moment to create displays with a full range of color (though some manufacturers have added “pure” cyan, magenta or yellow for added fidelity).
And that’s the point of this post. A high-resolution spectral display would open up a whole new world of experience. I’d like to share some of those potential experiences.
Cyan… Is that a kind of pepper?
Poor, neglected cyan. You could almost say we don’t have a word for it; most native English speakers will just call it blue. The cyan you see above is but a pale impostor compared to the true beauty behind turquoise, tropical beaches, and ’80s spandex.
If you have some sunlight and a crystal or prism handy, I encourage you to take a good long look at the color in between green and blue. Now compare that part of the rainbow with the spectrum in the first image above.
What you will (or maybe you won’t?) notice is that the cyan on your screen is washed out—quite literally whiter. The problem? Green contains red. Wait, what?
Cyan is between blue and green. So naturally, to create cyan on the screen, you will mix blue and green together. But take a look at the perception chart again. See where the spectrum is green? The red receptor is at about half its maximum, the green receptor is a little way down from its maximum, and the blue receptor is practically at the bottom. But in the cyan part of the spectrum red is only about 1/5th max, and green and blue are about 1/3 max (blue looks lower because our eyes aren’t as sensitive to it overall; that’s why pure blue looks so dark compared to other colors).
So when you mix green and blue together, the red receptors are firing a lot more strongly than under monochromatic cyan. Since red, green, and blue together make white, adding perceived redness to cyan makes it look more white.
How hot is your pink?
Fluorescent colors have a certain je ne sais quois that just doesn’t come through in print or display. While je might not sais quois, I think je might ai une idée…
Fluorescence is when a material absorbs light of one wavelength, and emits it at another. A photon is absorbed by an atom or molecule, bumping one of its electrons to a higher energy state. A little while while later (understatement), the electron settles back down by releasing another photon. If the electron is excited enough, it might come back down in more than one step. Each step emits a photon with a different wavelength, longer than the original photon’s.
Under sunlight or a black light, fluorescent dyes and pigments become excited by ultraviolet light which we can’t see, and then emit frequencies of light which we can see. From our point of view, it appears as if more light is being reflected than falls on the object in the first place. This effect is exaggerated with a dark room and a black light.
The color of a fluorescent object is a chimera of sorts. The substance might have a normal, reflected color on its own which is is mixed with the very different frequencies of light from fluorescence. I suspect that the aforementioned je ne sais quois is exactly this juxtaposition of colors, creating novel ratios of receptor signals.
Magenta is a good example of this. It’s created with a combination of red and blue, colors on the opposite ends of the spectrum. Your brain tries to treat it as a single color, but it still has that special something that makes it stand out from the rest.
Haha! My socks are glowing!
Black lights are one of the more spectacular failures of imaging and display technologies. On the imaging side, most cameras simply record them as blue. Very bright blue, occasionally white, even. When was the last time you were blinded by a black light? Of course, this can be solved with filters, but that’s too advanced for most people to bother.
Displays could probably do an OK job making black lights themselves look right, but the “black lit” objects wouldn’t be made up of the right frequencies.
It would really be cool if full spectrum displays extended into the near ultraviolet. TV or movie scenes with black lights, sun light, office fluorescents, etc. would have a more natural effect on objects lit by the display.
Sunsets and rainbows
Beautiful things, they are, though I’ve never seen a photo do them justice. Never. Sure, I can imagine what it might have actually looked like, but regardless of film, lens, and photographer quality, the shots are always underwhelming.
I don’t know whether we’ll ever be able to truly recreate the breathtaking majesty of the universe. I certainly hope so, and I believe it’s within the realm of possibilities. What I can say for certain, though, is that we’ll never get anywhere close without looking at the bigger picture of color.
* Some research indicates that we actually have ultraviolet-sensitive receptors as well, but our lenses block out the light before it reaches them.
** I wonder if some sort of rapid cycling through the spectrum would be noticeable to our retinas. Perhaps spinning prism + DLP? Implants? Nanobots? =) I’m sure our visual system would learn to work with that kind of signal. If you build it (signals to the brain) they will come (new neural pathways to interpret them).
