CHROMiX

Graphing visible range of colors of the standard eye

I’d LOVE to graph in LAB 3D mode the visible range of colors of the standard eye.

Any ideas?

I don’t care whether it’s thousands of graphed dots that form a shape, or an actual ICC profile.

I found CIERGB.icc on my system, but can’t find much information on what it is. Guessing it’s for the CIE RGB triangular area defined in 1920. That leaves out a lof of saturated greens and blues, in particular.

Found a D50_XYZ.icc from color.org, but it’s large. (ProPhotoRGB large.)

I’m happy to generate a bunch of random numbers representing colors… If there’s a way to test a particular defined color as to whether it’s visible or not.

It isn’t really possible to graph the gamut of human vision as a 3D Lab graph as who can say for sure, where exactly the limits of human vision lies. For example, where would one define the outer limits of our perception, at +/-100 a*/b* or +/-107a*/b* or somewhere else on the Lab scale? Of course, it isn’t possible to see anything approaching the outer limits of the Lab gamut as it includes colours which are able to be defined but for which no physical example could ever possibly exist. Colours like 0 or 1L*, 127a*, 0b* which would be a hue darker than the darkest possible black imaginable yet simultaneously the most incredibly vibrant red, orders of magnitude more intense and saturated than the reddest red any laser can produce.

So unfortunately, I think it isn’t possible to give you a 3D Lab graph of the gamut of human vision. If you do ever happen to find such a graph though, please so let me know.

Aaron, what you are describing is the color space not a gamut within that space. That space is designed to represent all colors visible under a given viewing condition. That colors exist that we can not see is a function of the space, not a region within that space. (As defined by what is visible). What darlingm is looking for is a chromaticity diagram (or an ICC profile) of the human eye.

Interestingly, however, there are colors that simply can not be represented using integer encoding (like ICC profiles); Bruce Lindbloom’s site has a nice diagram (Java required) and an explanation for why the diagram hangs over the edges of the graph boundaries.

Lab Gamut Display

Details about the Lab Gamut Display

I think the closest you could get is one of the CIE XYZ profiles (there are three; one each for D50, D55 and D65) Or maybe the CIELAB D50 profile.

CIE XYZ profiles

I have no idea how these look in ColorThink though… can ColorThink diagram connection spaces?

-Brian

Those XYZ profiles are in the form of ICC profiles, so they can be opened and viewed in ColorThink. Not sure if they are applicable to what’s being discussed here. The primary colors involved are at the extreme edges of Lab space.

The Bruce Lindbloom site like Brian suggested above is the best place to go for information on human perceptual gamut. Unfortunately he does not provide a profile or color list to download to give you something to play with in ColorThink. I have been able to work with this data in the past. ProPhoto comes somewhat close to the shape of human perceptible gamut. You can take ProPhoto and sort of round the edges like you see in the Lindbloom illustration, reduce the red saturation, reduce the green saturation, and note that the yellows and especially the cyan-greens go outside the boundaries of Lab, then you could sort of get an idea of what the human gamut looks like. CIERGB does not follow the visible gamut very closely - leaving out a lot of greens to blues as you say.