3 ms·
> The peak sensitivities are closer to green, a different green, and violet. This is fascinating. My first instinct is to say that the peak of those curves are
by tobr 6y ago
> The peak sensitivities are closer to green, a different green, and violet.
This is fascinating. My first instinct is to say that the peak of those curves are not as important as where they discriminate best/where they are most different from the other curves - which does seem to give a more clear red, green and blue. But I’m sure it’s not that simple when you take into account how the brain processes these signals.
Would it be possible to create a good color screen using subpixels emitting light at the peak sensitivity frequencies?
- kps 6y agoNo, because light at the L (‘red’ cone) peak still stimulates M (‘green’) around 75%, and the brain makes that yellow. To see red, you need a lower frequency that stimulates L twice as much as M.
- klodolph 6y agoAny three-primary system is going to cover a triangle of colors. RGB primaries cover a fairly large triangle. If you wanted a larger gamut, you might consider switching to four primaries. You can still encode colors using the RGB space and convert to e.g. RGBY in the display or whatever.
- zokier 6y agoI do note that this is a limitation mostly for physical systems/models. For non-physical color spaces you can make a triangular shape cover far more of the human visible gamut by having the primaries be "imaginary", i.e. not real visible colors. ACES AP0 is one great example of that
- klodolph 6y agoYes, if you had something like a RGBY display, you would probably use a high-gamut RGB space to encode the data and do a transformation prior to display.