<--back--      home      --next-->

This effect is relevent to our colormap generation because of how we rely on interpolation between control points. What's interpolation this case?

Well, its a combination of scaling and component-wise addition between two colors. Once we know gamma, we know how to scale and add luminances of individual phosphors. And, the mechanics of the Cathode Ray Tube itself performs the addition of luminances from the different phosphors.

But we do color interpolation by interpolation of the RGB components, which means that the saturation of the intermediate colors is necessarily less than the saturation of the control point colors, which in turn means that the brightness can vary non-monotonically between two control points, due to the Helmholtz-Kohlrousch effect. Also, if we're interpolating between two control points with very different saturations, we want to make sure that the luminance variation produced by interpolation is consistent with the pattern of grayscale luminance variation that we started with. Basically: because interpolation is based on addition, we want to make sure we're working with quantities that we know how to add, and that's luminance, not brightness.

In all honesty, the difference between the two may be a pretty subtle effect, but its something we should be able to control for. The literature says that luminance is the important quantity to control, so its luminance that I want match.