<--back--      home      --next-->

For instance, spatial acuity is much better for luminance than for hue and saturation. So if you want to show details, and do so in a way that conveys the over-all shape of the object, you probably want a colormap with monotonically increasing luminance [first example].

But luminance is also how we determine shape from shading. So, if we want to indicate values on a surface of an object by colormapping it, but we also want to show the shape of the object by shading it, then the colormap probably shouldn't have any luminance variations in itself, as these could be misleading.

The colormap here shows that problem, since you can see that red on this part of the isosurface is much darker than yellow, but this doesn't correspond to variations in shading caused by shape.