Re: 3d highlights on dark colors

[Gerd Moellmann]

Miles Bader <address@hidden> writes:

> Ok, I figured out a reasonable way to do this without using any
> expensive math functions.  The attached patch uses basically the same
> parameters, but instead of adjusting the existing multiplicative factor,
> it calculates an additive factor to add to all the color components,
> using only multiplication and (one) division.
> 
> The resulting effect is *almost* the same as the previous code; however,
> because this approach adds the same value to all color components
> (R,G,B), highlights will tend to be slightly whiter than the reference
> color.  To my eye this looks slightly *better*, as it gives them a
> slightly `specular' appearance...  [Your opinion may vary :-)]

I think it would be better to introduce two variables, say
top-shadow-contrast and bottom-shadow-contrast.  These would replace
the hard coded constants 1.2 and 0.6 that are currently used to make
lighter and darker colors.