I'm convinced that preparing Global Illumination models should develop from mapping the sides of dice to the surface of the sphere.

Imacen - Thanks for the link!

The first hurdle I see here is learning of how dice are created. My first hope is that a die can be generated with any number of sides greater than 4. I just posted a question about this over at the site Imacken linked to earlyer, at http://www.applied-synergetics.com/ashp/forum/viewtopic.php?p=2#2.

The second hurdle is at mapping the center of each side of each die to a coordinate on the surface of the Light Control Sphere.

While we're waiting for the more advanced formulae input of hurdle 1, let's start in on the second hurdle by taking on the 4 sided die as a warmup. In the above pic. are the images of the usual D&D(Dungeons and Dragons) dice I've been mentioning. The "4 Sided" is the one in the middle. Side 1 is easy, we can plot it at 0,0,0. But then where the other 3 lights go is questionable. My first instinct is to divide the 180 degrees, from the top of the sphere to the bottom, by the number of sides we have in the shape(180 / 4 = 45, 180 - 45 = 135.) Thus we could use the latitudinal line at 135 by dividing it's 360 degrees, of Z rotation, by the number of sides left to plot (360 / 3 = 1 light every 120 degrees.) But then I think ahead to the model of the 6 sided. The 4 planes of its "Sides" would make contact with the surface of the sphere at the equator(The 90th latitude). This makes me ask what is divided by 4 to get 90. The answer of 360, transfered back to the model of the 4 Sided, would mean we should have used the 120th line of latitude(360 / the 3 three sides of the 4 Sided = the 120 latitude). I tend to lean toward this second theory, as with it we come up with 120 degrees twice, though I'm not proof posotive that it is perfect. And would actually like a better formula, as all die shapes wo'nt be as easy as finding just one line of latitude aside from 0,0,0...

= )