> That worked Slime.  Thanks for the help again.  Here is a link to the
> results:
>
> http://www.wwwmwww.com/TRON/walls11.png


Looks great! You might want to move the cones farther out (and scale them
larger) since that one on the left doesn't look like it's very far away.

> I need to raise the AA a bit.  However I still don't understand why the
> light_group idea didn't work.

Honestly, I don't know why it didn't work either. The results you posted
were strange. Sure it's not a max_trace_level issue or something?

> And I'm also confused by "2.3.6.3  Simple
> media examples".  In their first example they have a light source and a
> tranparent container with "rgbt 1".  Isn't that the same as rgbt
> <1,1,1,something> where the something is 1?

Yes, but in your code you had the something = 2.5, which is different. If we
call the transparency value ("something") t, then the calculation for the
lighting of a transparent object is

color = t * (color behind surface) + (1-t) * (color of surface)

When t = 1, you can see that this turns into just the color behind the
surface. But when t > 1, the color of the surface is multiplied by a
negative number, having a subtractive effect. When you set the color of the
surface to black then it's only subtracting zero which does nothing.

 - Slime
 [ http://www.slimeland.com/ ]

Post a reply to this message