On Thu, 24 Feb 2000 23:41:46 -0500, "TonyB"
<ben### [at] panamac-comnet> wrote:

>I need to know what a "bounding slab" is. I also need to know what a "light
>buffer" is. I want to know because they're making the parsing of a scene I'm
>working on *really* slow. They happen right at the end after having read in
>the POV file. I'm trying to make a forest with some Gilles Tran trees, and I
>have 5 different trees which I place in different ways. I have some with a
>recursion (is this the correct word?) of 3, and those go in the background,
>some with 4 that go in the foreground and some with 6 that go right in the
>front. When I put them all together the big slowdown happens. Individually
>and they all parse pretty quickly. I don't get it.

A bounding slab is the smallest box that an object fits in and that
has edges and walls parallel to the world coordinate axes. From what I
remember from FT-POV, a light buffer is similar to a vista buffer in
that it is a 2D projection of bounding boxes, but not on the screen
plane, rather on a very small sphere surrounding a point light source.
Then, when performing shadow-ray testing of a point against an object,
it is first checked if the shadow ray hits the projection (which is a
fast enough operation only involving comparison of a couple of angles)
and only if it does the actual ray-shape intersection test is
performed for the shadow ray. Put simply, there is a major benefit in
render time at the cost of parse time and memory consumption.


Peter Popov
pet### [at] usanet
ICQ: 15002700

Post a reply to this message