To get full advantage of meshes, there are two things you have to do:

  1. You should make them as big as possible (ie. put as many triangles as
possible inside one mesh). Putting a few triangles inside a mesh and copying
this mesh does not help anything (in fact, it actually produces a slower
rendering than just using a union of triangles).
  2. You should see if you can generate the scene by copying this big mesh
several times (this saves memory).

  In practice you have to find a compromise between these two things.
You could put *everything* inside one single mesh for maximum speed, but
then it would take a lot of memory (besides being very slow to parse).
  To save memory you should create a mesh as a part of the scene, and then
copy this mesh around the scene. However, as you get exact copies of this
mesh, the result might not be good (ie. if you can see the repeating meshes
in the scene), so you need to model the scene carefully. You could create
several different meshes and make copies of them around the scene in order
to introduce more variation.

-- 
#macro M(A,N,D,L)plane{-z,-9pigment{mandel L*9translate N color_map{[0rgb x]
[1rgb 9]}scale<D,D*3D>*1e3}rotate y*A*8}#end M(-3<1.206434.28623>70,7)M(
-1<.7438.1795>1,20)M(1<.77595.13699>30,20)M(3<.75923.07145>80,99)// - Warp -

Post a reply to this message