I should have mentioned that conversion between VRML and POV-Ray is easy.
They are both ASCII 3d scripting languages with some similarities, and
there are a number of tools that will convert automatically (pov2vrml,
AccuTrans, etc).

VRML continuous mesh avatars can use java to manipulate the mesh vertices or
VRML's own scripting to do it. I have examples of both. At the very least
this would help you work out a way to get POV-Ray to do it.

I have been using VRML for several years and have become a little unhappy
with the development effort there in recent years. I've always thought
POV-Ray had one of the coolest communities and has a very economical
language. It always seemed to me that being able to model and move things
in POV-Ray in realtime would make a lot of sense, so that when you are
happy with the layout of the scene you could render that as the raytraced
final image. It surprises me that this never happened. When I was manager
of a 3d multiuser world for kids a little while back I had the wonderful
experience of watching kids as young as 5 years old building things by
moving them around inside the virtual world. It is a very powerful way of
doing things. Imagine if we could do that with POV-Ray!

(Sorry, that last bit is a little bit off-topic.)

Post a reply to this message