>> >Full-blown CSG is very complicated and difficult to implement properly.
>> >However, both your suggestions (truncated cone and torus) can be handled
>> >by a "Revolve" node that takes a 2D shape and revolves it around the Y
>> >axis.
>> (1) thanks for the technique-pointer; I will take a look at it...
>>
>> (2) I know CSG will/would be hard to implement, I just think it is a
>> generally useful modeling tool (witness it's presence in so many
packages)
>> which helps make the expression of ideas by authors easy AND helps keep
the
>> display overhead on the client... where it belongs.
Yes, "revolve," or surface of revolution (SREV) is a good easy thing
to do, if we are not going to do NURBS (which are similar to CSG
constructive solid geom in complexity.)
CSG is useful and wide-spread, but there are other techniques, too. (BREP).
One of the things that CSG, or "solid modelling" in general, is good at is
mass properties (weight, density) and collision detection. However, we
should
focus on what sort of physics based modellers there are available before
deciding what new primitives to add for them.
YON, [email protected], Jan C. Hardenbergh, Oki Advanced Products 508-460-8655
http://www.oki.com/people/jch/ =|= 100 Nickerson Rd. Marlborough, MA 01776
Imagination is more important than knowledge - Albert Einstein (1879-1955)