> Well, closed ("well-behaving") meshes DO have a well defined interior, and
> the "insideness" algorithm is rather trivial (just trace a ray from the
> point in question to some direction (inside_vector), if there are an even
> number of intersections, the point is outside, else inside). It can fail,
> though, if the ray happens to on the exactly same plane as some of the
> triangles being tested. This is what seems to be happening here. The
> inside_vector should be chosen carefully to not be parallel to any of the
> triangles in a mesh.

Oh, so that is what I don't know from never using meshes. I wonder why
inside_vector is not allowed for arbitrary objects (mostly unions of
patch primitives).


-- 
merge{#local i=-11;#while(i<11)#local
i=i+.1;sphere{<i*(i*i*(.05-i*i*(4e-7*i*i+3e-4))-3)10*sin(i)30>.5}#end
pigment{rgbt 1}interior{media{emission x}}hollow}//  Mark Weyer

Post a reply to this message