In article <3D77E242.8F775B59@computermuseum.fh-kiel.de>,


> I don't think the root solver is the problem, because it
> works perfectly with, for example, standard spheres, but not
> with the *same* spheres modeled by a poly.

The sphere primitive doesn't use the poly root solver, it is completely 
unrelated.


> I have continued to experiment with polys and I'm irritated
> even more -- just look at my intersection of three orthogonal
> semitransparent poly-planes. Even these  most simple  polys
> create cutting-faces that are too large (by  exactly  0.0001 --
> compare with the white sphere: the same value as with my drop).

You are depending on an object that is already prone to precision 
problems to be accurate at scales that would stress many primitives. It 
happens that 10e-4 is a tolerance value used for several things in the 
poly code, including the root tolerance. Just use larger values or 
different shapes.

-- 
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/

Post a reply to this message