Ilia Guzei wrote:


> Shay:
> This will not be the case if you start with an
> icosahedron 
 
> Ilia:
> Unfortunately, it is the case.  Just to prove it, take
> a look at a triangle with the following apical positions
> <snip>

Yes, I see that you are correct. It seems that producing this with this type
of subdivision is impossible. Still, I made a distribution of 60k and
another at around 164k which were very close. Springs would not help
either, as topologically equal length sides of the subdivided shape are not
possible.

Probably the best way to start, however. On a subdivided mesh with all
points lying on a unit length sphere, the differences were around .0015, so
some repulsion correction would decrease that difference to beyond
measurable in a few steps.

 -Shay

Post a reply to this message