I want to double check something.

Is it the case that the number of vertex normals and the number of
vertices should be the same in a SubGeometry object?

I'm fixing the ColladaParser to read triangles. Collada indexes
normals differently from vertices, and I'm having to create a parallel
array - since there is only one index array.  I assume that this is a
limitation/specification of Molehill's api's.

(Yes, I could auto calculate normals, but they end up backwards - I
assume that's because the vertex winding is opposite what away3d
expects. I could reverse the winding in the parser, but I decided it
would make more sense to read the Collada-provided normals as the
artists may tweak them in our pipeline someplace).

Reply via email to