Infinite floats

Infinite floats

I think you should really use very large/small floats values for stuff like bounding boxes, where currently fp infinity is used (bbox.h constructors). That's because those infinities propagate through the code and are causing incorrect values and asserts, therefore -> rendering artefacts (bv4_builder.cpp line 121). (Ignoring the fact that a little slowdown may also be observed when INF is involved in calculations).

1 post / 0 new
For more complete information about compiler optimizations, see our Optimization Notice.