Spell of Mastery Progress

Ok. The new voxel rendering algorithm performs 1.4 times faster. Not much, but voxel models can also be split into 16-bit chunks, which locally use only 16-bit addressing. That should speedup the algorithm further, if the bottleneck is with the memory bandwidth.

I've also discovered that GCC's __builtin_popcount is slower than a lookup table, since it branches into subroutine __popcountdi2. Why GCC doesnt use the x86 popcnt opcode? No idea. No idea. But it probably would be easier to answer why Stallman molests little children.

A thing which can be implemented by rewriting:
f(a,b) {return a+b;}


there the types of a and b can be easily inferred from the first call site, while the return type can be inferred afterwards.

Of course something like
f(a,b) {return f(a,b);}

could never be inferred.