3 ms·
I think he's saying that the MMX versions of these functions actually perform worse than what the compiler could produce from the plain C code, because the comp
by pavlov 15y ago
I think he's saying that the MMX versions of these functions actually perform worse than what the compiler could produce from the plain C code, because the compiler would be able to use SSE2 for all floating-point math. Hence, these elaborately hand-optimized code paths are actually "pessimizations".
- exDM69 15y ago... for new hardware. If you still target Pentium 2 boxes, this is bad news for you :)