Faradd optimization¶
squall, AMD Opteron 8220 @ 2.8 GHz | |||
gcc 4.5.2 | gcc 4.6.3 | icc 12.1.3 | |
loop | 117300 | 117500 | 100700 |
sse2 intrinsics | 103800 | 105700 | 100200 |
4 times unrolled, sse2 intrinsics | 103500 | 101200 | 99500 |
wanglung, INTEL Xeon E5520 @ 2.3 GHz | |||
gcc 4.5.3 | gcc 4.6.3 | gcc 4.7.0 | |
loop | 53300 | 51500 | 51300 |
sse2 intrinsics | 45500 | 44000 | 44100 |
4 times unrolled, sse2 intrinsics | 44500 | 41000 | 43000 |