2 ms·
Note that simply using a SIMD vector class library does not make it "go faster". In fact it can make things worse (due to latency). What you usually need is a p
by Cloudef 4y ago
Note that simply using a SIMD vector class library does not make it "go faster". In fact it can make things worse (due to latency). What you usually need is a problem and then a solution (algorithm) that parallelizes well.
- kolbe 4y agoExplain? My experience has been that you don't need much. I've benchmarked Agner's exp, and if you have 4 calculations to do, then calling it with avx2 will be 4x faster than calling std::exp four times.