3 ms·
What you want to look at in that table is reciprocal throughput, which is almost everywhere doubled for 512-bit wide instructions.
by Denvercoder9 4y ago
What you want to look at in that table is reciprocal throughput, which is almost everywhere doubled for 512-bit wide instructions.
- kolbe 4y agoI think you're right, but I've passed the hacker news edit threshold. May my misinformation live on forever.
- celrod 4y agoI'm guessing that 20% is still enough for your zen4 to be faster than raptor lake running the avx2 path, while also probably using less power.
- dathinab 4y agono you want to look at benchmarks of realistic real word applications pure throughput doesn't matter if in most realistic use cases you will never reach it
- kolbe 4y agoI use it via the MKL and https://github.com/vectorclass/ https://github.com/vectorclass/ I think they have very efficient pipelines.