> Except literally running at half capacity. I’ve been writing SIMD code for some time now, and I disagree. You can read the documentation I’ve generated http

> Except literally running at half capacity. I’ve been writing SIMD code for some time now, and I disagree. You can read the documentation I’ve generated https://github.com/Const-me/IntelIntrinsics https://github.com/Const-me/IntelIntrinsics AVX intrinsics begin with _mm256_. You’ll find that for some AVX operations, such as _mm256_fmadd_ps, _mm256_adds_epi8, _mm256_blendv_ps, both latency and throughput of Ryzen is equivalent to Skylake. For some others, e.g. _mm256_add_ps, _mm256_mul_ps, Ryzen is close to some older Intel, Haswell or Broadwell. Only for very rarely used stuff, e.g. _mm256_broadcast_ps, _mm256_madd_epi16, Ryzen is significantly slower than Intel.