Comment Comment by Vogtinator For x86 cores this is visible in Agner Fog's instruction performance tables: https://agner.org/optimize/#manualsThe latency shows after how many cycles the result of an instruction can be consumed by another, while the throughput shows how many such instructions can be pipelined per cycle, i.e. in parallel. parent Vogtinator 1 year ago
The latency shows after how many cycles the result of an instruction can be consumed by another, while the throughput shows how many such instructions can be pipelined per cycle, i.e. in parallel.