[eqvinox]

It's also that running a benchmark by wrapping it in a shell script and invoking "date" externally adds a constant offset to all timings. It's negligible if the benchmark itself takes long enough, but not when you go down to 0.25s or even 0.06s.

[igouy]

And it's not like we can't find programming language benchmark comparisons that do better than that:

https://benchmarksgame-team.pages.debian.net/benchmarksgame/