2 ms·
>but if up to 15% isn't significant, where do we draw that line?. It's not like we don't have a yardstick for actually noticeable speed updates. Pre-JIT JS to
by coldtea 6y ago
>but if up to 15% isn't significant, where do we draw that line?.
It's not like we don't have a yardstick for actually noticeable speed updates. Pre-JIT JS to JIT JS for example was a big several times speedup. PHP 6 vs PHP7 was several times as well (no JIT involved here, just other optimizations).
15% is only 'significant' in the context of an already heavily optimized program (which FF's JIT is). Not significant as in "the users will notice it".