Arne Goedeke wrote:
>I benchmarked the sprintf changes, and there is really no difference. I
>think in theory on cpus with small caches performance should improve, but thats
>just a guess.

Actually, even on a processor with a large enough cache, there will be a
noticable difference if multiple processes are running; it will allow
better hitrates on the cache despite context switches.
This is, obviously, notoriously difficult to benchmark.
-- 
Stephen.
  • Re:... Marcus Comstedt (ACROSS) (Hail Ilpalazzo!) @ Pike (-) developers forum
  • Re:... Arne Goedeke
  • Re:... Arne Goedeke
  • Re:... Arne Goedeke
  • old... Arne Goedeke
  • old... Martin Nilsson (Opera Mini - AFK!) @ Pike (-) developers forum
  • Re:... Arne Goedeke
  • Re:... Martin Nilsson (Opera Mini - AFK!) @ Pike (-) developers forum
  • Re:... Arne Goedeke
  • Re:... Arne Goedeke
  • Re:... Stephen R. van den Berg
  • Re:... Stephen R. van den Berg
  • Re:... Arne Goedeke
  • old... Henrik Grubbstr�m (Lysator) @ Pike (-) developers forum
  • old... Per Hedbor () @ Pike (-) developers forum
  • Re:... Stephen R. van den Berg
  • Re:... Martin Nilsson (Opera Mini - AFK!) @ Pike (-) developers forum
  • Re:... Per Hedbor () @ Pike (-) developers forum
  • Re:... Per Hedbor () @ Pike (-) developers forum
  • Re:... Stephen R. van den Berg
  • Re:... Marcus Comstedt (ACROSS) (Hail Ilpalazzo!) @ Pike (-) developers forum

Reply via email to