in reply to Re: Finding all substrings
in thread Finding all substrings

Um, just reading the Benchmark output man:
Rate original new original 7710/s -- -80% new 38610/s 401% --
If that doesn't mean 400% faster then I guess I need more education.


Replies are listed 'Best First'.
Re: (3) Finding all substrings (Russ: more != less)
by Russ (Deacon) on Apr 24, 2002 at 19:49 UTC
    Benchmark is telling you:
    • new is 80% faster than original
    • original is 401% slower than new
    Don't confuse "this much more" with "this much less." :-)

    Brainbench 'Most Valuable Professional' for Perl

      The issue here has nothing to do with "less" and "more", and everything to do with units of measure. When samtregar says his routine is 400% faster, he is obviously implying a unit of measure based on rate, ie. number of executions per second. Dominus's quip assumed an inverse unit of measure based on seconds per execution. Personally, I think samtregar's assumption about units was the better one. Most people are accustomed to thinking about things in terms of rate, ie. if I say I made my script run 4x faster, there's really no confusion. If I say I made it run 75% faster, things get muddled. But once it goes over 100%, there's no question as to what is meant.
                     s aamecha.s a..a\u$&owag.print