[Python-checkins] r46083 - sandbox/trunk/stringbench sandbox/trunk/stringbench/stringbench.py

[Fredrik Lundh]

M.-A. Lemburg wrote:

>> (and pybench is also extremely unreliable if you look at
>> individual tests; more about that later).
> 
> Hmm, awaiting your evidence ;-)

the current case is a portion of the pybench unicode bench-
mark, where pybench reports a massive slowdown for a portion
of the library that hasn't really changed, and timeit reports
stable timings for *exactly* the same piece code:

   before, according to timeit: 140 ms
    after, according to timeit: 139 ms
slowdown, according to pybench: 62%

whatever pybench is measuring in this case, it's not measuring
the speed of the actual operation...

we're looking into this as we speak.

</F>