On Mon, 29 Oct 2012 09:56:57 -0400
Brett Cannon <brett@python.org> wrote:

> To see if the bad iterative_count and threaded_count results were
> consistently bad, I ran the benchmark suite on my MacBook Pro to see how
> "reliable" the benchmarks were. The output is below.
>
> Basically 6 benchmarks (regex_effbot, queens, startup_nosite,
> iterative_count, threaded_count, and telco) had a variance of more than 15%
> performance between my 2 computers, although queens, iterative_count, and
> threaded_count were the only ones that swung between neutral/good to bad
> depending on the machine (the rest either want from bad to very bad, or
> very good to more very good).

This is using different compilers on the 2 computers, right?

Yes: gcc 4.6.3 on Linux and Clang 3.1 on OS X.
">

(original) (raw)



On Mon, Oct 29, 2012 at 3:22 PM, Antoine Pitrou <solipsis@pitrou.net> wrote:

On Mon, 29 Oct 2012 09:56:57 -0400
Brett Cannon <brett@python.org> wrote:

\> To see if the bad iterative\_count and threaded\_count results were
\> consistently bad, I ran the benchmark suite on my MacBook Pro to see how
\> "reliable" the benchmarks were. The output is below.
\>
\> Basically 6 benchmarks (regex\_effbot, queens, startup\_nosite,
\> iterative\_count, threaded\_count, and telco) had a variance of more than 15%
\> performance between my 2 computers, although queens, iterative\_count, and
\> threaded\_count were the only ones that swung between neutral/good to bad
\> depending on the machine (the rest either want from bad to very bad, or
\> very good to more very good).

This is using different compilers on the 2 computers, right?

Yes: gcc 4.6.3 on Linux and Clang 3.1 on OS X.