AthlonRob wrote:

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
NotDashEscaped: You need GnuPG to verify this message

On Wed, 30 Jun 2004 13:48:41 GMT, f/f george <george@yourplace.com>
wrote:

I think if you run the mp3 palyers WHILE doing the benchmarking you
MAY come out with a closer to accurate time. The benchmark is based
on whatever the computer is doing while benchmarking, if you stop
doing your regular stuff it will say it is faster than it really is.
Maybe the Celerons are fast in the beginning but then slow down
because of the small cache and that is affecting the benchmarks? If
so then the programmers should have caught that!

Actualy, not quite... the benchmarks are based on the CPU time it
takes to perform some calculations, so using the system ideally
shouldn't have an effect on this.  In reality, it does have an effect
depending on the operating system and what, exactly, is being done at
the same time as
the benchmarks.

The workunit estimated times are just estimates.  Each workunit is a
little bit different, I think they're using just one estimate for all
workunits at the moment.

Thanks Rob, I figured it was something like that. My initial post was
motivated by curiosity, I just wondered what different CPUs were getting in
the way of benchmarks.

Cheers,
--
~misfit~