Here are a few interesting benchmarks on QEMU, Valgrid simulation, Bochs and native speeds on x86. Also, a new stable release, Valgrind 2.2.0, is available. 2.2.0 brings many improvements over 2.0.0, and includes the new Massif heap-profiling tool.
QEMU, Valgrid, Bochs Benchmarks
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Why are the versions of the software used in this benchmark so old?
Surely all of these packages are faster now. Additionally, it appears to be done by the guy who wrote QEMU, so it’s not exactly an independent benchmark.
Look at the results of the Alpha processor. A 667 MHz Alpha runs x86 code faster than a 500 MHz P3. This might not be a surprise to some, but it was for me.
It would be more interesting to present benchmarks that aren’t this outdated. How do current versions of the software compare?
Qemu has changed a bit indeed but Bochs hasn’t changed much since last year and Valgrind hasn’t either (at least not in the simulation area that’s been measured).
Look at the results of the Alpha processor. A 667 MHz Alpha runs x86 code faster than a 500 MHz P3. This might not be a surprise to some, but it was for me.
No no no… they are *native* alpha executables, and yeah the 666MHz Alpha goes about 150-300% faster than the PIII-500, but that is not at all surprising..
When emulating x86 under QEMU, you can see the alpha goes about 70-150% the speed of (ie. *roughly* equivalent to) a pentium 90.
Ah, I was reading the tables wrong. My bad.
This benchmarking page has been up on the QEMU for months.
it’s nice you all noticed that these are quite old (Jul 10, 2003) – you could of course renew them!