[Mono-list] Newbye questions about mono's speed

Edward Ned Harvey (mono) edward.harvey.mono at clevertrove.com
Fri Sep 5 13:56:38 UTC 2014


> From: mono-list-bounces at lists.ximian.com [mailto:mono-list-
> bounces at lists.ximian.com] On Behalf Of gbenrais
> 
> <http://benchmarksgame.alioth.debian.org/u64q/program.php?test=nbody
> &lang=csharp&id=3>

That program, at a glance, seems to only do basic operations, without touching on any big (potentially un-optimized) classes.  So I suspect, the biggest factor on this program will be enabling the "Optimizations" during compile.  Set your project to "Release" and not "Debug."  Probably will make an 8x performance difference.

Also, mono and .NET have different startup overhead times.  Not sure how significant that is.


> *Windows machine :*
> Mono JIT compiler version 3.3.0 (master/6cd4ddc)

Wait, what?  Don't run mono on windows unless you need to.  Mono on windows exists for various reasons, but if you're a newbie to mono, you're better off to just ignore it until you know what you're doing and have a need for it.  For the program you referenced, no reason to do it.  Just use Visual Studio & .NET.

Mono 3.3.0 vs 2.10.8.  That's a very significant difference.

x86 vs x86_64, that's a very significant difference.

Completely different machines, different architectures, CPU speed, memory speed.  Big huge performance differences.  You're not really measuring the mono performance relative to anything.  You're largely just measuring machine performance with no point of reference.


More information about the Mono-list mailing list