I found something similar discussed, but it was while .NET 4.0 RC. As for current moment does someone tries to do some performance comparison and can share with as.It is very interesting as besides other changes such core modules like CLR and GC are updated. Does that gives any performance advantage.
+4
A:
Performance is pretty specific to particular scenarios. You cannot simply measure "performance". You have to measure "performance of [detailed, repeatable situation]".
The only real way to answer this is to measure it for your specific code.
Marc Gravell
2010-05-26 19:14:13
So that is the question. Does someone tries that ?
Incognito
2010-05-26 19:20:19
How the hell can someone else profile your code?
DeadMG
2010-05-26 19:31:06
@DeadMG I am not talking about any particular code. Where you have read that ???? I mean general comparison of runtime ! If you have project which was lets say working on .NET 2.0 and after migrating to .NET 4.0 did you get any performance gains? Hope now it is much clear to you.
Incognito
2010-05-27 04:17:04
@Incognito - that depends on *exactly* what bits of the runtime you are using. It is a safe assumption that some things have been tweaked and are faster, and maybe some things have had extra checks / options injected into them, making them slightly slower. Likewise it is safe to assume that at least *some* JIT/CLI changes have happened - that again could make it go either way, depending on **exactly** what you are using.
Marc Gravell
2010-05-27 06:41:58
The guy above you JUST said that two runtimes can only be compared with a specific piece of code. I.E., yours.
DeadMG
2010-05-27 08:55:35