[cod] Linux vs. Windows

[Steven Hartland]

lol I think you need to do some background reading.

Its a know issue that 2.4 kernels often report inaccurate
CPU usage so if ur usage seems too good to be true it
is :P

Basic detail is its all to do with timeslices, if a process
yields before its up it will register zero cpu usage. So
you can write a process which actually uses 99% CPU
but will show up as 0% in both top and ps ( as they both
use the kernel for CPU usage info ).

The nature of game servers seems to make them very
susceptible to inaccurate CPU measurement.

Sorry to burst ur bubble :P

2.6 kernels with their default higher HZ are significantly
less affected by this issue as is FreeBSD.

For Windows users watch out for people quoting total
CPU % when either a dual processor or HT enabled
as this will mean the real usage is 2 * or 4 * ( dual cpu HT )
that which they quote. e.g. 25% on a dual proc HT enabled
machine is actually using 100% of a single CPU.

    Steve / K

----- Original Message ----- 
From: <patrick at fragzzhost.com>
To: <cod at icculus.org>
Sent: Saturday, November 20, 2004 7:59 PM
Subject: Re: [cod] Linux vs. Windows


dont be jealus my friend :)

Btw chris check ur damm phone hehe
----- Original Message ----- 
From: "Steven Hartland" <killing at multiplay.co.uk>


> Let me guess a 2.4 kernel right? If so that CPU usage is
> just wrong sorry.
> 
>    Steve / K
> ----- Original Message ----- 
> From: "Chris Adams" <chris at fragzzhost.com>
> 
> 
>> 8% here for CS Source 32 players 3.06 CPU :-)



================================================
This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. 

In the event of misdirection, illegible or incomplete transmission please telephone (023) 8024 3137
or return the E.mail to postmaster at multiplay.co.uk.