Just reporting information for hardware comparison, in case anyone is 
interested (don't knock it if you know this already):

128 tick, 26 players, Windows 7, Phenom II X4 965 (3.4GHz quad-core)
= Unstable var. Dropping to 22 players gave acceptable 0.6 - 0.9 var.

128 tick, 24 players, Windows 7, Sandy Bridge (2.2GHz hyperthreaded quad-core, 
so 8 threads)
= Unstable var. Dropping to 20 players gave acceptable 0.3 - 0.9 var.

Would be interesting to know if these performance limits are down to hardware 
or the 128-tick implementation of the CS:GO dedicated server.
I am planning to test on a 12-thread Ivy Bridge-E system to answer this 
question. Has anyone tried this on a 4.7GHz 4770K to see what happens?

> 
> Message: 5
> Date: Wed, 26 Mar 2014 03:13:25 -0700 (PDT)
> From: L33TGaming <[email protected]>
> To: [email protected]
> Subject: Re: [Csgo_servers] Server performance problems
> Message-ID: <[email protected]>
> Content-Type: text/plain; charset=us-ascii
> 
> Just ordered a Xeon 2667v2 server. Lets see how this server goes with 32 real
> players.

                                          
_______________________________________________
Csgo_servers mailing list
[email protected]
https://list.valvesoftware.com/cgi-bin/mailman/listinfo/csgo_servers

Reply via email to