You can always use Drek's srcds calculator for recommend rates.
http://www.reece-eu.net/drekrates.php
Allan
-Original Message-
From: hlds-boun...@list.valvesoftware.com
[mailto:hlds-boun...@list.valvesoftware.com] On Behalf Of Kyle Sanderson
Sent: Thursday, July 22, 2010 1:48 AM
To:
Yes Windows here, assigning cores wont solve high cpu load it will just limit
it :(
- Original Message -
From: Mike Stiehm mikesti...@gmail.com
To: 'Half-Life dedicated Win32 server mailing list'
hlds@list.valvesoftware.com
Sent: Thursday, July 22, 2010 1:11 AM
Subject: Re: [hlds]
Thats insane, how many servers would you normally run on that setup?
Date: Wed, 21 Jul 2010 22:35:08 -0600
From: zellers.shi...@gmail.com
To: hlds@list.valvesoftware.com
Subject: [hlds] Server lag, one server on one box
So I noticed a few of you people were experiencing server lag.
That calculator seems bogus to me.
This is what it produces for a 32slot SCRDS server with a 100Mbit pipe, 2.4Ghz
CPU and 8GB mem:
sv_minrate 0
sv_maxrate 185547
sv_minupdaterate 4
sv_maxupdaterate 618
I could output random numbers like that as well...
Saint K.
-Original Message-
No one needs more than 66. 100 is for jurists. Think 'organic food'. Work on
more locations to improve latency first.
On Jul 22, 2010 8:02 AM, Saint K. sai...@specialattack.net wrote:
That calculator seems bogus to me.
This is what it produces for a 32slot SCRDS server with a 100Mbit pipe,
That's absolutely positively moronic.
sv_minrate 0
sv_maxrate 92773
sv_minupdaterate 4
sv_maxupdaterate 309
.
Kyle.
On Wed, Jul 21, 2010 at 11:21 PM, Allan Button abut...@netaccess.ca wrote:
You can always use Drek's srcds calculator for recommend rates.
http://www.reece-eu.net/drekrates.php
Minimum rates should always be 10
On Jul 22, 2010 10:42 AM, Kyle Sanderson kyle.l...@gmail.com wrote:
That's absolutely positively moronic.
sv_minrate 0
sv_maxrate 92773
sv_minupdaterate 4
sv_maxupdaterate 309
.
Kyle.
On Wed, Jul 21, 2010 at 11:21 PM, Allan Button abut...@netaccess.ca
You would be right but for me it did fix the high CPU spike I was seeing. I
have an 8 core system and did let windows assign the affinity. everything
was nice and level as for CPU usage prior the the update. After the update
just one single instance or srcds would spike to as much as 85% (of all 8
im getting reports... that is isnt.. people are still able to replicate it
on our servers.
On Wed, Jul 21, 2010 at 3:10 PM, clad iron cladi...@gmail.com wrote:
i guess this got fixed, woot...
An optional update for Team Fortress 2 is now available. Here are the
specific changes:
Server
ya we run about16 tf2 servers running on 3 different dedicated servers, we
have issues of dropouts as well, that stems back from the engineer update.
doesn't happen to everyone but there are timeouts that just occur when the
cpu spikes up.
On Thu, Jul 22, 2010 at 10:13 AM, Mike Stiehm
I use a GSP and do not have my own dedicated server, but I too seem to be
affected by this. My servers are running really bad and I have noticed my
server fps fluctuating wildly between 512 and 64. After opening a ticket
with my GSP they have included information indicating that my server is
using
That's a typical response lol. Fps fluctuation is pretty normal. But if you are
hitting 100 percent the whole server will feel it and lag.
Sent from my iPhone 4
On Jul 22, 2010, at 11:16 AM, M33CROB ad...@phoenixrisingtf2.com wrote:
I use a GSP and do not have my own dedicated server, but I
Thanks for the suggestion Eric, I've restarted a selection of our
most popular publics with a mixture of -threads 1 and -threads 3
and will monitor the results.
It will take ~24 hours to get enough stats to be able to formulate a
meaningful conclusion from, so will report back some time
I started a server with this command line:
-console -game tf +maxplayers 32 -threads 1 +map pl_goldrush
However according to Task Manager it still has 10 or 11 threads just like
the servers started without it. Is that normal?
On Thu, Jul 22, 2010 at 2:29 PM, Eric Smith er...@valvesoftware.com
Check % CPU / thread. To run an entire application in one thread is
ludicrous. Most likely those extraneous threads are necessary, but bell... I
didn't program it
On Jul 22, 2010 2:56 PM, 1nsane 1nsane...@gmail.com wrote:
I started a server with this command line:
-console -game tf +maxplayers
still get timeouts/cpu lagspikes with both threads 3 and threads 1.
On Thu, Jul 22, 2010 at 3:51 PM, HL-SDK Synths syntron...@gmail.com wrote:
Check % CPU / thread. To run an entire application in one thread is
ludicrous. Most likely those extraneous threads are necessary, but bell...
I
I am aware of this and I would never want to run the server in a single
thread. Running a mysql query and pausing the game, that's no so awesome.
On Thu, Jul 22, 2010 at 4:51 PM, HL-SDK Synths syntron...@gmail.com wrote:
Check % CPU / thread. To run an entire application in one thread is
A beta update to the Counter-Strike 1.6 dedicated server is now available. To
access the beta, run the update command with -beta hl1beta. The specific
changes include:
- Potential fix for svc_bad network errors (thanks to Sorin)
Jason
___
To
Windows server still seems to be binding only to primary IP and ignores the +ip
command line arg.
Example Command line:
srcds.exe -game swarm -console +ip 203.26.94.120 -port 27315 +maxplayers 4
+exec server.cfg +map asi-jac6-sewerjunction
However typing status in server console produces:
Change +ip to -ip
Also recommend adding -nohltv if you're binding quite a few to one IP address
.. auto port assignments overlap after a while and the server won't start.
-Original Message-
From: hlds-boun...@list.valvesoftware.com
[mailto:hlds-boun...@list.valvesoftware.com] On Behalf
On 23/07/2010, at 10:38 AM, Nick Turner wrote:
Change +ip to -ip
Also recommend adding -nohltv if you're binding quite a few to one IP address
.. auto port assignments overlap after a while and the server won't start.
-ip works, thanks.
ML.
21 matches
Mail list logo
