I have a 1700x, 4790, and a 6700 setup for csgo. If you ask me, the 4790 has
the best performance ever. While you can host more servers with 1700x but
you can't host more than 18 slots in a 128 tick server.
Will get a Coffee Lake very soon. Will see how that performs
--
Sent from:
Hi all,
Does anyone have any idea how new CPUs perform on srcds? Typical var on a full
server, etc ...?
I think it is pretty established that Intel architectures have typically
outperformed AMD's for the last 10 years, but what about Ryzen & Threadripper?
My guess would be that the 8700K is
of the servers?
Date: Sun, 5 Jul 2015 18:08:47 -0700
From: billgates8...@gmail.com
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance
Windows has always been, since 2014, the OS CSGO servers performs
the best.
This is due to multithreading of the servers, where
Windows has always been, since 2014, the OS CSGO servers performs the best.
This is due to multithreading of the servers, where the main thread and
network thread is separate.
You can also optimize the servers via these cvars, if you haven't already
done so:
sv_occlude_players 0
Anyone know why linux dont have the same multithreading of the servers?
Date: Sun, 5 Jul 2015 18:08:47 -0700
From: billgates8...@gmail.com
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance
Windows has always been, since 2014, the OS CSGO servers
What's the reason for the following cvars to be set?
sv_occlude_players 0
sv_force_transmit_players 0
On Mon, Jul 6, 2015 at 2:08 AM, L33TGaming billgates8...@gmail.com wrote:
Windows has always been, since 2014, the OS CSGO servers performs the best.
This is due to multithreading of the
What is a FF policy? please do explain a little bit more here.
--
View this message in context:
http://csgo-servers.1073505.n5.nabble.com/Server-performance-increased-tp8195p8202.html
Sent from the CSGO_Servers mailing list archive at Nabble.com.
the Valve MM servers are 60+ ping and
64 tick and we don't have leetway/altpug/ESEA)
Date: Mon, 31 Mar 2014 00:31:04 +0200
From: gbs.dead...@gmail.com
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
You can run the process with taskset: http
kill_d(){
PROCESS_NAMES=srcds_linux srcds_i686 srcds_i486 srcds_amd hlds_i686
hlds_i486 hlds_amd srcds_run
for name in $PROCESS_NAMES; do
PIDS=`pgrep $name`
for p in $PIDS; do
renice -20 $p
done
done
}
___
Csgo_servers mailing list
.
From: sza...@hotmail.com
To: csgo_servers@list.valvesoftware.com
Date: Thu, 27 Mar 2014 23:54:16 +0100
Subject: Re: [Csgo_servers] Server performance problems
I tested 64 tick 32 slots a month ago on a 2.66GHz.
Did not work
.
From: sza...@hotmail.com
To: csgo_servers@list.valvesoftware.com
Date: Thu, 27 Mar 2014 23:54:16 +0100
Subject: Re: [Csgo_servers] Server performance problems
I tested 64 tick 32 slots a month ago on a 2.66GHz.
Did not work
Hey,
first you need to install a task-manager like utitily. I prefer htop,
but you can use top.
Then you find out the Process-ID of the CS:GO server, and run the
following command as root:
# renice -20 pid
That should help to get a more stable tickrate.
Am 30.03.2014 13:51,
Thx for the fast answers.
Is it possible to change the priority 'on the fly' , so while the Server
is running and full? :)
And:
I have to change it really via 'root' and not as 'server-user'?
Am 30.03.2014 13:56, schrieb Moritz:
Hey,
first you need to install a task-manager like utitily.
Yes and yes. You can only lower the priority as normal user but sudo or
root is required to raise it.
-ics
mi...@hs-s.com kirjoitti:
Thx for the fast answers.
Is it possible to change the priority 'on the fly' , so while the
Server is running and full? :)
And:
I have to change it really
To clarify.. -20 is highest, 20 is lowest. It's reverse thing.
-ics
ics kirjoitti:
Yes and yes. You can only lower the priority as normal user but sudo
or root is required to raise it.
-ics
mi...@hs-s.com kirjoitti:
Thx for the fast answers.
Is it possible to change the priority 'on the
Thx a lot. I will test it now and i'll report if this have the effect i
hope it will. :)
Am 30.03.2014 15:37, schrieb ics:
To clarify.. -20 is highest, 20 is lowest. It's reverse thing.
-ics
ics kirjoitti:
Yes and yes. You can only lower the priority as normal user but sudo
or root is
One more time the process-priority thing:
is there a way to start the server with a higher cpu priority? is this
also '-high' as on clientside?
Am 30.03.2014 15:44, schrieb mi...@hs-s.com:
Thx a lot. I will test it now and i'll report if this have the effect
i hope it will. :)
Am
No, not as far as I know.
Am 30.03.2014 19:27, schrieb mi...@hs-s.com:
One more time the process-priority thing:
is there a way to start the server with a higher cpu priority? is this
also '-high' as on clientside?
Am 30.03.2014 15:44, schrieb mi...@hs-s.com:
Thx a lot. I will test it now
You can have the server create a .pid file and use a script to renice
or taskset it.
On Sun, Mar 30, 2014 at 8:13 PM, Moritz f...@flysoftiii.de wrote:
No, not as far as I know.
Am 30.03.2014 19:27, schrieb mi...@hs-s.com:
One more time the process-priority thing:
is there a way to start the
Thx guys.
I'll try it this way! :)
thx a lot!!
Am 30.03.2014 20:36, schrieb Valentin G.:
You can have the server create a .pid file and use a script to renice
or taskset it.
On Sun, Mar 30, 2014 at 8:13 PM, Moritz f...@flysoftiii.de wrote:
No, not as far as I know.
Am 30.03.2014 19:27,
You can run the process with taskset:
*http://linux.die.net/man/1/taskset http://linux.die.net/man/1/taskset*
If you are using your own modified srcds_run (shell script), you can add
this command. When you do your own start script, which runs the srcds_linux
directly, you have to set the
Are non-standard tick rates that well supported on CS:GO?
I was thinking of going 100-tick, but wasn't sure if the game would manage it
well.
P.S.: Does BF3 really have 10-tick? I have a hard time believing that.
_
Drop your
Mar 2014 16:07:09 +
Subject: Re: [Csgo_servers] Server performance at 100-tick
Are non-standard tick rates that well supported on CS:GO?
I was thinking of going 100-tick, but wasn't sure if the game would manage it
well.
P.S.: Does BF3 really have 10-tick? I have a hard time believing
Yes and it has it's own issues.
http://www.reddit.com/r/battlefield_4/comments/1uiqje/while_the_10hz_tick_rate_remains_net_code_will/
Should be noted that for example valve's L4D1 and 2 ran on 30 ticks, and
there were issues with sliding of events. Like you hit a zombie off
and it didn't take
BF3 and BF4 use the same netcode, and they update their clients at 10htz.
BF3 and BF4 are *ten* tick. It's insane.
Sorry to reply with non-csgo_server stuff, just like to point out that
using BF servers as benchmarks is deceptive. They are 'expensive' on
resources, but really are laughably low
102.4 - Like in the CS:GO Beta. Works fine for us. We decreased the
Tick nearly 2 Month ago due to the unbelievable increase of CPU Usage
for just one gameserver!
Am 28.03.2014 17:12, schrieb Gordon Reynolds:
BF3 and BF4 use the same netcode, and they update their clients at 10htz.
BF3 and
+0200
From: ics i...@ics-base.net
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
Message-ID: 533308bb.9010...@ics-base.net
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Interesting results you got. I'm running 128 tick 24 slot
Tested Windows today, i had better fps and less cpu use with Linux, windows
is worse.
--
View this message in context:
http://csgo-servers.1073505.n5.nabble.com/Server-performance-problems-tp6375p6886.html
Sent from the CSGO_Servers mailing list archive at Nabble.com.
Im talking bout 32slot with 128ticks, when i run 10slot on linux i get
0.1-0.3 var.
--
View this message in context:
http://csgo-servers.1073505.n5.nabble.com/Server-performance-problems-tp6375p6888.html
Sent from the CSGO_Servers mailing list archive at Nabble.com.
:03:38 +0100
Subject: Re: [Csgo_servers] Server performance problems
Just tried 24 slot DM with:
2x Intel Xeon E5-2643 v2 (25M Cache, 3.50 GHz, Turbo 3.80 GHz)
2x 8GB RDIMM (1600MT/s)
2x SSD 240GB
2x Broadcom 57810-k Dual port 10Gb Network Card
No stable tickrate.
Date: Sun, 23 Mar
I tested 64 tick 32 slots a month ago on a 2.66GHz.Did not work.
From: lewis...@gmail.com
Date: Thu, 27 Mar 2014 18:05:26 -0400
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
Drop your tic from 128 to 100 and report back. I believe dm's don't need
After yesterday's update. My server is experiencing sv drops, E3 1270v2 here.
From: sza...@hotmail.com
To: csgo_servers@list.valvesoftware.com
Date: Thu, 27 Mar 2014 23:54:16 +0100
Subject: Re: [Csgo_servers] Server performance problems
I tested 64 tick 32 slots a month ago on a 2.66GHz.Did
I creat the csgo server recommendations for best windows os ? (bad english
sorry)
Date: Wed, 26 Mar 2014 19:04:59 +0200
From: i...@ics-base.net
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
Interesting results you got. I'm running 128 tick
: Wed, 26 Mar 2014 19:04:59 +0200
From: i...@ics-base.net
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
Interesting results you got. I'm running 128 tick 24 slot server on
Q9559 (2.83GHz) and it used to be fine before Valve's update
Im running i5 4670k with 4.5GHz OC, 0.3-1VAR with 32 players on 128ticks, no
drops
--
View this message in context:
http://csgo-servers.1073505.n5.nabble.com/Server-performance-problems-tp6375p6831.html
Sent from the CSGO_Servers mailing list archive at Nabble.com.
Well then we just need Valve to buy us all supers-ervers to run their fucking
gameservers.
I need 12 cores running 4.5GHz, mmkay?
Date: Wed, 26 Mar 2014 17:00:26 -0700
From: alexander.hj...@playstar.se
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance
Buy 4 mini-itx with i5s and you got 16cores for 2k euro :)
--
View this message in context:
http://csgo-servers.1073505.n5.nabble.com/Server-performance-problems-tp6375p6833.html
Sent from the CSGO_Servers mailing list archive at Nabble.com.
___
:56 -0400
From: evourr evo...@gmail.com
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
Message-ID: 61E2D9AFEAF4414786A53C8216886333@SNSC3PC
Content-Type: text/plain; charset=utf-8
Yes, it will drastically reduce cpu usage.
- Original
Saying you are able to tell the difference between 128 and 120 tick, which is
effectively 0.5ms and 128 and f.e. 64 tick . which is about 8ms ..is
a totally different story.
--
View this message in context:
But subjectively based upon your thinking...can you tell the difference of
8ms? How we feel in game is all extrapolated, meaning the game fills up gaps
between server updates. When the tickrate is fluctuating, the client
receives information at an unregular interval. Just as an example, you
thats not how tickrate works ... you certainly won't get 64 ticks in the
first 100ms, then nothing happens and the server will send you the missing
56tick in the last 250ms.
128 tick = server updates the game ( on his end..) every ~8ms.
If the server has such a high var ( i guess 4.0..) making
to make the answer even clearer:
128@1.5var vs. 128@0.5var = max. performance difference of about 2.56% ...
1x128tick vs. 2x128tick = DOUBLE amount of slots. For a performance loss of
max ( !! ) 2.56%.
The only reason people aim for as low var as possible ... so players can
sleep in peace. Or
L33TGaming wrote
Do you understand var? Var is not the deviation from server tickrate
rather
deviation from the last timeframe. If your server tickrate drops to a
sustained 64 from 128, your var isn't going to be 64.
When you get 128 FPS server or client wise, you're not getting 8ms per
:38 +0100
Subject: Re: [Csgo_servers] Server performance problems
Just tried 24 slot DM with:2x Intel Xeon E5-2643 v2 (25M Cache, 3.50 GHz, Turbo
3.80 GHz)2x 8GB RDIMM (1600MT/s)2x SSD 240GB2x Broadcom 57810-k Dual port 10Gb
Network Card
No stable tickrate.
Date: Sun, 23 Mar 2014 15:55:39 +0200
Hmm I see but isn't that related with the same map problems from css?
Where most of times stripper was needed as a workaround?
Date: Sun, 23 Mar 2014 15:55:39 +0200
From: i...@ics-base.net
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
It's
To: csgo_servers@list.valvesoftware.com csgo_servers@list.valvesoftware.com
Sent: Sunday, March 23, 2014 10:35:41 AM
Subject: Re: [Csgo_servers] Server performance problems
Hmm I see but isn't that related with the same map problems from css?
Where most of times stripper was needed
You need a better cpu, i did setup a new node today with a Inte Desktop CPU
we had in spare and run 128tick and 32slots with no tick drops at all, you
can try them here.
192.121.44.55 - Standard 192.121.44.56 - DM FFA 192.121.44.57 - DM FFA
Pistols
--
View this message in context:
You wanna share how exactly you did it or are you just here to brag?
Date: Sat, 22 Mar 2014 12:49:03 -0700
From: alexander.hj...@playstar.se
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
You need a better cpu, i did setup a new node today
I can't verify because two are empty and one has only three players. You
won't see drops unless there are a decent number of players.
On Mar 22, 2014 3:49 PM, AlexHjelm alexander.hj...@playstar.se wrote:
You need a better cpu, i did setup a new node today with a Inte Desktop CPU
we had in spare
He had 128Tick with 32 bots.His post is just advertisement/braging. :P
Date: Sat, 22 Mar 2014 15:54:08 -0400
From: goabs...@absurdminds.net
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
I can't verify because two are empty and one has only three
2014 15:54:08 -0400
From: goabs...@absurdminds.net
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
I can't verify because two are empty and one has only three players. You
won't see drops unless there are a decent number of players.
On Mar 22, 2014
He had 32 bots before he posted here. :P
Date: Sat, 22 Mar 2014 16:01:24 -0400
From: goabs...@absurdminds.net
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
I tried all three and none even had bots. If you want people to test it out, at
least
Hi, i did open this server up with bots now.
192.121.44.56:27015
Bots do use more server resources than regular players.
--
View this message in context:
http://csgo-servers.1073505.n5.nabble.com/Server-performance-problems-tp6375p6739.html
Sent from the CSGO_Servers mailing list archive at
Same here. its not stable. :(
Date: Sat, 22 Mar 2014 16:28:23 -0400
From: goabs...@absurdminds.net
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
Immediately when I joined, the tickrate went yellow, abs I see it spiking from
1 to 2.8. In other
Try now, i tested a unstable sourcemod addon on it :)
Should be 100% stable now, no yellow here.
--
View this message in context:
http://csgo-servers.1073505.n5.nabble.com/Server-performance-problems-tp6375p6742.html
Sent from the CSGO_Servers mailing list archive at Nabble.com.
I don't see any performance difference. I see you're even in the server
with me, so surely you see the spikes. I just saw the tickrate go to double
digits it dropped so low.
On Mar 22, 2014 4:31 PM, AlexHjelm alexander.hj...@playstar.se wrote:
Try now, i tested a unstable sourcemod addon on it
Hm I got stable tick after the reboot. But I think the tickrate will go more
unstable the longer the server is up.
Date: Sat, 22 Mar 2014 16:45:01 -0400
From: goabs...@absurdminds.net
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server performance problems
I don't see any
I dont have that problem, mine is stuck on 128ticks all the time, no drops at
all on the sv.
So i guess you have some problems or something?
--
View this message in context:
http://csgo-servers.1073505.n5.nabble.com/Server-performance-problems-tp6375p6745.html
Sent from the CSGO_Servers
Lol, anyone saying that they get no tick drops on their server is a liar.
It doesn't matter how godlike your machine is, you are going to get tick
drops. It's inevitable, Mr. Anderson.
On Sat, Mar 22, 2014 at 4:48 PM, AlexHjelm alexander.hj...@playstar.sewrote:
I dont have that problem, mine
Calling people here words wont do you any good.
Try the server and see for yourself before you call people stuff.
--
View this message in context:
http://csgo-servers.1073505.n5.nabble.com/Server-performance-problems-tp6375p6748.html
Sent from the CSGO_Servers mailing list archive at
I wasn't calling any one person any name. I'm saying that it's damn near
impossible for any server, no matter the quality, to not have any tick rate
drops with CS:GO. Every server experiences at least minor drops, especially
when players are connected. Anyone saying their server gets no Var at
First off, I'll like to mention that CS:GO servers on Linux are 60% - 80%
slower than the same machine on Windows. Even if you had a 1270 on linux,
you are not able to host 128 tick 32 player servers. It's entirely possible
to get stable 128 tick with 32 players, all you need is windows and a
Mate, if you're here to tell us Oh your servers are crap, look at
mine...they are so good, just shut it and get out of here. If you're a good
person, tell the rest of the community how you accomplished it.
--
View this message in context:
I just wanted to point out, that you all are arguing about something that's
barely noticable to the regular player. Like ...not even close. A 128 tick
server with a var of even 2.0 what? That's like a tick of 120-128 (
wild guess )? Are you REALLY arguing about 8 ticks or what?
The whole
I for one can definitely tell the difference. I normally have net_graph
disabled so I'm not constantly focusing on the variance, but I've tested
myself and been pretty spot on. If you're getting high fps and playing on a
high hertz monitor it's actually really easy to tell. Yeah, it might not be
So you tell me you can notice (!!) a worse hitreg with 120 tick compared to
128. Go join The Avengers pls.
( btw: im also playing with high fps and a 120hz monitor)
--
View this message in context:
http://csgo-servers.1073505.n5.nabble.com/Server-performance-problems-tp6375p6759.html
Sent
Yes, that's exactly what I'm telling you.
On Mar 23, 2014 1:21 AM, Antix mtmu...@web.de wrote:
So you tell me you can notice (!!) a worse hitreg with 120 tick compared to
128. Go join The Avengers pls.
( btw: im also playing with high fps and a 120hz monitor)
--
View this message in
Go join The Avengers then, since you must be super human. I guess Mr. Fury
will have a place for you.
Just to clear things out about how ridiculous it is, to claim being able to
tell a difference between 120 tick and 128 tick.
128 tick = 7,8125 ms between each tick
120 tick = 8,3 ms between
Hey guys,
since the Valve-Staff seems to read reddit I've tried to post a topic
about the server performance problems there. Maybe this helps.
http://www.reddit.com/r/GlobalOffensive/comments/1xwzye/csgo_server_performance_a_plea_from_the_server/
If you feel like agreeing with what I wrote
.
Thanks
Nick Ireland
From: csgo_servers-boun...@list.valvesoftware.com
[mailto:csgo_servers-boun...@list.valvesoftware.com] On Behalf Of Jonathon King
Sent: Thursday, 14 June 2012 2:26 PM
To: csgo_servers@list.valvesoftware.com
Subject: [Csgo_servers] Server Performance
Hello all
PM
To: csgo_servers@list.valvesoftware.com
Subject: [Csgo_servers] Server Performance
Hello all. Just wondering if some of you other server operators could let me
know what kind of CPU usage you are seeing on populated servers, and your
hardware.
This is what I’m seeing with a full 32
is
meaningless.
Thanks
Nick Ireland
From: csgo_servers-boun...@list.valvesoftware.com
[mailto:csgo_servers-boun...@list.valvesoftware.com] On Behalf Of Jonathon
King
Sent: Thursday, 14 June 2012 2:26 PM
To: csgo_servers@list.valvesoftware.com
Subject: [Csgo_servers] Server Performance
[mailto:
csgo_servers-boun...@list.valvesoftware.com] On Behalf Of Jonathon King
Sent: Thursday, 14 June 2012 2:26 PM
To: csgo_servers@list.valvesoftware.com
Subject: [Csgo_servers] Server Performance
Hello all. Just wondering if some of you other server operators could
let me know
Of Jonathon
King
Sent: Thursday, 14 June 2012 2:26 PM
To: csgo_servers@list.valvesoftware.com
Subject: [Csgo_servers] Server Performance
Hello all. Just wondering if some of you other server operators could
let me know what kind of CPU usage you are seeing on populated servers
;)
Thanks
Nick Ireland
-Original Message-
From: csgo_servers-boun...@list.valvesoftware.com
[mailto:csgo_servers-boun...@list.valvesoftware.com] On Behalf Of Alistair C
Sent: Thursday, 14 June 2012 8:06 PM
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server Performance
: Thursday, 14 June 2012 8:06 PM
To: csgo_servers@list.valvesoftware.com
Subject: Re: [Csgo_servers] Server Performance
Seems at odds with:
https://support.steampowered.com/kb_article.php?ref=5386-HMJI-5162
WHY HIGHER FPS?
The key reason to run higher FPS is the render time. At 1000FPS
is
that the default setting? How do I set fps max in command line? Should it
be exactly the same as the tickrate, or slightly above?
*From:* Nick Ireland nick.k.irel...@gmail.com
*Sent:* Thursday, June 14, 2012 10:31 AM
*To:* csgo_servers@list.valvesoftware.com
*Subject:* Re: [Csgo_servers] Server
Hello all. Just wondering if some of you other server operators could let me
know what kind of CPU usage you are seeing on populated servers, and your
hardware.
This is what I’m seeing with a full 32 player dust2 deathmatch plugin server
(66 tick).
http://i.imgur.com/11coR.png
It seems pretty
78 matches
Mail list logo
