At 08:42 PM 5/19/2009, bob dolet wrote:
Hello everyone, i am having issues with srcds running CSS servers, every so
often (once an hour?) there will be a short warp in game, lasting around
500ms. This is happening on my servers, as well as a friends running
different hardware and different
At 08:42 PM 5/19/2009, bob dolet wrote:
Hello everyone, i am having issues with srcds running CSS servers, every so
often (once an hour?) there will be a short warp in game, lasting around
500ms. This is happening on my servers, as well as a friends running
different hardware and different
At 05:10 PM 4/28/2009, Kveri wrote:
Hello,
I have some new info about 1000fps servers that we were discussing in
November 2008.
Old topic is here:
http://list.valvesoftware.com/mailman/private/hlds_linux/2008-November/058527.html
So, I was able to get constant 1000fps server, by constant I mean
At 11:25 AM 4/20/2009, Steffen Tronstad wrote:
Dont think about using realtime patch, it will
kill your CPU in notime (or, og ahead and try it, but expect loaaad)
HRTS = On
IRQ balancing wouldnt do that much difference
afaik. I dont change it from default value of whatever kernel i am using.
At 11:25 AM 4/20/2009, Steffen Tronstad wrote:
Dont think about using realtime patch, it will
kill your CPU in notime (or, og ahead and try it, but expect loaaad)
HRTS = On
IRQ balancing wouldnt do that much difference
afaik. I dont change it from default value of whatever kernel i am using.
, but it catches a majority of them.
Nowadays you just hope your game is shitty enough that nobody wants
to bother cheating in it (SOF 3: Payback is an example of a shitty game)
:-)
G. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org | gary
at cpanel.net
http
, but it catches a majority of them.
Nowadays you just hope your game is shitty enough that nobody wants
to bother cheating in it (SOF 3: Payback is an example of a shitty game)
:-)
G. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org | gary
at cpanel.net
http
At 06:01 AM 4/16/2009, ServerAlex wrote:
your maxrate is too low for 32 players. you need at least 5. and
minrate 13000 isnt recommended either.. take 2 or 25000.
additionally you shouldn't install two servers on such a slow machine.
24-slots tf2 takes up to 60-80% usage on a q6600 for
At 06:01 AM 4/16/2009, ServerAlex wrote:
your maxrate is too low for 32 players. you need at least 5. and
minrate 13000 isnt recommended either.. take 2 or 25000.
additionally you shouldn't install two servers on such a slow machine.
24-slots tf2 takes up to 60-80% usage on a q6600 for
At 06:05 AM 4/10/2009, Ronny Schedel wrote:
Sorry, but Pentium 3 is not low end, it is ancient. This is what they find
when they dig for lost civilizations.
Pentium 3's have better syscall latency than P4s.
I'm running a TF2 server on a 1133MHz Pentium-3 computer (2GB of ram,
100mbit
At 06:05 AM 4/10/2009, Ronny Schedel wrote:
Sorry, but Pentium 3 is not low end, it is ancient. This is what they find
when they dig for lost civilizations.
Pentium 3's have better syscall latency than P4s.
I'm running a TF2 server on a 1133MHz Pentium-3 computer (2GB of ram,
100mbit
At 07:30 PM 3/24/2009, Guy Watkins wrote:
I think a Mask of 255.255.255.255 is wrong. I don't see how that can route
or broadcast. I think all IPs should be on the same subnet. But I have not
used virtual interfaces on Linux.
It's not bad for aliases.
At 07:30 PM 3/24/2009, Guy Watkins wrote:
I think a Mask of 255.255.255.255 is wrong. I don't see how that can route
or broadcast. I think all IPs should be on the same subnet. But I have not
used virtual interfaces on Linux.
It's not bad for aliases.
At 02:53 AM 3/19/2009, bob dolet wrote:
Hello everyone, running debian lenny with 2.6.26.8-rt16 patch, 1000Hz,
hrtimer/srcds set to realtime. Running tournament mode with no random
damage, valve maps, no mods, at random times, -not often- on a machine that
isnt heavily loaded at all, each cpu
At 02:53 AM 3/19/2009, bob dolet wrote:
Hello everyone, running debian lenny with 2.6.26.8-rt16 patch, 1000Hz,
hrtimer/srcds set to realtime. Running tournament mode with no random
damage, valve maps, no mods, at random times, -not often- on a machine that
isnt heavily loaded at all, each cpu
At 02:45 PM 3/8/2009, Ben B wrote:
okay, i compiled the kernel, and I see all of them.
you think i should use tsc or hpet? there are opinions for both throughout
the list it seems.
TSC is fast to read because it lives in the CPU..
HPET lives off somewhere behind a bridge, so reading it requires
At 02:45 PM 3/8/2009, Ben B wrote:
okay, i compiled the kernel, and I see all of them.
you think i should use tsc or hpet? there are opinions for both throughout
the list it seems.
TSC is fast to read because it lives in the CPU..
HPET lives off somewhere behind a bridge, so reading it requires
At 08:27 AM 3/7/2009, listac...@lvwnet.com wrote:
I feel like such a linux nub... how do you set your clocksource?
On any fairly recent 2.6-series kernel you can put this right on the
kernel boot commandline:
clocksource=$whatever
like
clocksource=hpet
or
clocksource=tsc
so that it is set
At 08:27 AM 3/7/2009, listac...@lvwnet.com wrote:
I feel like such a linux nub... how do you set your clocksource?
On any fairly recent 2.6-series kernel you can put this right on the
kernel boot commandline:
clocksource=$whatever
like
clocksource=hpet
or
clocksource=tsc
so that it is set
into the kernel :P
On Sat, Mar 7, 2009 at 8:34 AM, Gary Stanley g...@velocity-servers.netwrote:
At 08:27 AM 3/7/2009, listac...@lvwnet.com wrote:
I feel like such a linux nub... how do you set your clocksource?
On any fairly recent 2.6-series kernel you can put this right on the
kernel boot
into the kernel :P
On Sat, Mar 7, 2009 at 8:34 AM, Gary Stanley g...@velocity-servers.netwrote:
At 08:27 AM 3/7/2009, listac...@lvwnet.com wrote:
I feel like such a linux nub... how do you set your clocksource?
On any fairly recent 2.6-series kernel you can put this right on the
kernel boot
At 01:12 PM 3/7/2009, Ben B wrote:
dont ask me, this is how it was when i found it. So if I compile my own
kernel, it should have them clocksources?
What's the motherboard? is it a virtual machine?
On Sat, Mar 7, 2009 at 11:01 AM, Gary Stanley
g...@velocity-servers.netwrote:
At 12:47 PM 3
At 01:12 PM 3/7/2009, Ben B wrote:
dont ask me, this is how it was when i found it. So if I compile my own
kernel, it should have them clocksources?
What's the motherboard? is it a virtual machine?
On Sat, Mar 7, 2009 at 11:01 AM, Gary Stanley
g...@velocity-servers.netwrote:
At 12:47 PM 3
At 10:33 PM 3/6/2009, The Universes wrote:
I'm running 4 32 man TF2 servers at 500fps (66 tick) and I'm wondering
if having kernel at 1000HZ (not tickless) with PREEMPT is necessary?
Is that putting unnecessary strain on my CPU (Q9300)?
My second question is if you guys use HPET as your
At 10:33 PM 3/6/2009, The Universes wrote:
I'm running 4 32 man TF2 servers at 500fps (66 tick) and I'm wondering
if having kernel at 1000HZ (not tickless) with PREEMPT is necessary?
Is that putting unnecessary strain on my CPU (Q9300)?
My second question is if you guys use HPET as your
At 03:23 AM 2/24/2009, Kveri wrote:
guys I don't see any point running a server with pingboost (1,2 or 3)
with 1000Hz kernel, why? It's already boosted, so why using pingboost?
Kveri
Different methods to service the timers that drive it. 1 uses an
alarm() call, 2 uses a select()
and 3 uses a
At 03:23 AM 2/24/2009, Kveri wrote:
guys I don't see any point running a server with pingboost (1,2 or 3)
with 1000Hz kernel, why? It's already boosted, so why using pingboost?
Kveri
Different methods to service the timers that drive it. 1 uses an
alarm() call, 2 uses a select()
and 3 uses a
At 02:12 PM 2/24/2009, Steven Hartland wrote:
Make sure that ALL power management option in the machine BIOS
are disabled. If you have any Power management enabled then
your server will be totally unpredictable, causing lag.
Regards
Steve
- Original Message -
From: kERPLUNK
At 05:19 PM 2/24/2009, Matthias Bleile wrote:
which clocksource do you use?
It's not the clocksource. It's power management stuff enabled in the
BIOS, or maybe some type
of chipset errata. Or it's something enabled in the kernel, like CPU speed.
At 02:12 PM 2/24/2009, Steven Hartland wrote:
Make sure that ALL power management option in the machine BIOS
are disabled. If you have any Power management enabled then
your server will be totally unpredictable, causing lag.
Regards
Steve
- Original Message -
From: kERPLUNK
At 05:19 PM 2/24/2009, Matthias Bleile wrote:
which clocksource do you use?
It's not the clocksource. It's power management stuff enabled in the
BIOS, or maybe some type
of chipset errata. Or it's something enabled in the kernel, like CPU speed.
At 03:47 PM 2/23/2009, A. Eijkhoudt wrote:
not under esxi it doesn't..:)
We're using an ESXi cluster, so good to know ;)
But seriously, I still think the CPU usage is ridiculous for TF2 now. I've
fiddled a lot with HPET/Hz/etc. settings under Linux and I can't seem to
get it below these values
At 10:57 AM 2/22/2009, Christopher Szabo wrote:
Hi!
I'm running a Counter-strike 1.6 publicserver (32 slots) and I have
a problem with my settings.
Everybody has a high choke, 30-100. I have tried many rate settings
but none is working well.
The connection isn't the problem because it's a
At 03:47 PM 2/23/2009, A. Eijkhoudt wrote:
not under esxi it doesn't..:)
We're using an ESXi cluster, so good to know ;)
But seriously, I still think the CPU usage is ridiculous for TF2 now. I've
fiddled a lot with HPET/Hz/etc. settings under Linux and I can't seem to
get it below these values
At 10:57 AM 2/22/2009, Christopher Szabo wrote:
Hi!
I'm running a Counter-strike 1.6 publicserver (32 slots) and I have
a problem with my settings.
Everybody has a high choke, 30-100. I have tried many rate settings
but none is working well.
The connection isn't the problem because it's a
At 01:41 AM 2/19/2009, kERPLUNK wrote:
Hi, i have a serious problem with my players ping/lag.
I think, that comes from my server-side fps fluctuation.
Is there some way to make it really really stable?
No.
I tried a lot of things, like:
- Changing sys_ticrate to 0, 100, 120, 333, 500, 1000,
At 01:41 AM 2/19/2009, kERPLUNK wrote:
Hi, i have a serious problem with my players ping/lag.
I think, that comes from my server-side fps fluctuation.
Is there some way to make it really really stable?
No.
I tried a lot of things, like:
- Changing sys_ticrate to 0, 100, 120, 333, 500, 1000,
Problem #1
I need some information about HLDS Counter-Strike 1.6 servers. I'm not
certain what i should use in the start-command regarding the ticrate,
pingboost and what is tos? I'm running two publicservers on each dedicated
server. Both servers has 1000Hz in kernel.
Machine #1
Problem #1
I need some information about HLDS Counter-Strike 1.6 servers. I'm not
certain what i should use in the start-command regarding the ticrate,
pingboost and what is tos? I'm running two publicservers on each dedicated
server. Both servers has 1000Hz in kernel.
Machine #1
At 01:13 PM 2/10/2009, rav...@arkanox.net wrote:
I'm registered to a bunch of mailing list related to server software
such as DSPAM, Drupal, WordPress, DirectAdmin, etc. Each email for a
mailing list is dropped in a specific folder in my inbox using
filters. It would be a nightmare for me to
At 01:13 PM 2/10/2009, rav...@arkanox.net wrote:
I'm registered to a bunch of mailing list related to server software
such as DSPAM, Drupal, WordPress, DirectAdmin, etc. Each email for a
mailing list is dropped in a specific folder in my inbox using
filters. It would be a nightmare for me to
At 03:00 AM 2/9/2009, Mikael Pedersen wrote:
Now that we have opened the can of worms, maybe someone can explain my
weird top output. I have an Intel C2D E7200 equipped machine, which
runs a TF2 server and a L4D server, L4D on the first core and TF2 on
the other. Here is my top output:
top -
At 03:00 AM 2/9/2009, Mikael Pedersen wrote:
Now that we have opened the can of worms, maybe someone can explain my
weird top output. I have an Intel C2D E7200 equipped machine, which
runs a TF2 server and a L4D server, L4D on the first core and TF2 on
the other. Here is my top output:
top -
At 07:37 PM 2/8/2009, Nephyrin Zey wrote:
TF2 takes considerably more CPU than when i first started hosting servers,
easily.
Same with source :(
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
At 07:37 PM 2/8/2009, Nephyrin Zey wrote:
TF2 takes considerably more CPU than when i first started hosting servers,
easily.
Same with source :(
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
At 08:19 PM 2/8/2009, A. Eijkhoudt wrote:
Gary Stanley wrote:
At 07:37 PM 2/8/2009, Nephyrin Zey wrote:
TF2 takes considerably more CPU than when i first started hosting servers,
easily.
Same with source :(
Surely these binaries can be re-optimized at some point, because the
increase
At 08:19 PM 2/8/2009, A. Eijkhoudt wrote:
Gary Stanley wrote:
At 07:37 PM 2/8/2009, Nephyrin Zey wrote:
TF2 takes considerably more CPU than when i first started hosting servers,
easily.
Same with source :(
Surely these binaries can be re-optimized at some point, because the
increase
At 08:49 PM 2/8/2009, A. Eijkhoudt wrote:
Gary Stanley wrote:
- VAC2
- They don't do profiling
- Compiler bugs (see the thread about cpu usage + fPIC and clobbering
registers)
- Expensive locking
- Un-optimized loops (re: compiler bugs)
- Complex/long code paths that can be optimize
At 08:49 PM 2/8/2009, A. Eijkhoudt wrote:
Gary Stanley wrote:
- VAC2
- They don't do profiling
- Compiler bugs (see the thread about cpu usage + fPIC and clobbering
registers)
- Expensive locking
- Un-optimized loops (re: compiler bugs)
- Complex/long code paths that can be optimize
At 09:09 PM 2/8/2009, Ben wrote:
Gary Stanley wrote:
We have a better chance of Duke Nukem Forever being released. We've
also been begging
for AMD64 server binaries (again?) (for re-release?) for AGES.
Should be thankful that Valve even have the Linux server support that
they do (which
At 09:09 PM 2/8/2009, Ben wrote:
Gary Stanley wrote:
We have a better chance of Duke Nukem Forever being released. We've
also been begging
for AMD64 server binaries (again?) (for re-release?) for AGES.
Should be thankful that Valve even have the Linux server support that
they do (which
At 10:25 PM 2/8/2009, you wrote:
Don't get me started guys!
Suspects (might have no improvement, might be all the difference in the
world):
- Their 'all in one binary' means that aside from the math functions they
specifically program, 90% of code is generated for a featureless i486
platform.
At 10:25 PM 2/8/2009, you wrote:
Don't get me started guys!
Suspects (might have no improvement, might be all the difference in the
world):
- Their 'all in one binary' means that aside from the math functions they
specifically program, 90% of code is generated for a featureless i486
platform.
At 08:40 PM 1/27/2009, Nephyrin Zey wrote:
Hey Chris,
Do the l4d binaries still use -fPIC? I would be curious to see what
kind of a performance impact this might be having, i know various
performance oriented libraries (such as nvidia's OpenGL library, mpeg
decoders) opt to create non-PIC shared
At 08:40 PM 1/27/2009, Nephyrin Zey wrote:
Hey Chris,
Do the l4d binaries still use -fPIC? I would be curious to see what
kind of a performance impact this might be having, i know various
performance oriented libraries (such as nvidia's OpenGL library, mpeg
decoders) opt to create non-PIC shared
At 04:23 AM 1/28/2009, J.Miribel wrote:
Same here. acpi=off as kernel param and I'll have only one core showing up.
You can use acpi=ht to turn acpi code on only for multi core functions,
but I still get some FPS drops..
Only thing you need to do is uncheck the ACPI Processor, you don't
really
At 04:23 AM 1/28/2009, J.Miribel wrote:
Same here. acpi=off as kernel param and I'll have only one core showing up.
You can use acpi=ht to turn acpi code on only for multi core functions,
but I still get some FPS drops..
Only thing you need to do is uncheck the ACPI Processor, you don't
really
At 09:46 PM 1/24/2009, Chris Green wrote:
The first thing I can think of that you could do is rewrite the time
function to cache time and only update it every so often.. I have a
proof of concept here. It reduces
syscall time by 40% for HZ=1000 for gettimeofday/clock_gettime.
The left4dead
At 09:46 PM 1/24/2009, Chris Green wrote:
The first thing I can think of that you could do is rewrite the time
function to cache time and only update it every so often.. I have a
proof of concept here. It reduces
syscall time by 40% for HZ=1000 for gettimeofday/clock_gettime.
The left4dead
not to use all that power the x64 processors offer?
Kveri
Joseph Laws wrote:
Yes, I'm very interested in the status of the AMD64 binaries as well.
Gary Stanley wrote:
What ever happened to AMD64, valve?
The AMD64 binaries allowed hosters to use additional gpr's on the
stack
not to use all that power the x64 processors offer?
Kveri
Joseph Laws wrote:
Yes, I'm very interested in the status of the AMD64 binaries as well.
Gary Stanley wrote:
What ever happened to AMD64, valve?
The AMD64 binaries allowed hosters to use additional gpr's on the
stack
What ever happened to AMD64, valve?
The AMD64 binaries allowed hosters to use additional gpr's on the
stack.. also there are other benefits, native vsyscalls/vdso, no more
memory segmentation, high memory support (no PAE)
Can you re-release them? I found the older binaries performed
What ever happened to AMD64, valve?
The AMD64 binaries allowed hosters to use additional gpr's on the
stack.. also there are other benefits, native vsyscalls/vdso, no more
memory segmentation, high memory support (no PAE)
Can you re-release them? I found the older binaries performed
At 04:04 PM 1/20/2009, Matthias Bleile wrote:
Hi everybody.
So i wasted 1 week of free time and i still wasn't able to find out..
We got a Intel Quad Core and 4 hlds Servers.
Each server is assigned to 1 CPU using taskset.
The Server assigned to CPU0 gives excellent performance, as seen here:
At 04:04 PM 1/20/2009, Matthias Bleile wrote:
Hi everybody.
So i wasted 1 week of free time and i still wasn't able to find out..
We got a Intel Quad Core and 4 hlds Servers.
Each server is assigned to 1 CPU using taskset.
The Server assigned to CPU0 gives excellent performance, as seen here:
?
On Tue, Jan 20, 2009 at 4:23 PM, Gary Stanley
g...@velocity-servers.netwrote:
At 04:04 PM 1/20/2009, Matthias Bleile wrote:
Hi everybody.
So i wasted 1 week of free time and i still wasn't able to find out..
We got a Intel Quad Core and 4 hlds Servers.
Each server is assigned to 1 CPU
?
On Tue, Jan 20, 2009 at 4:23 PM, Gary Stanley
g...@velocity-servers.netwrote:
At 04:04 PM 1/20/2009, Matthias Bleile wrote:
Hi everybody.
So i wasted 1 week of free time and i still wasn't able to find out..
We got a Intel Quad Core and 4 hlds Servers.
Each server is assigned to 1 CPU
At 02:21 PM 1/16/2009, Peter Lindblom wrote:
Hi !
isnt running a amd binary stupid when having a Intel cpu ?? maybe Im wrong
here but I wouldnt ..
and I hope you know that 64-bit OS and 32-bit game source isnt that 100%
good ..
Peter
That statement is ridiculous. All 32bit syscalls are
At 02:21 PM 1/16/2009, Peter Lindblom wrote:
Hi !
isnt running a amd binary stupid when having a Intel cpu ?? maybe Im wrong
here but I wouldnt ..
and I hope you know that 64-bit OS and 32-bit game source isnt that 100%
good ..
Peter
That statement is ridiculous. All 32bit syscalls are
At 08:29 PM 1/15/2009, Bruce Potter wrote:
Howdy,
Quick (but I'm sure not simple) question regarding processor choice.
I've been running some TF2 servers (and now L4D servers as well) on an
AMD-based server with CentOS for a while. Currently have 2 x dual
core 2.8GHz previous-gen Opterons in the
At 08:29 PM 1/15/2009, Bruce Potter wrote:
Howdy,
Quick (but I'm sure not simple) question regarding processor choice.
I've been running some TF2 servers (and now L4D servers as well) on an
AMD-based server with CentOS for a while. Currently have 2 x dual
core 2.8GHz previous-gen Opterons in the
At 03:17 PM 1/7/2009, Ryan Burke wrote:
My educated guess is that they are using non-blocking calls to recvfrom
using the MSG_DONTWAIT flag. Normally recv and recvfrom will block until
data is received. This stops the flow of the program unless it is
multi-threaded with on thread dealing with
At 03:17 PM 1/7/2009, Ryan Burke wrote:
My educated guess is that they are using non-blocking calls to recvfrom
using the MSG_DONTWAIT flag. Normally recv and recvfrom will block until
data is received. This stops the flow of the program unless it is
multi-threaded with on thread dealing with
. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org
http://leaf.dragonflybsd.org/~gary
There currently are 7 different ways to get time from a computer.
All of them can't agree on how long a second is supposed to be -Me
. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org
http://leaf.dragonflybsd.org/~gary
There currently are 7 different ways to get time from a computer.
All of them can't agree on how long a second is supposed to be -Me
68.142.72.250:27011
Adding master server 72.165.61.189:27011
*status*
hostname: Thetu.bz
version : 1.0.4.4/14 3691 insecure (secure mode enabled, disconnected from
Steam3)
udp/ip : 127.0.0.1:27015
map : ctf_2fort at: 0 x, 0 y, 0 z
players : 0 (24 max)
Try using -ip
G. Monk Stanley
gary
68.142.72.250:27011
Adding master server 72.165.61.189:27011
*status*
hostname: Thetu.bz
version : 1.0.4.4/14 3691 insecure (secure mode enabled, disconnected from
Steam3)
udp/ip : 127.0.0.1:27015
map : ctf_2fort at: 0 x, 0 y, 0 z
players : 0 (24 max)
Try using -ip
G. Monk Stanley
gary
, etc :)
G. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org
http://leaf.dragonflybsd.org/~gary
There currently are 7 different ways to get time from a computer.
All of them can't agree on how long a second is supposed to be -Me
, etc :)
G. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org
http://leaf.dragonflybsd.org/~gary
There currently are 7 different ways to get time from a computer.
All of them can't agree on how long a second is supposed to be -Me
with info
(ktrace data) and I'll write some patches to get committed upstream
to RELENG_7/HEAD
G. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org
http://leaf.dragonflybsd.org/~gary
There currently are 7 different ways to get time from a computer.
All of them can't agree
with info
(ktrace data) and I'll write some patches to get committed upstream
to RELENG_7/HEAD
G. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org
http://leaf.dragonflybsd.org/~gary
There currently are 7 different ways to get time from a computer.
All of them can't agree
At 12:11 PM 11/22/2008, Joseph Love wrote:
I had a machine with 7.0 release (before upgrading it to 7.1) and when
running the l4d server, it would spit out a ton of messages to the
console that looked like this:
linux_sys_futex: unknown op 128
linux_sys_futex: unknown op 129
I have a patch for
At 12:11 PM 11/22/2008, Joseph Love wrote:
I had a machine with 7.0 release (before upgrading it to 7.1) and when
running the l4d server, it would spit out a ton of messages to the
console that looked like this:
linux_sys_futex: unknown op 128
linux_sys_futex: unknown op 129
I have a patch for
it all the time, no matter what. Only way to do is
is to make gettimeofday coarse based. That will round the usec field
to the last stored value, IIRC.
G. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org
http://leaf.dragonflybsd.org/~gary
There currently are 7
it all the time, no matter what. Only way to do is
is to make gettimeofday coarse based. That will round the usec field
to the last stored value, IIRC.
G. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org
http://leaf.dragonflybsd.org/~gary
There currently are 7
At 09:38 AM 11/17/2008, Saint K. wrote:
Hi,
I recently installed Debian Lenny on a dual Qcore Xeon (new onces) 2.33GHz
machine. My issue currently is that an empty TF2 server uses already around
8% CPU load on a single core. With 5 players this rises to 25% already, and
when coming close to 20
At 09:38 AM 11/17/2008, Saint K. wrote:
Hi,
I recently installed Debian Lenny on a dual Qcore Xeon (new onces) 2.33GHz
machine. My issue currently is that an empty TF2 server uses already around
8% CPU load on a single core. With 5 players this rises to 25% already, and
when coming close to 20
At 02:00 AM 11/13/2008, John wrote:
Gary:
With -pingboost 2, HL1 actually uses select() for its delays.
-pingboost 2 uses alarm(), -pingboost 1 uses select()
I was careful to check this before I originally posted; what I said about
was accurate, as least at the OS level. You can confirm
At 02:00 AM 11/13/2008, John wrote:
Gary:
With -pingboost 2, HL1 actually uses select() for its delays.
-pingboost 2 uses alarm(), -pingboost 1 uses select()
I was careful to check this before I originally posted; what I said about
was accurate, as least at the OS level. You can confirm
919.96 10
You're never going to get 1000 all the time, no matter who says what.
from the usleep() man page:
BUGS
Probably not accurate on many machines down to the
microsecond. Count on precision only to -4 or maybe -5.
G. Monk Stanley
gary at summit-servers dot com | gary
919.96 10
You're never going to get 1000 all the time, no matter who says what.
from the usleep() man page:
BUGS
Probably not accurate on many machines down to the
microsecond. Count on precision only to -4 or maybe -5.
G. Monk Stanley
gary at summit-servers dot com | gary
At 05:52 PM 11/12/2008, you wrote:
You're never going to get 1000 all the time, no matter who says what.
from the usleep() man page:
BUGS
Probably not accurate on many machines down to the
microsecond. Count on precision only to -4 or maybe -5.
It is not possible to achieve a
At 05:52 PM 11/12/2008, you wrote:
You're never going to get 1000 all the time, no matter who says what.
from the usleep() man page:
BUGS
Probably not accurate on many machines down to the
microsecond. Count on precision only to -4 or maybe -5.
It is not possible to achieve a
At 11:21 PM 11/12/2008, Guy Watkins wrote:
} -Original Message-
} From: [EMAIL PROTECTED] [mailto:hlds_linux-
} [EMAIL PROTECTED] On Behalf Of Gary Stanley
} Sent: Wednesday, November 12, 2008 10:16 PM
} To: Half-Life dedicated Linux server mailing list
} Subject: Re: [hlds_linux
At 11:21 PM 11/12/2008, Guy Watkins wrote:
} -Original Message-
} From: [EMAIL PROTECTED] [mailto:hlds_linux-
} [EMAIL PROTECTED] On Behalf Of Gary Stanley
} Sent: Wednesday, November 12, 2008 10:16 PM
} To: Half-Life dedicated Linux server mailing list
} Subject: Re: [hlds_linux
, scheduler latency, and a bunch of other things
cause the drops.
G. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org
http://leaf.dragonflybsd.org/~gary
There currently are 7 different ways to get time from a computer.
All of them can't agree on how long a second is supposed
, scheduler latency, and a bunch of other things
cause the drops.
G. Monk Stanley
gary at summit-servers dot com | gary at DragonflyBSD dot org
http://leaf.dragonflybsd.org/~gary
There currently are 7 different ways to get time from a computer.
All of them can't agree on how long a second is supposed
At 08:40 PM 10/30/2008, Chris Green wrote:
The left4dead dedicated linux servers are built using a newer
version of the gcc compiler, and are optimized for higher end
processors and built with more optimization turned on. In addition,
some linux-specific bottlenecks (gettimeofday, way too many
At 08:40 PM 10/30/2008, Chris Green wrote:
The left4dead dedicated linux servers are built using a newer
version of the gcc compiler, and are optimized for higher end
processors and built with more optimization turned on. In addition,
some linux-specific bottlenecks (gettimeofday, way too many
At 07:04 PM 10/27/2008, kama wrote:
On Sat, 25 Oct 2008, Gary Stanley wrote:
At 04:29 AM 10/25/2008, kama wrote:
Hi!
The server crashes under FreeBSD 6.x while using SMP. I have not tried any
other version of FreeBSD. Without SMP they dont crash, at least not as
often
101 - 200 of 413 matches
Mail list logo
