[hlds_linux] The 1000fps problem
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 no drops, not single one (sorry for arguing too much in November, I was wrong). But there are some issues I encountered. 1. I think htop, top and others are not showing CPU usage correct. 0% of all cores when 10 fully loaded 12 player 1000fps servers with pingboost 3, it's funny. 2. segfaults, sometimes it just hangs. I don't know why, debug says nothing, but I was able to minimize if not eliminate those segfaults (but I don't know how, yet). I'm using Gentoo 64bit, but it should be almost same for any other Linux disto. Here is the kernel config: http://kveri.com/1000fps_kernel_config If you're willing to try it, I'll be glad. First of all: it's vanilla kernel 2.6.26.8 (kernel.org) NO RT patch just this config and these few steps: include/asm/param.h (or any other param.h under include dir in kernel root): #ifdef __KERNEL__ # define HZ 2100/* Internal kernel timer frequency */ # define USER_HZ1 /* some user interfaces are */ # define CLOCKS_PER_SEC 1 /* in ticks like times() */ #endif include/linux/jiffies.h: change #elif HZ = 1536 HZ 3072 # define SHIFT_HZ 11 to #elif HZ = 1536 HZ 3072 # define SHIFT_HZ 1 and change # define LATCH ((CLOCK_TICK_RATE + HZ / 2) / HZ) / * For divider * / to #define LATCH ((CLOCK_TICK_RATE + 2200/2) / 2200) /* For divider */ kernel/timeconst.pl: change (comment it) print #if HZ != $hz\n; print #error \kernel/timeconst.h has the wrong HZ value!\\n; print #endif\n; to # print #if HZ != $hz\n; # print #error \kernel/timeconst.h has the wrong HZ value!\\n; # print #endif\n; and execute: kernel/timeconst.pl 2100 kernel/timeconst.h *this step is probably why my CPU stats aren't working correctly anymore*. Next, use TSC clocksource, for grub: add clocksource=tsc to your kernel command line in menu.lst/grub.conf for lilo: add append=clocksource=TSC to your /etc/lilo.conf, under this kernel image and execute /sbin/lilo. I didn't try hpet, but it probably worth trying, as well as changing CPU I/O schedulers in kernel. That config I posted above is heavily optimized for my hardware, but it can be optimized further, make sure it suits your hardware. My experiences: with RT patch I was able to get around 900 fps, but not as stable as without RT patch. I tried 4 different kernels (all 2.6.26.8): 1. without RT without HZ modification, without RT with HZ, with RT without HZ and with RT with HZ. Best one was without RT with HZ (that one I was discribing). Pingboost: I tried anything from pingboost 0 (no -pingboost) to pingboost 3, nothing except pingboost 3 gave me 1000fps constant. Sys_ticrate: well, it looks like anything more than HZ in kernel (2100) makes game accelerated, so I'm using sys_ticrate 2000 for now. If anyone tries this method/config please let me know what are your results, I'll be very interested. That HZ guide was from: http://www.howto-cs16-root.de/gen_kernel.htm some other tips: http://wiki.fragaholics.de/index.php/EN:Linux_Kernel_Optimization Please, anyone who has experience in this, share it too. Thanks Kveri -- Tato sprava bola prehladana na vyskyt virusov a nebezpecneho obsahu antivirovym systemom a zda sa byt cista. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
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 no drops, not single one (sorry for arguing too much in November, I was wrong). But there are some issues I encountered. 1. I think htop, top and others are not showing CPU usage correct. 0% of all cores when 10 fully loaded 12 player 1000fps servers with pingboost 3, it's funny. Because you butchered the kernel timecounter code with fictitious values. No matter what runs on the machine, you will now never get accurate usage values. 2. segfaults, sometimes it just hangs. I don't know why, debug says nothing, but I was able to minimize if not eliminate those segfaults (but I don't know how, yet). Another side effect of the above. It could also be because your system clock is drifting backwards due to you altering SHIFT_HZ. I'm using Gentoo 64bit, but it should be almost same for any other Linux disto. Here is the kernel config: http://kveri.com/1000fps_kernel_config If you're willing to try it, I'll be glad. First of all: it's vanilla kernel 2.6.26.8 (kernel.org) NO RT patch just this config and these few steps: include/asm/param.h (or any other param.h under include dir in kernel root): #ifdef __KERNEL__ # define HZ 2100/* Internal kernel timer frequency */ # define USER_HZ1 /* some user interfaces are */ # define CLOCKS_PER_SEC 1 /* in ticks like times() */ #endif USER_HZ always runs at usually whatever HZ is, IIRC. Or maybe it's whatever the PIT is running at. include/linux/jiffies.h: change #elif HZ = 1536 HZ 3072 # define SHIFT_HZ 11 to #elif HZ = 1536 HZ 3072 # define SHIFT_HZ 1 Changing Shift_HZ to 1 is evil, and breaks everything else including ntp and adjtimex. and change # define LATCH ((CLOCK_TICK_RATE + HZ / 2) / HZ) / * For divider * / to #define LATCH ((CLOCK_TICK_RATE + 2200/2) / 2200) /* For divider */ kernel/timeconst.pl: change (comment it) print #if HZ != $hz\n; print #error \kernel/timeconst.h has the wrong HZ value!\\n; print #endif\n; to # print #if HZ != $hz\n; # print #error \kernel/timeconst.h has the wrong HZ value!\\n; # print #endif\n; and execute: kernel/timeconst.pl 2100 kernel/timeconst.h *this step is probably why my CPU stats aren't working correctly anymore*. Next, use TSC clocksource, for grub: add clocksource=tsc to your kernel command line in menu.lst/grub.conf for lilo: add append=clocksource=TSC to your /etc/lilo.conf, under this kernel image and execute /sbin/lilo. TSC is better to read, but use HPET if you can. I didn't try hpet, but it probably worth trying, as well as changing CPU I/O schedulers in kernel. There are no such things are CPU I/O Schedulers. You are talking about disk I/O schedulers. That config I posted above is heavily optimized for my hardware, but it can be optimized further, make sure it suits your hardware. I wouldn't suggest it's optimized for your hardware, it's a generic config that is altered. My experiences: with RT patch I was able to get around 900 fps, but not as stable as without RT patch. I tried 4 different kernels (all 2.6.26.8): 1. without RT without HZ modification, without RT with HZ, with RT without HZ and with RT with HZ. Best one was without RT with HZ (that one I was discribing). RT patches are better due to the fact they reduce kernel latency. Personally, RT patches are far superior for things like precise nanosleep()/tsleep()/usleep() Pingboost: I tried anything from pingboost 0 (no -pingboost) to pingboost 3, nothing except pingboost 3 gave me 1000fps constant. Pingboost 1/2 use either an alarm() or select(). Alarm() and Select() do NOT use hrtimers, therefor that is why you are not seeing the behavior you think you are seeing. Select() uses jiffies as a clocksource, and jiffies provides very coarse resolution, about the same as the interrupt timer. Sys_ticrate: well, it looks like anything more than HZ in kernel (2100) makes game accelerated, so I'm using sys_ticrate 2000 for now. If anyone tries this method/config please let me know what are your results, I'll be very interested. That HZ guide was from: http://www.howto-cs16-root.de/gen_kernel.htm some other tips: http://wiki.fragaholics.de/index.php/EN:Linux_Kernel_Optimization Running 2100 interrupts a second does nothing but cause cacheline pingpongs due to excessive clock interrupts. IIRC, linux stops updating syscalls like gtod if HZ = 1024, so 2100 is useless. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Hi ! I'm trying to get a 1000FPS stable server, with pain.. I've followed many recommendations on this thread like : - RT patched kernel (from Ingo Molnar) - with and without HRT Support - Tickrate 1000 and pingboost 2 I've posted my configuration here : http://pastebin.com/f7c4dff1e here's the result : around 950 FPS Unstable. If I try Tickrate 1 I get 1000FPS some times... but not stable and with an accelerated game :( I also tried the chrt program (chrt -f -p 90 PID, tried FIFO / RR, all the same results), that works better with a Low Latency Desktop kernel than a Real Time kernel. (I fall around 500FPS with RT kernel with chrt program). I know this is a kind of business on Internet game servers, but my goal is LANs, and I have no business plans on LAN :/ Thanks for your help :) Daryl. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Same for me... What is your hardware ? Daryl a écrit : Hi ! I'm trying to get a 1000FPS stable server, with pain.. I've followed many recommendations on this thread like : - RT patched kernel (from Ingo Molnar) - with and without HRT Support - Tickrate 1000 and pingboost 2 I've posted my configuration here : http://pastebin.com/f7c4dff1e here's the result : around 950 FPS Unstable. If I try Tickrate 1 I get 1000FPS some times... but not stable and with an accelerated game :( I also tried the chrt program (chrt -f -p 90 PID, tried FIFO / RR, all the same results), that works better with a Low Latency Desktop kernel than a Real Time kernel. (I fall around 500FPS with RT kernel with chrt program). I know this is a kind of business on Internet game servers, but my goal is LANs, and I have no business plans on LAN :/ Thanks for your help :) Daryl. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
At 04:07 AM 11/17/2008, Daryl wrote: Hi ! I'm trying to get a 1000FPS stable server, with pain.. I've followed many recommendations on this thread like : - RT patched kernel (from Ingo Molnar) RT helps out with scheduler latency.. - with and without HRT Support Required at least. - Tickrate 1000 and pingboost 2 pingboost uses select() for timing of frames. the problem is that select uses jiffies resolution, not hrtimers. another thing is, FPS is measured by a couple of syscalls, namely gettimeofday() and nanosleep(). There's also a bug in the timespec_to_jiffies code where it automatically adds 3 jiffies to prevent timers firing off on all CPUs. I've posted my configuration here : http://pastebin.com/f7c4dff1e here's the result : around 950 FPS Unstable. If I try Tickrate 1 I get 1000FPS some times... but not stable and with an accelerated game :( Never going to get 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 different ways to get time from a computer. All of them can't agree on how long a second is supposed to be -Me ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
At 04:07 AM 11/17/2008, Daryl wrote: Hi ! I'm trying to get a 1000FPS stable server, with pain.. I've followed many recommendations on this thread like : - RT patched kernel (from Ingo Molnar) RT helps out with scheduler latency.. - with and without HRT Support Required at least. - Tickrate 1000 and pingboost 2 pingboost uses select() for timing of frames. the problem is that select uses jiffies resolution, not hrtimers. another thing is, FPS is measured by a couple of syscalls, namely gettimeofday() and nanosleep(). There's also a bug in the timespec_to_jiffies code where it automatically adds 3 jiffies to prevent timers firing off on all CPUs. I've posted my configuration here : http://pastebin.com/f7c4dff1e here's the result : around 950 FPS Unstable. If I try Tickrate 1 I get 1000FPS some times... but not stable and with an accelerated game :( Never going to get 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 different ways to get time from a computer. All of them can't agree on how long a second is supposed to be -Me ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Contact me if your interested in viewing a ROCK SOLID 1000 FPS server It's very possible. On Wed, Nov 12, 2008 at 7:52 AM, Kveri [EMAIL PROTECTED] wrote: There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years
Re: [hlds_linux] The 1000fps problem
Why do you need 980+ fps on public?
It's waste of resources, and no hardware can handle that.
Kveri
Faustas Buškevičius wrote / napísal(a):
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED] 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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low
jitter.
On a Core2-based machine, we typically see a stable ~982fps with -
pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will
dip
slightly. Typically with -pingboost 2, the dips are into the upper
990s.
-John
___
To unsubscribe, edit your list preferences, or view the list
archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Because he wants to. If you don't want to help him then keep your trap
shut.
Kveri wrote:
Why do you need 980+ fps on public?
It's waste of resources, and no hardware can handle that.
Kveri
Faustas Buškevičius wrote / napísal(a):
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED] 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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low
jitter.
On a Core2-based machine, we typically see a stable ~982fps with -
pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will
dip
slightly. Typically with -pingboost 2, the dips are into the upper
990s.
-John
___
To unsubscribe, edit your list preferences, or view the list
archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Yea.
Trap Shut.
On Fri, Nov 14, 2008 at 7:02 AM, Joseph Laws [EMAIL PROTECTED] wrote:
Because he wants to. If you don't want to help him then keep your trap
shut.
Kveri wrote:
Why do you need 980+ fps on public?
It's waste of resources, and no hardware can handle that.
Kveri
Faustas Buškevičius wrote / napísal(a):
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED]
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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low
jitter.
On a Core2-based machine, we typically see a stable ~982fps with -
pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will
dip
slightly. Typically with -pingboost 2, the dips are into the upper
990s.
-John
___
To unsubscribe, edit your list preferences, or view the list
archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
--
-
[EMAIL PROTECTED]
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
I think stable 980+ fps on 20 slot public with 20 players is not possible.
Kveri
Kveri wrote / napísal(a):
Why do you need 980+ fps on public?
It's waste of resources, and no hardware can handle that.
Kveri
Faustas Buškevičius wrote / napísal(a):
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED] 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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low
jitter.
On a Core2-based machine, we typically see a stable ~982fps with -
pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will
dip
slightly. Typically with -pingboost 2, the dips are into the upper
990s.
-John
___
To unsubscribe, edit your list preferences, or view the list
archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
This is the best we got it to with our configuration and pingboost 2. Output
from console stats (output from rcon stats are more consistently 1000). :
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.87 231.765059 7723 1000.00 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.05 229.285059 7723 980.39 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.02 230.855059 7723 984.25 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.26 232.515059 7723 981.35 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 141.28 232.725059 7723 956.02 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.59 234.155059 7723 505.31 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 144.81 237.625059 7723 492.85 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.50 237.015059 7723 946.07 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.39 235.565059 7723 492.37 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 141.01 227.235059 7723 975.61 20
statsDropped sway adr * mysway.net from server
Reason: Client sent 'drop'
CPU InOut Uptime Users FPSPlayers
0.00 127.23 203.215059 7723 975.61 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 126.72 197.655059 7723 928.51 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 129.82 198.305059 7723 970.87 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 136.97 213.335059 7723 948.77 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 140.09 223.735059 7723 1000.00 19
top:
PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND
16994 ndsg2 20 0 273m 256m 7608 S 38 7.8 1508:40 hlds_i686
On Fri, Nov 14, 2008 at 2:38 PM, Kveri [EMAIL PROTECTED] wrote:
I think stable 980+ fps on 20 slot public with 20 players is not possible.
Kveri
Kveri wrote / napísal(a):
Why do you need 980+ fps on public?
It's waste of resources, and no hardware can handle that.
Kveri
Faustas Buškevičius wrote / napísal(a):
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED]
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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates
Re: [hlds_linux] The 1000fps problem
I've been testing 1000fps gameservers myself and i found after long
research a kernel + patch combination that gives always 1000fps (with
pingboost2). It stays on 1000fps constantly even with many players on
server and multiple 1000fps servers on the same hardware. It didn't even
use much CPU.
I still consider the perfect kernel as some kind of business secret so
im not telling anything more than that its possible with pingboost2 +
real time patchset from /Ingo Molnár. Rest you have to figure out
yourself. =)/
kabukiUkie kirjoitti:
This is the best we got it to with our configuration and pingboost 2. Output
from console stats (output from rcon stats are more consistently 1000). :
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.87 231.765059 7723 1000.00 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.05 229.285059 7723 980.39 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.02 230.855059 7723 984.25 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.26 232.515059 7723 981.35 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 141.28 232.725059 7723 956.02 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.59 234.155059 7723 505.31 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 144.81 237.625059 7723 492.85 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.50 237.015059 7723 946.07 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.39 235.565059 7723 492.37 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 141.01 227.235059 7723 975.61 20
statsDropped sway adr * mysway.net from server
Reason: Client sent 'drop'
CPU InOut Uptime Users FPSPlayers
0.00 127.23 203.215059 7723 975.61 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 126.72 197.655059 7723 928.51 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 129.82 198.305059 7723 970.87 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 136.97 213.335059 7723 948.77 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 140.09 223.735059 7723 1000.00 19
top:
PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND
16994 ndsg2 20 0 273m 256m 7608 S 38 7.8 1508:40 hlds_i686
On Fri, Nov 14, 2008 at 2:38 PM, Kveri [EMAIL PROTECTED] wrote:
I think stable 980+ fps on 20 slot public with 20 players is not possible.
Kveri
Kveri wrote / napísal(a):
Why do you need 980+ fps on public?
It's waste of resources, and no hardware can handle that.
Kveri
Faustas Buškevičius wrote / napísal(a):
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED]
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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause()
Re: [hlds_linux] The 1000fps problem
2008/11/14 Valtteri Kiviniemi [EMAIL PROTECTED]
I've been testing 1000fps gameservers myself and i found after long
research a kernel + patch combination that gives always 1000fps (with
pingboost2). It stays on 1000fps constantly even with many players on
server and multiple 1000fps servers on the same hardware. It didn't even
use much CPU.
I still consider the perfect kernel as some kind of business secret so
im not telling anything more than that its possible with pingboost2 +
real time patchset from /Ingo Molnár. Rest you have to figure out
yourself. =)/
kabukiUkie kirjoitti:
This is the best we got it to with our configuration and pingboost 2.
Output
from console stats (output from rcon stats are more consistently 1000).
:
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.87 231.765059 7723 1000.00 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.05 229.285059 7723 980.39 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.02 230.855059 7723 984.25 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.26 232.515059 7723 981.35 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 141.28 232.725059 7723 956.02 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.59 234.155059 7723 505.31 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 144.81 237.625059 7723 492.85 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.50 237.015059 7723 946.07 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.39 235.565059 7723 492.37 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 141.01 227.235059 7723 975.61 20
statsDropped sway adr * mysway.net from server
Reason: Client sent 'drop'
CPU InOut Uptime Users FPSPlayers
0.00 127.23 203.215059 7723 975.61 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 126.72 197.655059 7723 928.51 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 129.82 198.305059 7723 970.87 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 136.97 213.335059 7723 948.77 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 140.09 223.735059 7723 1000.00 19
top:
PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND
16994 ndsg2 20 0 273m 256m 7608 S 38 7.8 1508:40 hlds_i686
On Fri, Nov 14, 2008 at 2:38 PM, Kveri [EMAIL PROTECTED] wrote:
I think stable 980+ fps on 20 slot public with 20 players is not
possible.
Kveri
Kveri wrote / napísal(a):
Why do you need 980+ fps on public?
It's waste of resources, and no hardware can handle that.
Kveri
Faustas Buškevičius wrote / napísal(a):
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED]
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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL)
Re: [hlds_linux] The 1000fps problem
ah crap... you too?!...
Ferenc Kovacs schreef:
2008/11/14 Valtteri Kiviniemi [EMAIL PROTECTED]
I've been testing 1000fps gameservers myself and i found after long
research a kernel + patch combination that gives always 1000fps (with
pingboost2). It stays on 1000fps constantly even with many players on
server and multiple 1000fps servers on the same hardware. It didn't even
use much CPU.
I still consider the perfect kernel as some kind of business secret so
im not telling anything more than that its possible with pingboost2 +
real time patchset from /Ingo Molnár. Rest you have to figure out
yourself. =)/
kabukiUkie kirjoitti:
This is the best we got it to with our configuration and pingboost 2.
Output
from console stats (output from rcon stats are more consistently 1000).
:
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.87 231.765059 7723 1000.00 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.05 229.285059 7723 980.39 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.02 230.855059 7723 984.25 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.26 232.515059 7723 981.35 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 141.28 232.725059 7723 956.02 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.59 234.155059 7723 505.31 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 144.81 237.625059 7723 492.85 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.50 237.015059 7723 946.07 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.39 235.565059 7723 492.37 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 141.01 227.235059 7723 975.61 20
statsDropped sway adr * mysway.net from server
Reason: Client sent 'drop'
CPU InOut Uptime Users FPSPlayers
0.00 127.23 203.215059 7723 975.61 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 126.72 197.655059 7723 928.51 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 129.82 198.305059 7723 970.87 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 136.97 213.335059 7723 948.77 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 140.09 223.735059 7723 1000.00 19
top:
PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND
16994 ndsg2 20 0 273m 256m 7608 S 38 7.8 1508:40 hlds_i686
On Fri, Nov 14, 2008 at 2:38 PM, Kveri [EMAIL PROTECTED] wrote:
I think stable 980+ fps on 20 slot public with 20 players is not
possible.
Kveri
Kveri wrote / napísal(a):
Why do you need 980+ fps on public?
It's waste of resources, and no hardware can handle that.
Kveri
Faustas Buškevičius wrote / napísal(a):
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED]
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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause()
Re: [hlds_linux] The 1000fps problem
2008/11/15 Nander Paardekooper [EMAIL PROTECTED]
ah crap... you too?!...
Ferenc Kovacs schreef:
2008/11/14 Valtteri Kiviniemi [EMAIL PROTECTED]
I've been testing 1000fps gameservers myself and i found after long
research a kernel + patch combination that gives always 1000fps (with
pingboost2). It stays on 1000fps constantly even with many players on
server and multiple 1000fps servers on the same hardware. It didn't even
use much CPU.
I still consider the perfect kernel as some kind of business secret so
im not telling anything more than that its possible with pingboost2 +
real time patchset from /Ingo Molnár. Rest you have to figure out
yourself. =)/
kabukiUkie kirjoitti:
This is the best we got it to with our configuration and pingboost 2.
Output
from console stats (output from rcon stats are more consistently 1000).
:
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.87 231.765059 7723 1000.00 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.05 229.285059 7723 980.39 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.02 230.855059 7723 984.25 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.26 232.515059 7723 981.35 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 141.28 232.725059 7723 956.02 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.59 234.155059 7723 505.31 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 144.81 237.625059 7723 492.85 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 143.50 237.015059 7723 946.07 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 142.39 235.565059 7723 492.37 20
stats
CPU InOut Uptime Users FPSPlayers
0.00 141.01 227.235059 7723 975.61 20
statsDropped sway adr * mysway.net from server
Reason: Client sent 'drop'
CPU InOut Uptime Users FPSPlayers
0.00 127.23 203.215059 7723 975.61 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 126.72 197.655059 7723 928.51 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 129.82 198.305059 7723 970.87 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 136.97 213.335059 7723 948.77 19
stats
CPU InOut Uptime Users FPSPlayers
0.00 140.09 223.735059 7723 1000.00 19
top:
PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND
16994 ndsg2 20 0 273m 256m 7608 S 38 7.8 1508:40 hlds_i686
On Fri, Nov 14, 2008 at 2:38 PM, Kveri [EMAIL PROTECTED] wrote:
I think stable 980+ fps on 20 slot public with 20 players is not
possible.
Kveri
Kveri wrote / napísal(a):
Why do you need 980+ fps on public?
It's waste of resources, and no hardware can handle that.
Kveri
Faustas Buškevičius wrote / napísal(a):
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED]
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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1
EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1
EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
Re: [hlds_linux] The 1000fps problem
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 this with strace.
I see output like this for -pingboost 2:
Indeed. I didn't bother to profile it. I was going on the original
email and what's in the sdk..
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}}, NULL) = 0
pause() = ? ERESTARTNOHAND (To be restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910, [ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}}, NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low jitter.
On a Core2-based machine, we typically see a stable ~982fps with -pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will dip
slightly. Typically with -pingboost 2, the dips are into the upper 990s.
-John
With no pingboosting on a spare, hacked up Linux machine (gentoo), I
get 1000/999fps all the time with HPET and TSC as a clocksource.. Of
course, I put a busy wait patch in
nanosleep to allow for more accurate nanosleep delays (old behavior),
plus a module I use to convert the very broken, jittery gettimeofday
and usleep with clock_gettime(CLOCK_MONO..) and clock_nanosleep().
I've also disabled SMI's on this machine to prevent pipeline stalls
due to SMI interrupts firing off every few uS or so.. but that's
pretty much an overkill :P
It looks like no pingboosting uses pingboost 2 (select boost..)
CPU InOut Uptime Users FPSPlayers
0.00 0.00 0.00 5 0 999.00 0
stats
CPU InOut Uptime Users FPSPlayers
0.00 0.00 0.00 5 0 1000.00 0
stats
CPU InOut Uptime Users FPSPlayers
0.00 0.00 0.00 5 0 1000.00 0
stats
CPU InOut Uptime Users FPSPlayers
0.00 0.00 0.00 5 0 999.00 0
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
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 this with strace.
I see output like this for -pingboost 2:
Indeed. I didn't bother to profile it. I was going on the original
email and what's in the sdk..
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}}, NULL) = 0
pause() = ? ERESTARTNOHAND (To be restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910, [ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}}, NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low jitter.
On a Core2-based machine, we typically see a stable ~982fps with -pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will dip
slightly. Typically with -pingboost 2, the dips are into the upper 990s.
-John
With no pingboosting on a spare, hacked up Linux machine (gentoo), I
get 1000/999fps all the time with HPET and TSC as a clocksource.. Of
course, I put a busy wait patch in
nanosleep to allow for more accurate nanosleep delays (old behavior),
plus a module I use to convert the very broken, jittery gettimeofday
and usleep with clock_gettime(CLOCK_MONO..) and clock_nanosleep().
I've also disabled SMI's on this machine to prevent pipeline stalls
due to SMI interrupts firing off every few uS or so.. but that's
pretty much an overkill :P
It looks like no pingboosting uses pingboost 2 (select boost..)
CPU InOut Uptime Users FPSPlayers
0.00 0.00 0.00 5 0 999.00 0
stats
CPU InOut Uptime Users FPSPlayers
0.00 0.00 0.00 5 0 1000.00 0
stats
CPU InOut Uptime Users FPSPlayers
0.00 0.00 0.00 5 0 1000.00 0
stats
CPU InOut Uptime Users FPSPlayers
0.00 0.00 0.00 5 0 999.00 0
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED] 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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low
jitter.
On a Core2-based machine, we typically see a stable ~982fps with -
pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will
dip
slightly. Typically with -pingboost 2, the dips are into the upper
990s.
-John
___
To unsubscribe, edit your list preferences, or view the list
archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED] 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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low
jitter.
On a Core2-based machine, we typically see a stable ~982fps with -
pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will
dip
slightly. Typically with -pingboost 2, the dips are into the upper
990s.
-John
___
To unsubscribe, edit your list preferences, or view the list
archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Why do you want 1000 FPS servers?
2008/11/13 Faustas Buškevičius [EMAIL PROTECTED]
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED] 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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low
jitter.
On a Core2-based machine, we typically see a stable ~982fps with -
pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will
dip
slightly. Typically with -pingboost 2, the dips are into the upper
990s.
-John
___
To unsubscribe, edit your list preferences, or view the list
archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
We belive 1000FPS servers give the best gaming/aiming experience for
pro-players.
Using 1000FPS on public server is waste of resources if you're hosting
company, but if you have gaming project and 1-2 pub servers - you make your
server be the best in the list of available public servers.
-.-.-.-.-.-.-.-.-.-.-.-
en3my
www.2Po.eu
- Original Message -
From: AnAkIn . [EMAIL PROTECTED]
To: Half-Life dedicated Linux server mailing list
hlds_linux@list.valvesoftware.com
Sent: Thursday, November 13, 2008 6:58 PM
Subject: Re: [hlds_linux] The 1000fps problem
Why do you want 1000 FPS servers?
2008/11/13 Faustas Buškevičius [EMAIL PROTECTED]
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED]
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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low
jitter.
On a Core2-based machine, we typically see a stable ~982fps with -
pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will
dip
slightly. Typically with -pingboost 2, the dips are into the upper
990s.
-John
___
To unsubscribe, edit your list preferences, or view the list
archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
You should already know that 1000 FPS server is just bullshit and that only
people that know nothing about servers buy them (a lot) so it gives more
money to GSP.
There was an article that explained it which was posted on the HLDS mailing
list a while ago (The 1000 FPS Fairy Tale).
2008/11/13 en3my [EMAIL PROTECTED]
We belive 1000FPS servers give the best gaming/aiming experience for
pro-players.
Using 1000FPS on public server is waste of resources if you're hosting
company, but if you have gaming project and 1-2 pub servers - you make your
server be the best in the list of available public servers.
-.-.-.-.-.-.-.-.-.-.-.-
en3my
www.2Po.eu
- Original Message -
From: AnAkIn . [EMAIL PROTECTED]
To: Half-Life dedicated Linux server mailing list
hlds_linux@list.valvesoftware.com
Sent: Thursday, November 13, 2008 6:58 PM
Subject: Re: [hlds_linux] The 1000fps problem
Why do you want 1000 FPS servers?
2008/11/13 Faustas Buškevičius [EMAIL PROTECTED]
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED]
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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low
jitter.
On a Core2-based machine, we typically see a stable ~982fps with -
pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will
dip
slightly. Typically with -pingboost 2, the dips are into the upper
990s.
-John
___
To unsubscribe, edit your list preferences, or view the list
archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives
Re: [hlds_linux] The 1000fps problem
Personally I can agree with that and 1000fps server never was one of
my goals. I'm after a server with _stable_ 500 fps (or 300, or
whatever number *I* set it to be at), as long as its _stable_.
Meaning, I would love to have a public server with _stable_ 300 fps,
it's just that the fps is constantly dipping to, say, 70, or rocketing
to, say, 980, neither of which is any good.
2008/11/13 AnAkIn . [EMAIL PROTECTED]:
You should already know that 1000 FPS server is just bullshit and that only
people that know nothing about servers buy them (a lot) so it gives more
money to GSP.
There was an article that explained it which was posted on the HLDS mailing
list a while ago (The 1000 FPS Fairy Tale).
2008/11/13 en3my [EMAIL PROTECTED]
We belive 1000FPS servers give the best gaming/aiming experience for
pro-players.
Using 1000FPS on public server is waste of resources if you're hosting
company, but if you have gaming project and 1-2 pub servers - you make your
server be the best in the list of available public servers.
-.-.-.-.-.-.-.-.-.-.-.-
en3my
www.2Po.eu
- Original Message -
From: AnAkIn . [EMAIL PROTECTED]
To: Half-Life dedicated Linux server mailing list
hlds_linux@list.valvesoftware.com
Sent: Thursday, November 13, 2008 6:58 PM
Subject: Re: [hlds_linux] The 1000fps problem
Why do you want 1000 FPS servers?
2008/11/13 Faustas Buškevičius [EMAIL PROTECTED]
What are the chances of sustaining 980+ fps on a public server with
20+ players and max rates ?
On Thu, Nov 13, 2008 at 1:09 PM, Kveri [EMAIL PROTECTED] wrote:
Interesting, it looks like a bug in documentation. I'll test it on
brand new dual E5335 xeon server.
Kveri
Sent from my iPhone
On 13 Nov 2008, at 08:00, John [EMAIL PROTECTED]
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 this with
strace.
I see output like this for -pingboost 2:
...
gettimeofday({1226558338, 85065}, NULL) = 0
gettimeofday({1226558338, 85091}, NULL) = 0
gettimeofday({1226558338, 85122}, NULL) = 0
gettimeofday({1226558338, 85147}, NULL) = 0
gettimeofday({1226558338, 85170}, NULL) = 0
select(1, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
gettimeofday({1226558338, 85971}, NULL) = 0
gettimeofday({1226558338, 85996}, NULL) = 0
recvfrom(5, 0xbfa3efe4, 4010, 0, 0xbfa3ff90, 0xbfa3efcc) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558338, 86058}, NULL) = 0
gettimeofday({1226558338, 86083}, NULL) = 0
gettimeofday({1226558338, 86102}, NULL) = 0
gettimeofday({1226558338, 86120}, NULL) = 0
gettimeofday({1226558338, 86161}, NULL) = 0
...
In constrast, -pingboost 1 gives output like this:
gettimeofday({1226558633, 60244}, NULL) = 0
gettimeofday({1226558633, 60272}, NULL) = 0
recvfrom(5, 0xbfb5ecb4, 4010, 0, 0xbfb5fc60, 0xbfb5ec9c) = -1 EAGAIN
(Resource temporarily unavailable)
gettimeofday({1226558633, 60340}, NULL) = 0
gettimeofday({1226558633, 60360}, NULL) = 0
gettimeofday({1226558633, 60388}, NULL) = 0
gettimeofday({1226558633, 60415}, NULL) = 0
gettimeofday({1226558633, 60442}, NULL) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
pause() = ? ERESTARTNOHAND (To be
restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
rt_sigaction(SIGALRM, {0x804a910, [ALRM], SA_RESTART}, {0x804a910,
[ALRM],
SA_RESTART}, 8) = 0
setitimer(ITIMER_REAL, {it_interval={0, 0}, it_value={0, 1000}},
NULL) = 0
sigreturn() = ? (mask now [])
select(1, [0], NULL, NULL, {0, 0}) = 0 (Timeout)
It sounds like Valve flipped the definitions of the functions since
creating
the versions you posted.
With our kernel configuration, load-balancing, etc, both -pingboost 1
and -pingboost 2 provide very stable framerates with extremely low
jitter.
On a Core2-based machine, we typically see a stable ~982fps with -
pingboost
1 and a stable 1000fps with -pingboost 2. Rarely, either method will
dip
slightly. Typically with -pingboost 2, the dips are into the upper
990s.
-John
___
To unsubscribe, edit your list preferences, or view the list
archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences, or view the list archives,
please visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
___
To unsubscribe, edit your list preferences
Re: [hlds_linux] The 1000fps problem
Maybe because your cpu cannot handle it? pingboost 3 is more performance hungry than pingboost 2. But it really doesn't matter if you have 900 or 1000 fps, you are not able to run 1000fps stable, maybe 980-1000 yes, but not 1000 stable. Btw in GSP terms, 900FPS = 1000FPS. Kveri xLnT wrote / napísal(a): Hi. Just tested one server with -pingboost 3 and +sys_ticrate 1000 on my 1000hz kernel, No Preemt, Tickless. And it was even worse then with -pingboost 2 and +sys_ticrate 1000. got about 900fps fairly stable with the pingboost 3 option. And 980fps stable with -pingboost 2 option. Wierd. Regards Faustas Buškevičius wrote: -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- - [EMAIL PROTECTED] ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http
Re: [hlds_linux] The 1000fps problem
Then game-hosting.com is faking their stats output. They have stable 1000fps and drops down to 500fps exactly. Not 501 or 499 but 500fps.. once in a while.. otherwise it's 1000fps stable. There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers
Re: [hlds_linux] The 1000fps problem
Hlds - cs 1.6 Sent from my iPhone On 12 Nov 2008, at 14:48, Joseph Laws [EMAIL PROTECTED] wrote: I'm sorry but which game are you guys going back and forth about? xLnT wrote: Then game-hosting.com is faking their stats output. They have stable 1000fps and drops down to 500fps exactly. Not 501 or 499 but 500fps.. once in a while.. otherwise it's 1000fps stable. There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I
Re: [hlds_linux] The 1000fps problem
They do, because of IRQ. Kveri en3my wrote / napísal(a): None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly
Re: [hlds_linux] The 1000fps problem
So you guys are looking for something like this? 09:59:21 stats 09:59:21 og L 11/12/2008 - 09:56:37: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:21 CPU InOut Uptime Users FPSPlayers 3.00 0.00 0.00 0 0 1000.00 0 09:59:22 stats 09:59:22 og L 11/12/2008 - 09:56:38: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:22 CPU InOut Uptime Users FPSPlayers 3.00 0.00 0.00 0 0 1000.00 0 09:59:22 stats 09:59:22 og L 11/12/2008 - 09:56:38: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:22 CPU InOut Uptime Users FPSPlayers 3.50 0.00 0.00 0 0 1000.00 0 09:59:23 stats 09:59:23 og L 11/12/2008 - 09:56:39: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:23 CPU InOut Uptime Users FPSPlayers 3.50 0.00 0.00 0 0 1000.00 0 09:59:23 stats 09:59:23 og L 11/12/2008 - 09:56:39: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:23 CPU InOut Uptime Users FPSPlayers 3.50 0.00 0.00 0 0 1000.00 0 09:59:24 stats 09:59:24 og L 11/12/2008 - 09:56:40: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:24 CPU InOut Uptime Users FPSPlayers 3.00 0.00 0.00 0 0 1000.00 0 09:59:25 stats 09:59:25 og L 11/12/2008 - 09:56:41: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:25 CPU InOut Uptime Users FPSPlayers 3.00 0.00 0.00 0 0 1000.00 0 or... 10:01:28 stats 10:01:28 CPU InOut Uptime Users FPSPlayers 4.00 0.00 0.00 0 0 500.00 0 10:01:33 stats 10:01:33 CPU InOut Uptime Users FPSPlayers 4.00 0.00 0.00 0 0 500.00 0 10:01:34 stats 10:01:34 CPU InOut Uptime Users FPSPlayers 2.00 0.00 0.00 0 0 500.00 0 10:01:34 stats 10:01:34 CPU InOut Uptime Users FPSPlayers 1.67 0.00 0.00 0 0 500.00 0 10:01:34 stats 10:01:35 CPU InOut Uptime Users FPSPlayers 1.67 0.00 0.00 0 0 500.00 0 10:01:35 stats 10:01:35 CPU InOut Uptime Users FPSPlayers 2.00 0.00 0.00 0 0 500.00 0 Kveri wrote: Hlds - cs 1.6 Sent from my iPhone On 12 Nov 2008, at 14:48, Joseph Laws [EMAIL PROTECTED] wrote: I'm sorry but which game are you guys going back and forth about? xLnT wrote: Then game-hosting.com is faking their stats output. They have stable 1000fps and drops down to 500fps exactly. Not 501 or 499 but 500fps.. once in a while.. otherwise it's 1000fps stable. There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under
Re: [hlds_linux] The 1000fps problem
The first part.. So you guys are looking for something like this? 09:59:21 stats 09:59:21 og L 11/12/2008 - 09:56:37: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:21 CPU InOut Uptime Users FPSPlayers 3.00 0.00 0.00 0 0 1000.00 0 09:59:22 stats 09:59:22 og L 11/12/2008 - 09:56:38: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:22 CPU InOut Uptime Users FPSPlayers 3.00 0.00 0.00 0 0 1000.00 0 09:59:22 stats 09:59:22 og L 11/12/2008 - 09:56:38: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:22 CPU InOut Uptime Users FPSPlayers 3.50 0.00 0.00 0 0 1000.00 0 09:59:23 stats 09:59:23 og L 11/12/2008 - 09:56:39: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:23 CPU InOut Uptime Users FPSPlayers 3.50 0.00 0.00 0 0 1000.00 0 09:59:23 stats 09:59:23 og L 11/12/2008 - 09:56:39: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:23 CPU InOut Uptime Users FPSPlayers 3.50 0.00 0.00 0 0 1000.00 0 09:59:24 stats 09:59:24 og L 11/12/2008 - 09:56:40: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:24 CPU InOut Uptime Users FPSPlayers 3.00 0.00 0.00 0 0 1000.00 0 09:59:25 stats 09:59:25 og L 11/12/2008 - 09:56:41: Rcon: rcon 1696617158 defaultrcon stats from 69.245.91.0:7130 09:59:25 CPU InOut Uptime Users FPSPlayers 3.00 0.00 0.00 0 0 1000.00 0 or... 10:01:28 stats 10:01:28 CPU InOut Uptime Users FPSPlayers 4.00 0.00 0.00 0 0 500.00 0 10:01:33 stats 10:01:33 CPU InOut Uptime Users FPSPlayers 4.00 0.00 0.00 0 0 500.00 0 10:01:34 stats 10:01:34 CPU InOut Uptime Users FPSPlayers 2.00 0.00 0.00 0 0 500.00 0 10:01:34 stats 10:01:34 CPU InOut Uptime Users FPSPlayers 1.67 0.00 0.00 0 0 500.00 0 10:01:34 stats 10:01:35 CPU InOut Uptime Users FPSPlayers 1.67 0.00 0.00 0 0 500.00 0 10:01:35 stats 10:01:35 CPU InOut Uptime Users FPSPlayers 2.00 0.00 0.00 0 0 500.00 0 Kveri wrote: Hlds - cs 1.6 Sent from my iPhone On 12 Nov 2008, at 14:48, Joseph Laws [EMAIL PROTECTED] wrote: I'm sorry but which game are you guys going back and forth about? xLnT wrote: Then game-hosting.com is faking their stats output. They have stable 1000fps and drops down to 500fps exactly. Not 501 or 499 but 500fps.. once in a while.. otherwise it's 1000fps stable. There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my
Re: [hlds_linux] The 1000fps problem
I'm sorry but which game are you guys going back and forth about? xLnT wrote: Then game-hosting.com is faking their stats output. They have stable 1000fps and drops down to 500fps exactly. Not 501 or 499 but 500fps.. once in a while.. otherwise it's 1000fps stable. There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related
Re: [hlds_linux] The 1000fps problem
There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better
Re: [hlds_linux] The 1000fps problem
Hi, the server do not using 70% it is using maybe 7% its a failure at sv_stats because you have set user_hz to 1000. look at your power @ top and disable the cpu feature at server commandline +sv_stats 0 greets tuorpeZ schrieb: Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising
Re: [hlds_linux] The 1000fps problem
It is entirely possible to maintain a constant 1000fps for HL1-based servers on Linux without faking it. In fact it would probably be harder to fake it than to actually do it. The right hardware combined with appropriate kernel optimization options and -pingboost 2 make it possible. -John - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Wednesday, November 12, 2008 5:41 AM Subject: Re: [hlds_linux] The 1000fps problem Then game-hosting.com is faking their stats output. They have stable 1000fps and drops down to 500fps exactly. Not 501 or 499 but 500fps.. once in a while.. otherwise it's 1000fps stable. There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM
Re: [hlds_linux] The 1000fps problem
This is the output from one of our servers, it is consistent throughout - and this is running Fedora 6 with a 2.6.x kernel (recompiled). The CPU usage is reported falsely, server is running on a Q6600. 20:08:05 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 81.89 119.89 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 85.18 125.19 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 84.30 123.78 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 84.06 122.97 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 84.22 121.49 12216 919.96 10 20:08:07 stats 20:08:07 CPU InOut Uptime Users FPSPlayers 99.90 83.16 118.83 12216 1000.00 10 20:08:07 stats 20:08:07 CPU InOut Uptime Users FPSPlayers 99.90 81.97 116.11 12216 1000.00 10 20:08:07 stats 20:08:07 CPU InOut Uptime Users FPSPlayers 99.90 82.55 116.54 12216 1000.00 10 20:08:07 stats 20:08:07 CPU InOut Uptime Users FPSPlayers 99.90 83.02 117.74 12216 1000.00 10 20:08:10 stats 20:08:10 CPU InOut Uptime Users FPSPlayers 99.90 85.15 121.78 12216 1000.00 10 20:08:11 stats 20:08:11 CPU InOut Uptime Users FPSPlayers 99.90 80.50 114.55 12216 1000.00 10 John wrote: It is entirely possible to maintain a constant 1000fps for HL1-based servers on Linux without faking it. In fact it would probably be harder to fake it than to actually do it. The right hardware combined with appropriate kernel optimization options and -pingboost 2 make it possible. -John - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Wednesday, November 12, 2008 5:41 AM Subject: Re: [hlds_linux] The 1000fps problem Then game-hosting.com is faking their stats output. They have stable 1000fps and drops down to 500fps exactly. Not 501 or 499 but 500fps.. once in a while.. otherwise it's 1000fps stable. There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow
Re: [hlds_linux] The 1000fps problem
@John What is the correct hardware? And what kernel optimization? Does the choise of distro and scheduler depend on it? Regards P.s Feel free to send me an email.. It is entirely possible to maintain a constant 1000fps for HL1-based servers on Linux without faking it. In fact it would probably be harder to fake it than to actually do it. The right hardware combined with appropriate kernel optimization options and -pingboost 2 make it possible. -John - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Wednesday, November 12, 2008 5:41 AM Subject: Re: [hlds_linux] The 1000fps problem Then game-hosting.com is faking their stats output. They have stable 1000fps and drops down to 500fps exactly. Not 501 or 499 but 500fps.. once in a while.. otherwise it's 1000fps stable. There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing
Re: [hlds_linux] The 1000fps problem
Core2 systems seem to achieve constant framerates best although it is obtainable on other hardware as well. In the latest kernel it should only take minor tweaking to do this, such as setting the processor type correctly adjusting a few other settings on the Processor type and features page. The I/O scheduler makes no difference and I can't answer the distro question. The -pingboost 2 is crucial to getting a solid 1000. I can't get into too many details, of course. It's also been awhile since I last explored this in any depth. -John - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Wednesday, November 12, 2008 1:44 PM Subject: Re: [hlds_linux] The 1000fps problem @John What is the correct hardware? And what kernel optimization? Does the choise of distro and scheduler depend on it? Regards P.s Feel free to send me an email.. It is entirely possible to maintain a constant 1000fps for HL1-based servers on Linux without faking it. In fact it would probably be harder to fake it than to actually do it. The right hardware combined with appropriate kernel optimization options and -pingboost 2 make it possible. -John - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Wednesday, November 12, 2008 5:41 AM Subject: Re: [hlds_linux] The 1000fps problem Then game-hosting.com is faking their stats output. They have stable 1000fps and drops down to 500fps exactly. Not 501 or 499 but 500fps.. once in a while.. otherwise it's 1000fps stable. There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate
Re: [hlds_linux] The 1000fps problem
At 03:10 PM 11/12/2008, Ryan Devonshire wrote: This is the output from one of our servers, it is consistent throughout - and this is running Fedora 6 with a 2.6.x kernel (recompiled). The CPU usage is reported falsely, server is running on a Q6600. 20:08:05 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 81.89 119.89 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 85.18 125.19 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 84.30 123.78 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 84.06 122.97 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 84.22 121.49 12216 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 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 ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
At 03:10 PM 11/12/2008, Ryan Devonshire wrote: This is the output from one of our servers, it is consistent throughout - and this is running Fedora 6 with a 2.6.x kernel (recompiled). The CPU usage is reported falsely, server is running on a Q6600. 20:08:05 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 81.89 119.89 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 85.18 125.19 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 84.30 123.78 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 84.06 122.97 12216 1000.00 10 20:08:06 stats 20:08:06 CPU InOut Uptime Users FPSPlayers 99.90 84.22 121.49 12216 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 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 ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
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 forever constant 1000 in practice, no. But as you have seen, it is possible to come very close. With -pingboost 2, HL1 actually uses select() for its delays. It likely has some minor precision errors as well (though these have been reduced by recent improvements in the kernel), but regardless, most of the time this will be trumped by internal factors, such as the game simply taking longer than 1ms to handle a tick, and external factors, such as a high overall machine load or delays in writing to the disk. -John ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
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 forever constant 1000 in practice, no. But as you have seen, it is possible to come very close. With -pingboost 2, HL1 actually uses select() for its delays. It likely has some minor precision errors as well (though these have been reduced by recent improvements in the kernel), but regardless, most of the time this will be trumped by internal factors, such as the game simply taking longer than 1ms to handle a tick, and external factors, such as a high overall machine load or delays in writing to the disk. -pingboost 2 uses alarm(), -pingboost 1 uses select() ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
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 forever constant 1000 in practice, no. But as you have seen, it is possible to come very close. With -pingboost 2, HL1 actually uses select() for its delays. It likely has some minor precision errors as well (though these have been reduced by recent improvements in the kernel), but regardless, most of the time this will be trumped by internal factors, such as the game simply taking longer than 1ms to handle a tick, and external factors, such as a high overall machine load or delays in writing to the disk. -pingboost 2 uses alarm(), -pingboost 1 uses select() ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
} -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] The 1000fps problem } } 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 forever constant 1000 in practice, no. } But } as you have seen, it is possible to come very close. } } With -pingboost 2, HL1 actually uses select() for its delays. It likely } has } some minor precision errors as well (though these have been reduced by } recent improvements in the kernel), but regardless, most of the time this } will be trumped by internal factors, such as the game simply taking } longer } than 1ms to handle a tick, and external factors, such as a high overall } machine load or delays in writing to the disk. } } -pingboost 2 uses alarm(), -pingboost 1 uses select() alarm() or ualarm() ? ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
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] The 1000fps problem
}
} 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 forever constant 1000 in practice, no.
} But
} as you have seen, it is possible to come very close.
}
} With -pingboost 2, HL1 actually uses select() for its delays. It likely
} has
} some minor precision errors as well (though these have been reduced by
} recent improvements in the kernel), but regardless, most of the time this
} will be trumped by internal factors, such as the game simply taking
} longer
} than 1ms to handle a tick, and external factors, such as a high overall
} machine load or delays in writing to the disk.
}
} -pingboost 2 uses alarm(), -pingboost 1 uses select()
alarm() or ualarm() ?
Here's pingboost 1:
void Sys_Sleep_Select( int msec )
{
struct timeval tv;
tv.tv_sec = 0;
tv.tv_usec = 1000 * msec;
select( 1, NULL, NULL, NULL, tv );
}
Here's 2:
void Sys_Sleep_Timer( int msec )
{
struct itimerval tm;
tm.it_value.tv_sec=msec/1000;
tm.it_value.tv_usec=(msec%1000)*1E3;
tm.it_interval.tv_sec = 0;
tm.it_interval.tv_usec = 0;
paused=0;
if( setitimer(ITIMER_REAL,tm,NULL)==0)
{ // set the timer to trigger
pause(); // wait for the signal
}
paused=1;
}
Of course, that was taken from the SDK, so there's no telling what is
what now.. the best option is probably pingboost 3, select and
setitimer have too much jitter.
I wrote a module that uses more aggressive timing, sort of like
UDPSoft's old booster, and it almost eliminates FPS jitter :)
Info was taken from a very old email from alfred..
From:
[EMAIL PROTECTED]@list.valvesoftware.com
on behalf of Alfred [EMAIL PROTECTED]
Sent: 13. juli 2002 07:18
To: [EMAIL PROTECTED]@list.valvesoftware.com
Subject: Re: [hlds_linux] Pingboost modes...
All the pingboot modes attempt to reduce the latency caused by the server.
The default implementation adds around 20msec to each players ping
(under linux).
Mode 1 reduces this by using a different wait method (a select() call).
This method reduces the latency to 10msec.
Mode 2 uses a similar but slightly different method (and alarm() type call).
Again, the result it 10msec worth of latency being added. NOTE that
this method
has the potential to hang a server in certain (terminal) situations.
If anyone has used this mode recently (not the first test we did!)
and it hangs please speak up
Mode 3 minimises the latency to the minimum possible level by processing
a frame EVERY time a packet arrives. This causes the lowest possible latency,
but can also cause extreme CPU usages (it does a complete frame for
every packet,
with each player sending lots of packets per second and 30 players
this adds up to insane amounts of frames). Use this mode at your own risk,
it will consume all available CPU, don't complain that cstrike uses
too much CPU
if you use this mode :-)
In a future release this mode will be tweaked to let the admin
balance latencies
agains CPU usage (by processing a frame every N packets).
There is also an external modules called pingbooster by UDPSoft
(or is it UDPSoftware?). They implement something like mode 3.
As this is an external module, and was built for an older version
of HL (1108) it may not work properly any longer, and future
releases may (accidently) break it.
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
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] The 1000fps problem
}
} 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 forever constant 1000 in practice, no.
} But
} as you have seen, it is possible to come very close.
}
} With -pingboost 2, HL1 actually uses select() for its delays. It likely
} has
} some minor precision errors as well (though these have been reduced by
} recent improvements in the kernel), but regardless, most of the time this
} will be trumped by internal factors, such as the game simply taking
} longer
} than 1ms to handle a tick, and external factors, such as a high overall
} machine load or delays in writing to the disk.
}
} -pingboost 2 uses alarm(), -pingboost 1 uses select()
alarm() or ualarm() ?
Here's pingboost 1:
void Sys_Sleep_Select( int msec )
{
struct timeval tv;
tv.tv_sec = 0;
tv.tv_usec = 1000 * msec;
select( 1, NULL, NULL, NULL, tv );
}
Here's 2:
void Sys_Sleep_Timer( int msec )
{
struct itimerval tm;
tm.it_value.tv_sec=msec/1000;
tm.it_value.tv_usec=(msec%1000)*1E3;
tm.it_interval.tv_sec = 0;
tm.it_interval.tv_usec = 0;
paused=0;
if( setitimer(ITIMER_REAL,tm,NULL)==0)
{ // set the timer to trigger
pause(); // wait for the signal
}
paused=1;
}
Of course, that was taken from the SDK, so there's no telling what is
what now.. the best option is probably pingboost 3, select and
setitimer have too much jitter.
I wrote a module that uses more aggressive timing, sort of like
UDPSoft's old booster, and it almost eliminates FPS jitter :)
Info was taken from a very old email from alfred..
From:
[EMAIL PROTECTED]@list.valvesoftware.com
on behalf of Alfred [EMAIL PROTECTED]
Sent: 13. juli 2002 07:18
To: [EMAIL PROTECTED]@list.valvesoftware.com
Subject: Re: [hlds_linux] Pingboost modes...
All the pingboot modes attempt to reduce the latency caused by the server.
The default implementation adds around 20msec to each players ping
(under linux).
Mode 1 reduces this by using a different wait method (a select() call).
This method reduces the latency to 10msec.
Mode 2 uses a similar but slightly different method (and alarm() type call).
Again, the result it 10msec worth of latency being added. NOTE that
this method
has the potential to hang a server in certain (terminal) situations.
If anyone has used this mode recently (not the first test we did!)
and it hangs please speak up
Mode 3 minimises the latency to the minimum possible level by processing
a frame EVERY time a packet arrives. This causes the lowest possible latency,
but can also cause extreme CPU usages (it does a complete frame for
every packet,
with each player sending lots of packets per second and 30 players
this adds up to insane amounts of frames). Use this mode at your own risk,
it will consume all available CPU, don't complain that cstrike uses
too much CPU
if you use this mode :-)
In a future release this mode will be tweaked to let the admin
balance latencies
agains CPU usage (by processing a frame every N packets).
There is also an external modules called pingbooster by UDPSoft
(or is it UDPSoftware?). They implement something like mode 3.
As this is an external module, and was built for an older version
of HL (1108) it may not work properly any longer, and future
releases may (accidently) break it.
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
so it's not stable, 1000fps almost stable, dropping to 500fps can be achieved by using low latency desktop in kernel, 1000Hz sys_ticrate 1000 and pingboost 3. But I was to able to achieve this only with 0 players on server. Kveri xLnT wrote / napísal(a): Then game-hosting.com is faking their stats output. They have stable 1000fps and drops down to 500fps exactly. Not 501 or 499 but 500fps.. once in a while.. otherwise it's 1000fps stable. There is no chance of running 1000fps stable server on any hardware, YOU CAN'T GET 1000FPS STILL server. I'm running 998-1000fps server on celeron 1,8GHz. Kveri tuorpeZ wrote / napísal(a): Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem
Re: [hlds_linux] The 1000fps problem
I think -pingboost 2 uses alarm(), -pingboost 1 uses select() Kveri John wrote / napísal(a): 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 forever constant 1000 in practice, no. But as you have seen, it is possible to come very close. With -pingboost 2, HL1 actually uses select() for its delays. It likely has some minor precision errors as well (though these have been reduced by recent improvements in the kernel), but regardless, most of the time this will be trumped by internal factors, such as the game simply taking longer than 1ms to handle a tick, and external factors, such as a high overall machine load or delays in writing to the disk. -John ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
@Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause
Re: [hlds_linux] The 1000fps problem
Hello, Has someone succeed to get a stable 1000fps hlds on a core2duo ? I tryed a lot of different kernel configurations but I never achieve to get FPS close to 1000 for a long time. Here what I've already tested (64bits) : Kernel 2.6.24.5 default Kernel 2.6.24.5 1000HZ Kernel 2.6.27.4 1000HZ No preempt Kernel 2.6.27.4 1000HZ preempt Kernel 2.6.27.4 1500HZ preempt (With USER_HZ=1500) Kernel 2.6.26.6 1000HZ preempt + RT patch Neither of those kernels achieve to get a stable +/- 900 fps except for the RT kernel : CPU InOut Uptime Users FPSPlayers 71.00 0.00 0.00 15 2 930.23 0 stats CPU InOut Uptime Users FPSPlayers 70.00 0.00 0.00 15 2 855.43 0 BUT 70% CPU without players ! Other kernels cause some drops to 300FPS. I think that my box (c2duo 2x2Ghz) simply can't compute 1000fps for HLDS... I think that it is only possible on a Xeon/Opteron CPU. Am I wrong ? Did someone succeed to get 1000fps on other than a server class CPU (ie Xeon/Opteron) ? --- tuorpeZ en3my a écrit : None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive
Re: [hlds_linux] The 1000fps problem
None. Default 2.6.26.6 kernel with config i've included in prev. mail. One more thing i forgot to suggest - try to disable almost every feature in MB's BIOS - COM, LPT, USB, Sound, etc. I don't have proof that they really mess your FPS, but belive they do :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Philipp Reddigau [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Tuesday, November 11, 2008 11:13 AM Subject: Re: [hlds_linux] The 1000fps problem @Enemy Wich Kernel patches are you using? best regards, Philipp en3my schrieb: Speaking on the subject... I was trying to archive 1000FPS performance on my dual quad box for few months under Debian Linux. Best result i've got is by using latest kernel (im using 2.6.26.5) with kernel config you can get at http://2po.eu/downloads/custom/www.2po.eu_linux_kernel_2.6.26.6.config But i suggest anyone to experiment with IO Schedulers, all the settings in Processor type and features, try to disable all the Power Managment, try kernel param acpi=off (append=acpi=off for lilo.conf). When dealing with lot of servers on one physical box (i have 8 physical cores and allow to run about 20 HLDS's and few HLDSS's at same time) you must experiment with affinities (man taskset in Linux) and priorities (man renice). And as it was suggested before in this maillist - if you're using 4+ GB of ram, use 64bit distro. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 10:13 PM Subject: Re: [hlds_linux] The 1000fps problem Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards
Re: [hlds_linux] The 1000fps problem
There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
it is possible. linux hernel 1000Hz, low latency desktop, -pingboost 3, sys_ticrate 1000 xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Use sys_ticrate 1000 not 1 Kveri Sent from my iPhone On 10 Nov 2008, at 12:43, xLnT [EMAIL PROTECTED] wrote: I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- - [EMAIL PROTECTED] ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
-pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- - [EMAIL PROTECTED] ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Hi. Just tested one server with -pingboost 3 and +sys_ticrate 1000 on my 1000hz kernel, No Preemt, Tickless. And it was even worse then with -pingboost 2 and +sys_ticrate 1000. got about 900fps fairly stable with the pingboost 3 option. And 980fps stable with -pingboost 2 option. Wierd. Regards Faustas Buškevičius wrote: -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- - [EMAIL PROTECTED] ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- - [EMAIL PROTECTED] ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
Take it easy :) I didn't say they hacked it. Since CPU values have been known to be reported incorrectly, there's no reason why fps values would *always* be reliable and accurate. In response to OP, try running 1000hz, not tickless kernel. Kveri has suggested doing so in the past. I think it's worth a shot. Please let us know of your findings. On Mon, Nov 10, 2008 at 7:15 PM, en3my [EMAIL PROTECTED] wrote: 2 Faustas: I think if they've hacked the output of stats - fps value, they would not change the stats - cpu use field. Don't they? :) Any way, not me nor my friends have a single complain about GH.com servers. That's why i belive the text i see in stats command and i belive stable 1000 fps is possible to archive (under Linux, don't know about others OS's). 2 J T: I'm not advertising any companies hosting. I've just told that perfect server is a real thing :) -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: Faustas Buskevicius [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 6:55 PM Subject: Re: [hlds_linux] The 1000fps problem -pingboost 3 only gives speeded servers when used in combination with sys_ticrate 1. Using -pingboost 3 with sys_ticrate 1000 will not cause any speed related problems at all. On an unrelated, personal and subjective side note, I haven't seen more than 0.75% CPU usage via rcon stats on any of the mentioned GSPs HLDS servers with 10 players on them two years ago. Which made me think the whole output might have been false. On Mon, Nov 10, 2008 at 4:12 PM, J T [EMAIL PROTECTED] wrote: I don't get it? Are you advertising that game-hosting.com gives better performance? I can has performance? Use sys_ticrate 1000 not 1, Kveri ? On Mon, Nov 10, 2008 at 6:47 AM, en3my [EMAIL PROTECTED] wrote: If you've managed to get 970+ FPS without constant drops to 500-750 you are lucky one or linux/HLDS guru :) I belive even CS pro-players won't be able to feel the drops on your server. Best example to ensure yourself that stable 1000FPS is possible is to rent HLDS from www.game-hosting.com (im using hosting in Sweden). I don't know how many HLDSs they have on one box/cpu core, but performance they give is just amazing. -.-.-.-.-.-.-.-.-.-.-.- en3my www.2Po.eu - Original Message - From: xLnT [EMAIL PROTECTED] To: Half-Life dedicated Linux server mailing list hlds_linux@list.valvesoftware.com Sent: Monday, November 10, 2008 1:43 PM Subject: Re: [hlds_linux] The 1000fps problem I have no other services running but opensshserver, Using -pingboost 3 you get a speeded server. And this provider don't have speeded servers. Regards. Kveri wrote: but with no other services running, like apache mysql etc. Just hlds. Kveri xLnT wrote / napísal(a): There is one provider that have 1000fps constantly (with drops down to 500fps sometimes) on their server, or perhaps they are faking the output of rcon stats. I have tried asking them, but no luck. So youre telling me its not possible? Regards. Gary Stanley wrote: At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, scheduler latency, and a bunch of other things cause the drops. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux -- - [EMAIL PROTECTED] ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http
[hlds_linux] The 1000fps problem
Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. Here's a paste from a server. http://paste.prco23.org/32971 This is a single server on a Debian Lenny setup, Dual Quadcore 2.5Ghz, 8GB RAM, 73GB SAS 15krpm. Anyone got their server to run 1000fps stable.. with few drops? Regards. ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
xLnT schrieb: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. Here's a paste from a server. http://paste.prco23.org/32971 This is a single server on a Debian Lenny setup, Dual Quadcore 2.5Ghz, 8GB RAM, 73GB SAS 15krpm. Anyone got their server to run 1000fps stable.. with few drops? Regards. You should test your server with 10 players. I think no one then reaches stable 950FPS. - Marcel ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, 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 to be -Me ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Re: [hlds_linux] The 1000fps problem
At 05:52 AM 11/9/2008, xLnT wrote: Hi. I am really bothered, what is it that makes hlds NOT to use max resources from the server? A gameserver reaches 1000fps sometimes.. but mostly its about 970-985fps. You're not going to able be able to run it that high all the time.. Interrupt latency, 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 to be -Me ___ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
