On Wed, Jul 20, 2011 at 12:47 PM, Peter Reinhold peter_va...@reinhold.dkwrote:
On Wed, 20 Jul 2011 08:43:57 +0200, Peter Reinhold wrote:
Regarding, playback of demo recreated from server replay block files
My problem now is that the viewpoint seems to be locked, I can't move
around or
On 07/19/2011 02:17 AM, Peter Reinhold wrote:
On Mon, 18 Jul 2011 23:27:44 -0700, Tony Paloma wrote:
Will. Combine the file parts in order to generate the .dem file.
I can't get this to work, i've tried both with and without the .dmx
file, and TF2 tells me
--
] playdemo demofile.dem
Our 32x servers require a dedicated core of a i7 2.8GHz each to maintain
66 update rate, and even then they drop into the 40s in times of high
action. Orangebox's server code is simply not very optimized. Valve
updating their compilers to use SSE3/4 instructions and tree
vectorization might
The engine was originally compiled for 32 slots, but maxplayers was
limited to 24. DF's plugin simply removed the maxplayers cap, letting
you use all 32 slots. A while back they bumped it to 33 slots ( to
accommodate 32+sourcetv/replay ) and it's been like that ever since.
They removed the 24x
This function caps the -maxplayers value, the actual max amount of slots
the server (and client) can handle is some #define somewhere
- Neph
On 07/19/2011 03:29 PM, Loïc PERY wrote:
the code is clear in hl2mp_gameinterface.cpp ...
void CServerGameClients::GetPlayerLimits( int
A 'frame' in the orangebox engine does this, server side:
while (sleep(1ms)) { while (ShouldRunATick()) { RunATick() } }
You can refer to the disassembly or even the old HL2 leak if you don't
believe me. Everyone who claims otherwise is just wrong.
So the only advantage FPS gives is how
Kernels can affect performance in other ways, i personally use a vanilla
2.6.39 kernel with high resolution timers and get great performance out
of it.
On 07/15/2011 11:40 PM, William Balkcom wrote:
With that being said would having a custom kernel no longer be required to
achieve the best
Nemrun 1.8 is available here:
http://nephyrin.net/tools/nemrun/1.8/
Features wise, it adds -onbootcmd which lets you specify a
command/script to run immediately before the server (re)boots (such as
cleaning up your replays folder, or logging server reboots, or whathaveyou).
The biggest
This is definitely a different crash from the engie crash this plugin
works around, though I haven't seen it myself. T_T
- Neph
On 05/30/2011 09:51 PM, bottige...@gmail.com wrote:
Here's another backtrace. This time it crashed at a function that
sounds related to bots, and this was with
The CPU usage that SRCDS manages to chug is borderline
suicidal for the end result, not to mention the countless places in
the engine where memory continues to leak like a sieve
I agree with this, in general, but I don't think 'syscalls' is the issue
here. You need to exclude everything but
I haven't had any bot-related crashes since running this - though I have
had numerous crashes since the replay update, including in the voting
system. If you're getting crashes only with bots on, try turning on core
dumps and getting a backtrace
- Neph
On 05/24/2011 07:58 AM,
You fool you'll kill us all!
On 05/18/2011 09:24 AM, Saul Rennison wrote:
sm_sbfpe_class engineer
Infinite loop? ;D
Thanks,
- Saul
On 18 May 2011 03:18, Nephyrin Zeynephy...@doublezen.net wrote:
I got a core dump of the bots crash, and it's related to the bots being
engineer and
I got a core dump of the bots crash, and it's related to the bots being
engineer and attempting to work out their sentry placement.
This plugin will temporarily prevent bots from spawning as engie, and
should prevent that particular crash, allowing you to use bots again
without fear:
IE:
*** glibc detected *** ./srcds_linux: malloc(): smallbin double linked
list corruption blah blah blah
We have one token bot named Brutus that runs around and is much beloved,
but we've had to turn him off as servers are crashing twice a day with
this. IIRC this was a problem several
First of all, great job on the replay system. I was terrified it was
going to work similarly to SourceTV (rape all the cpus!), but it
actually has worked well without a hiccup for me thus far. That said, I
have a few feature requests that I think my community would love:
replay_savevoice
-
This conversation ACTUALLY HAPPENED hours prior to tony discovering this:
(10:06:15 PM) DrunkenF00lTFC: // Upload replay file on main thread. I
see no problem with this.
(10:06:29 PM) Nephyrin: .. does it really
(10:06:33 PM) DrunkenF00lTFC: lol no.
(10:06:42 PM) DrunkenF00lTFC: well actually
I'm making a few big changes to nemrun, does anyone have any requests
for 2.0?
Right now I've got
- Fix the *third* typo that breaks -srvdir (lolol)
- Remove necessity of using 'screen', use named pipes instead (but still
support screen) (also more friendly for working with other scripts)
-
it to -shutdownscript with nemrun to handle killing your
servers, in which case everything should work (but let me know if you
run into issues/bugs)
- Neph
On 03/14/2011 12:39 AM, Peter Reinhold wrote:
On Fri, 11 Mar 2011 16:52:23 -0800, Nephyrin Zey wrote:
Adds -afternotifycmd for running a script after
Adds -afternotifycmd for running a script after an update has completed.
Fixes -srvdir so people with servers installed in subdirs arn't screwed
(What, do you guys expect me to update *and* test this stuff?)
Actually fixes the empty lockfile bug. Probably.
1.7 download:
We may, in fact, be as high as six users. I am astounded! Be sure to let
me know if you have any feature requests :-P
PS Join the hang out in #hlserveradmins @ irc.gamesurge.net
- Neph
On 02/21/2011 11:35 AM, Kaspars wrote:
Sorry but I'm using this script longer than you :) I'm just staying
For all three of you out there using this script, I've made some minor
updates, which can be found here:
http://nephyrin.net/tools/nemrun/1.6/
Mainly, it now works with L4D2's newer steam.inf format
- Neph
___
To unsubscribe, edit your list
Nemrun works as well as it ever did - valve just stopped bumping version
information, leaving out of date servers listed on the master server,
and breaking autoupdate detection (even their own autoupdate detection).
You're going to have to have a script that literally hammers
hldsupdatetool
You do need to send a line break - which nemrun does automatically.
There was a version of nemrun that had a bug related to
-sharedkillcommand, but it was fixed in an update.
screen -S bob -X stuff $'quit\n'
On Mon, Sep 27, 2010 at 1:09 AM, kama k...@pvp.se wrote:
I wrote a simple gnu screen
Mana pointed out to me that the cssbeta uses 'cstrike_beta' as a
gamedir, but 'cssbeta' as a updatetool ident, causing nemrun to fail to
auto update it properly. I'll fix this in the next release, but you can
also make the change yourself:
Find:
case $GAME in
Add right
Could everyone kindly stop spreading false information if you don't
know what you're talking about?
On some bios, you need to have the speedstep technology (which
turboboost is a part of) enabled to make use of turboboost. You can
enable speedstep in the bios, and the associated turbo features,
My servers have been using ext4 for a year or so without incident
- Neph
On Sat, Aug 28, 2010 at 11:27 AM, ics i...@ics-base.net wrote:
They are fine.
-ics
28.8.2010 21:03, Marc Aldorasi kirjoitti:
I'm setting up a server, and am planning on running srcds on it. / is
probably going to
The DNS round robin still has the bad IP and breaks scripts including
nemrun. So... Back to manually specifying IPs because the DNS cannot be
trusted.
- Neph
On 08/31/2010 12:03 PM, Zoid Kirsch wrote:
The HL2MasterServer at 216.207.205.99:27011 has been moved to
63.234.149.83:27011. You
http://nephyrin.net/tools/nemrun/1.5.2/
Valve just broke their master server dns (hl2master.steampowered.com).
It didn't just go MIA - it actually sits there rejecting packets which
breaks nemrun.
This update fixes this by:
- Handling rejected packets (assumes the master is defunct and
It's too late, Milton! The trust is gone. Gone!
On 08/31/2010 04:25 PM, Milton Ngan wrote:
The DNS round robin still has the bad IP and breaks scripts including
nemrun. So... Back to manually specifying IPs because the DNS cannot be
trusted.
I'm quite positive about the optional updates - the master server out
of date message means your server was denied listing for being too old
- so unless valve configured something wrong for that update, it was
definitely required (or required to stay listed at least).
For the notifications, the
+1 to Everything Gary said - RT kernels are generally a waste. They
might ensure more accurate wakeups, but the sleep(1) call really
limits how accurate those can be anyway even with hires timers, a
ld_preload to mess with sleep() could get you much more
accurate/efficient wakeups, but that's more
Tony is correct - optional updates are not detected at all by
srcupdatecheck. The master server simply is not aware of them. To
detect them would require A: A custom repository script abusing code
deep inside hldsupdatetool (valve would not appreciate this), or B:
Valve releasing some kind of XML
25000 rate might be okay for CSS, but it is too low for a big tf2
server, especially 32x. maxrate and client rates need to be like 60k
to have no choke whatsoever on 32x, and even steam sets client rates
to 30k by default if they have a cable connection.
I would recommend:
sv_minrate 1
Nemrun v1.5 is now out. This fixes a argument parsing bug, and and
issue with two games running on one install not updating properly.
Get it here:
http://nephyrin.net/tools/nemrun/
- Neph
___
To unsubscribe, edit your list preferences, or view the
of just one -
which isnt' the nicest thing for a script to be doing to valve's
servers. So... fix'd.
Get it here:
http://nephyrin.net/tools/nemrun/1.5.1/
(only the srcupdatecheck script has changed since 1.5)
On 08/26/2010 03:53 PM, Nephyrin Zey wrote:
Nemrun v1.5 is now out. This fixes
Note that 'banning *for rcon hacking attempts*' is just a retarded
phrase, it just means they used the wrong password too many times. If
you have any kind of sane rcon password, 15 attempts per day is not
going to compromise your password anytime this century. RCON uses your
server's port
For nemrun, there's -steamdir, the path to steam, and -srvdir which
is the path to the server relative to steam (what is passed to ./steam's
-dir, from steam's working directory*).
If your setup is:
/myservers/steam
/myservers/left4dead2/srcds_linux
/myservers/left4dead2/left4dead2/steam.inf
Getting delisted wouldn't cause this issue. I've had connectivity with
steam be iffy before - try letting the server run for a few minutes and
see if it connects on its own.
nemrun's update daemon (or, more specifically, the srcupdatecheck
script) registers a fake server on the list
:45 AM, Nephyrin Zey wrote:
Get it here:
http://nephyrin.net/tools/nemrun/1.3/
Mainly, it fixes a few bugs related to new updates, namely srcds_i486
- srcds_linux, and a bug with handling the new master protocol.
All changes:
v1.3 Released 07/11/2010 @ 0300PDT
FEATURES
- Passing -verify_all
Stop updating your game valve, wtf.
On 07/30/2010 03:12 PM, Saint K. wrote:
Are doin my fkin head in. I've been very patient, but seriously, what the fuck is up with
the weekly, and sometimes twice a week. I really can't be arsed anymore to host l4d2 due
to this cycle. I've taken down all
On 07/25/2010 10:46 AM, Katrina Payne wrote:
Just for the sake of completeness of this thread, I ran a quick and dirty
search.
Only relevant hit going on Neph Source Query Valve:
http://developer.valvesoftware.com/wiki/Source_Server_Query_Library
First hit going on Neph Script Valve:
Get it here:
http://nephyrin.net/tools/nemrun/1.3/
Mainly, it fixes a few bugs related to new updates, namely srcds_i486 -
srcds_linux, and a bug with handling the new master protocol.
All changes:
v1.3 Released 07/11/2010 @ 0300PDT
FEATURES
- Passing -verify_all will now cause the
This was due to master servers going away - even if it quit after X
attempts, it would start back up and resume timing out against the same
dead server. Now that it uses the DNS round robin though, I'll have it
restart to get a new IP after X failures (next time i bother updating
it, that is)
To anyone using nemrun or the included srcupdatecheck script, either
redownload v1.2 of srcupdatecheck from nephyrin.net/tools/nemrun, or
change the lines
host = blah
port = blah
near the top of the script to:
host = hl2master.steampowered.com
port = 27011
Previously hl2master only pointed
You mean the company that spent their time making this game and only
wants $20 for it? That has offered it for sale for $3 multiple times?
You realize the company is made up of *people* that spent their time
programming, designing, testing, etc this game that you're too cheap to
pay $3 for,
Speaking of unresolved problems...
- Unconnected commands
- Basically any invalid packet causing the server to break in some way
-- A2S_INFO spam (query caching just makes it take slightly more packets
to lag everything out)
-- A2C_PRINT spam causes lag, even before \b or bell character issues.
For anyone (anyone? bueller?) using my nemrun run script, I released a
new update that is primarily an update to srcupdatecheck to play nice
with valve's new master server changes they just implemented an hour ago:
http://nephyrin.net/tools/nemrun/1.2/
And yes, that's why everyone's getting
Those core files are truncated and pretty useless for diagnosing issues
- a full core dump from srcds at startup should be ~200megs. However,
core dumps can include sensitive information such as rcon passwords, so
be careful posting real/full ones to a public place.
On 02/01/2010 12:15 PM,
So earlier today one of my servers was lagging - badly. By time I showed
up the lag had cleared. Then again. Then again. Each time for about 5-10
minutes it would lag, and by time I'd shown up, it was gone. Finally, I
caught the lag happening directly. No unusual FPS or CPU usage spikes,
so i
it, download it, open it in wireshark) and found that 80% of
all traffic was 300byte packets from one ip.
- Neph
On Sun, Jan 24, 2010 at 12:10 AM, Nephyrin Zey nephy...@doublezen.net wrote:
So earlier today one of my servers was lagging - badly. By time I showed up
the lag had cleared
Tf2's default ban system is not very robust, but:
Usage: banid minutes userid | uniqueid { kick }
Use 0 minutes for permanent
So, banid 0 423 kick would kick and ban userid 423 forever. You can
also use their steamid. if you omit kick, it wont bother looking for and
kicking them if
This method would work fine, but if you share the files accross installs
you need to be careful of the permissions involved. Another option would
just be using rsync, if the diskspace is not an issue. If you're setting
up something really complex, using a unionfs might also be good.
I suspect
Lets add: 64-bit support, better instruction set utilization, better
optimization, and free candy to the wishlist. Until valve actually
commits people to this its probably not happening soon
- Neph
On 01/04/2010 04:00 PM, Kyle Sanderson wrote:
I too would like to know this information, the
The crafting is entirely client - steam, there's no way to stop it
that i know of, but you can certainly see when it happens and take
action. Make a script that times how long someone is in the spawnroom.
Have it also hook the player crafted item event. If they craft multiple
items and are in
Perhaps you could... use other master servers?
using
72.165.61.151:27011
72.165.61.136:27012
I've had no issues. I also haven't been auto-assigned to either of the
IPs you list at all recently, are you manually setting these?
- Neph
On 12/30/2009 09:12 AM, Arne Guski wrote:
Why is there no
My bad, reading comprehension ftw. There are other hl1 master servers,
though - 69.28.151.162, for instance. If the issue is bad master
servers, there should be others to use. If servers are being
auto-assigned a dead server, that's certainly a problem.
- Neph
On 12/30/2009 01:50 PM, Richard
I know of maybe 6 people that use this script, so I dont think its an
issue. Besides, if you're using an update daemon, it reregisters itself
every few seconds, so deregistering has no useful effect. The script has
to use a 3-in-a-row approach the responses are so unreliable from the
master
The latest version of nemrun:
http://nephyrin.net/tools/nemrun/
Supports starting an 'update daemon' that watches for updates and sends
reboot commands to servers. My servers usually are updated within
minutes of an update (and they beat the rush to the content servers, so
the DL doesn't take
A quick strace on steam reveals its doing it in the worst possible way:
getuid32() - getpwuid()
Because I actually have a impending use for changing this, I wrote a
dirty hack to let you change this with an env var:
http://infernalsoul.net/Junk/Dump/.c
compile with:
gcc -m32 -fPIC
Engine:
- Added checks to prevent transferring .smx, .gcf, and .sys files between
client/server
- Fixed upload/download exploits with spaces in the file extension or a path
separator at the beginning of the requested file (as reported on the HLDS
mailing lists)
This is sad. You can
Why not just have a IClientResource API where the client stores a
resource of a certain type (ie SPRAY) and the server decides where to
put it and what to name it, and makes sure one client isn't uploading 90
sprays, etc. The assumption that there needs to be a direct filesystem
API seems
I really dont want to get involved in this, but beware basically all
the advice you're receiving (including mine). There's tons of BS
regarding tickrate, RT kernels, fpsmeter.org, 'hit registration', and a
bunch of other useless nonsense by people who really just are echoing
what they read
On 12/01/2009 03:47 PM, Gary Stanley wrote:
SCHED_RR gives the same latency as SCHED_FIFO, in my tests. Under
load, this will be different, though. (send me a message and i'll
send you some code)
I use a multicore system with one core 'dedicated' to each instance. In
this case, FIFO
Actually, OnClientCommand doesn't get commands from unconnected
clients, so it cannot block them. In the newest sourcemod trunk, there
is a function you can use to hook *all* commands and veto them as necessary:
functag public Action:CommandListener(client, const String:command[],
argc);
Just to remind everyone, a lot of us run a pretty popular srcds
administration chat channel on gamesurge (linux and windows, everyone
welcome)
irc.gamesurge.net #hlserveradmins
Gamesurge is also home to:
#sourcemod
#sourceop (the tf2items.com guy)
#sourcebans
#hlcoders
and more. So join up!
Check out pred's CBaseServer extension on the sourcemod forms. With it
you could write a simple plugin to autoipban people who do this.
On 09/26/2009 11:06 PM, ics wrote:
There seems to be somekind of script laying around on certain players
which they can execute upon joining our TF2 servers.
I'm running three 33x linux servers on a q6600 with low var and solid
66 update rate except in the highest action.
You can connect to tf1.game.doublezen.net and check it out for
yourself. Make sure you set your rate to 6 or a similar uncapped
value before you judge the performance of a 32x,
I'd say that you're hitting your max on a computer that old. I'm
amazed you can get to 16-18. My 32x servers take up 70-90% of a core
on a q6600, so an old p3 wouldn't be able to come close
You might be able to do better on linux, the srcds performance has
been getting much worse patch after
Your server crashes every time you try to start it! April Fools! =D
or something like that, judging from the last time a phantom dll came
around.
- Neph
On 04/01/2009 01:39 PM, Rick Payton wrote:
So what got updated?
[mauiri...@jumbo-new srcds_ob]$ ./steam -command update -game tf -dir .
(not sure how many of you are still using this, but..)
Get it here:
http://nephyrin.net/tools/nemrun/1.1/
The latest version is always here:
http://nephyrin.net/tools/nemrun/latest/
Nemrun is an advanced srcds_run replacement that has numerous
features, including:
- Functioning autoupdate
-
will auto-restart if there is a crash.
ERROR: ld.so: object 'NephuCPUFix.so' from LD_PRELOAD cannot be preloaded:
ignored.
NephuCPUFix.so is in /orangebox/bin/
After typing r_sse2 in console it tells me that support for this is
disabled.
On Tue, Mar 24, 2009 at 8:09 AM, Nephyrin Zey nephy
hlds/srcds's handling of packets needs more sanity/rate checks, there
are numerous different types of packet you can flood at the server to
confuse it in different ways.
Also, the length 0 bans can just be done with this firewall rule:
-A INPUT -p udp -m udp --dport 27015 -m length --length 28
On 03/18/2009 02:02 PM, Daniel Vogel wrote:
This is not a question for a half life dedicated linux server mailing list.
(It's not even a question). And it's dumb, senseless and wastes ressources
and makes the archive harder to read and find useful topics.
Thank you Daniel, for your
Oh shit a post i consider spam. Quick, guys, 9 replies, start a new
topic, throw some replies in there too.
Spam fixed! Thank god for our quick action.
- Neph
On Wed, Mar 18, 2009 at 3:26 PM, DontWannaName! ad...@topnotchclan.com wrote:
Pong! lol, what happens when we say pong :D
This do anything on the server side or is srcds still single threaded?
Possibility one:
Srcds now takes advantage of multiple cores, chokepoints removed,
bitbuffers optimized, packet code optimized, now built with SSE2 code
by default since no pre-SSE2 cpu could even handle TF2 servers anyway,
;/
2009/3/19 Nephyrin Zey nephy...@doublezen.net
This do anything on the server side or is srcds still single threaded?
Possibility one:
Srcds now takes advantage of multiple cores, chokepoints removed,
bitbuffers optimized, packet code optimized, now built with SSE2 code
by default since
I have http://nemu.doublezen.net/cacti/scripts/rcon.py to fetch player
numbers and other data. You're welcome to use it. Other than that, it's
just configuring data sources/graphs/etc, which is a nightmare on its own~
- Neph
On 03/09/2009 12:43 PM, 1nsane wrote:
Bug Neph, he does/did.
On
On 02/24/2009 06:34 PM, Jason Ruymen wrote:
The Scout update should be live in about an hour or hopefully less from now.
Jason
Jason! irc.gamesurge.net/#hlserveradmins
Keep us company ;)
- Neph
___
To unsubscribe, edit your list preferences,
Reported in firefox and was marked as a web forgery two minutes later.
Valve should get an EV cert for steamcommunity or something.
- Neph
On 02/23/2009 10:09 AM, Rodrigo Peña wrote:
Hi,
Today in my forum I saw a post from a league and they linked it to they
steam community group. I click
On Feb 23, 2009, at 2:06 PM, Nephyrin Zey wrote:
Reported in firefox and was marked as a web forgery two minutes later.
Valve should get an EV cert for steamcommunity or something.
- Neph
On 02/23/2009 10:09 AM, Rodrigo Peña wrote:
Hi,
Today in my forum I saw a post from a league
Might this be bad code imported from l4d into tf2 that has caused this?
I am seeing zero coke and zero drop in update rate since the 30k bug was
fixed, except in situations where my server is getting poor FPS (which
is the same old generic server lag, not anything net related). What
weeks ago, none
of them had the problem resolved after yesterdays patch.
Regards,
-Gryz
On Tue, Feb 17, 2009 at 9:53 PM, Nephyrin Zey nephy...@doublezen.net
wrote:
Don't confuse this with the rate 30K bug. See the something is
limiting client-server bandwidth for more and contribute there so
Keep that Steam.cfg file in place, and try removing the ~/.steam
directory, which will force the steam binary to update itself and find
new content servers. Also, where is this system located?
- Neph
On Thu, Feb 19, 2009 at 2:20 PM, Jameson Quave jqu...@gmail.com wrote:
I have tried using the
This is going from what i know about srcds, though it should apply to
hlds, a little:
On modern kernels, i saw nothing but highly increased overhead when
using an RT kernel. This might be different for hlds (i'm on srcds)
but i think that with RCU preemption and the highest preemption level,
a
I don't think these stats track weapon used. The client updates the
server with its own stats (which is what tracks your
kills/points/max-ubers and everything else). But more importantly,
weapons are all in steam cloud, so they can see who was using what
loadouts. The bugged loadout configs were
Don't confuse this with the rate 30K bug. See the something is
limiting client-server bandwidth for more and contribute there so
the info is one place :-P
- Neph
On Tue, Feb 17, 2009 at 10:50 AM, 1nsane 1nsane...@gmail.com wrote:
I can has my lag fix? Please! :D
On Tue, Feb 17, 2009 at 5:29
This plugin doesn't do anything at all except on stats upload, which
is on map change only. Even if it was attempting to upon a player's
death, this plugin would block it, only speeding things up.
It's possible something super bizarre is going on, but i'm pretty sure
this plugin doesn't do or
is still appearing :/
Are you sure it is working?
2009/2/11 Pawel tra...@gmail.com
Thanks, good work :)
On Wed, Feb 11, 2009 at 3:42 AM, Nephyrin Zey
nephy...@doublezen.netwrote:
After ~24hours of running the plugin, many people have reported
experiencing *no* timeouts
This is for hex editing the server to remove player limits. This can
more easily be done with sourceop's PLR plugin, and the server/client
only handles 33 slots, so if you launch a +maxplayers 64 you're just
going to see some really uninteresting segfaults.
I've never heard of being able to
12,000 players on today.
Not that we don't need a new class pack, but suggesting that people
are losing interest or that your servers are going dead because of the
lack of one is a bit extreme :-P
- Neph
___
To unsubscribe, edit your list preferences,
After ~24hours of running the plugin, many people have reported
experiencing *no* timeouts, and nobody has reported an issue. I'll go
ahead and say that this is no longer experimental, though you should
still use it with caution :-P
Here's a 1.1 version of my plugin that has a left4dead compile,
We all know how well supposed timeout bug fixes work, but this one, i think,
does. Probably.
http://nephyrin.net/tools/NephDSU/NephDSU_v1.0.zip
This is an MM:S plugin, that adds a cvar: neph_nostats.
Setting neph_nostats to 1 blocks stats upload, which has been shown to be
causing timeouts (or,
-nogamestats was shown to *NOT* actually do what it says
Fixed
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Alliedmods.net is setting up forums and a wiki for srcds/hlds
administration in response to the mailing lists going down. See
https://forums.alliedmods.net/forumdisplay.php?f=130
___
To unsubscribe, edit your list preferences, or view the list archives,
Valve seems to think that as long as there are always servers they
shouldn't waste their time. They don't seem to understand that quality
servers and their associated gaming communities pull people into the
game and keep them there. How many CSS players are in a clan and
regularly participate in
Guys, the original thread in hlds (not _linux) has responses from
bur...@valvesoftware.com. He also said that due to the tremendous
amount of feedback, they're holding off on taking down the lists for
the time being.
- Neph
On Mon, Feb 9, 2009 at 7:55 PM, Ben fh-l...@3fl.net wrote:
You're
It's very rare, but still happens. Has happened maybe 5 times in the last 7
days for me, of 3 full 32x servers.
- Neph
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
TF2 takes considerably more CPU than when i first started hosting servers,
easily.
- Neph
___
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
Don't get me started guys!
Suspects (might have no improvement, might be all the difference in the
world):
- Their 'all in one binary' means that aside from the math functions they
specifically program, 90% of code is generated for a featureless i486
platform. Providing an advanced binary that's
A better solution would be is to just completely profile the engine
to see what it's spending it's time doing.
Done that - very call heavy, 8%+ time in bitbuffers, something like 4% in
thunks.
The engine also spends HZ/sec reading the clock. Do you really need
to call the clock that many
1 - 100 of 201 matches
Mail list logo
