Spawn Time For Large Servers Seems Slow

FrizzledFrizzled Join Date: 2002-11-02 Member: 4424Members
edited December 2002 in General Server Discussion
<div class="IPBDescription">Any suggestions?</div> I've heard that the game wasn't designed for servers with more than 20 people ... I've found that the game is very, Very well balanced with more than twenty people, save for one thing:

Alien Spawn times ... In a 15 on 15 server the aliens take about 10 seconds each to spawn. The problem is with 30 people on the map you have aliens dying more often than once every ten seconds. After that, Aliens start backing up in the spawn queue, waiting up to a minute and a half to spawn.

If I could cut the spawn time to seven seconds, i think that would work better ... is there any way to do this? Anyone have any suggestions?

If you want to see this in-action, check out the {H2O}-7 NS v1.03 server (not mine, but a great example of what happens with 30 people).

thanks,
_f

Comments

  • Lt_PredatorJ_LGSOLt_PredatorJ_LGSO Join Date: 2002-12-15 Member: 10851Members
    In my opinion , on that server (Which i have played on before , and experienced that problem) the admin should reset the "ns_alienrespawntime" var to a lower number , thus increasing the speed at which each hive regenerates khaara.
  • PJJPJJ Join Date: 2002-11-21 Member: 9579Members
    I think some code inside NS to scale the spawn time to depend on the number of people on a team would be very useful in the game. It probably wouldn't be too difficult to code once the Dev's decided what factor by which to scale the time. But that is just my opinion. =)
  • Lt_PredatorJ_LGSOLt_PredatorJ_LGSO Join Date: 2002-12-15 Member: 10851Members
    Good idea , i believe the Dev's have some time on this issue however , being that there arent many servers out there running more the 20 players do to the extreme resource issues. The aformentioned 30 player server is about the laggiest thing ive ever played on anyway. <!--emo&:p--><img src='http://www.unknownworlds.com/forums/html/emoticons/tounge.gif' border='0' valign='absmiddle' alt='tounge.gif'><!--endemo-->
  • ZdroneZdrone Join Date: 2002-11-02 Member: 3914Members, Constellation
    <!--QuoteBegin--Lt. PredatorJ (LGSO)+Dec 21 2002, 03:08 PM--></span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (Lt. PredatorJ (LGSO) @ Dec 21 2002, 03:08 PM)</td></tr><tr><td id='QUOTE'><!--QuoteEBegin-->The aformentioned 30 player server is about the laggiest thing ive ever played on anyway. <!--emo&:p--><img src='http://www.unknownworlds.com/forums/html/emoticons/tounge.gif' border='0' valign='absmiddle' alt='tounge.gif'><!--endemo--><!--QuoteEnd--></td></tr></table><span class='postcolor'><!--QuoteEEnd-->
    Bah, runs fine for me... Of course I own it <!--emo&:)--><img src='http://www.unknownworlds.com/forums/html/emoticons/smile.gif' border='0' valign='absmiddle' alt='smile.gif'><!--endemo-->

    I'm looking for a way to change the respawn rates on my own. Looks like its hard coded unfortunately.

    There is a 50 meg download (CS mappack) on the same line as the server that I need to move elsewhere, you may have been playing on it when someone was downloading.

    The funny thing is, I have 2, 18 person servers on a line the same speed as the single 30 person server. It runs quite a bit higher utilization than the 30 person line: <a href='http://www.h2ocs.com/mrtg/' target='_blank'>http://www.h2ocs.com/mrtg/</a>
  • verboseverbose Join Date: 2002-11-25 Member: 9968Members, Constellation
    <!--QuoteBegin--Zdrone+Dec 24 2002, 07:56 PM--></span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (Zdrone @ Dec 24 2002, 07:56 PM)</td></tr><tr><td id='QUOTE'><!--QuoteEBegin-->The funny thing is, I have 2, 18 person servers on a line the same speed as the single 30 person server. It runs quite a bit higher utilization than the 30 person line: <a href='http://www.h2ocs.com/mrtg/' target='_blank'>http://www.h2ocs.com/mrtg/</a><!--QuoteEnd--></td></tr></table><span class='postcolor'><!--QuoteEEnd-->
    Most boards/forums are hungry little buggers. If you have active boards, that could easily explain the bandwidth use of that box.
  • ZdroneZdrone Join Date: 2002-11-02 Member: 3914Members, Constellation
    <!--QuoteBegin--verbose+Dec 25 2002, 06:04 PM--></span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (verbose @ Dec 25 2002, 06:04 PM)</td></tr><tr><td id='QUOTE'><!--QuoteEBegin--><!--QuoteBegin--Zdrone+Dec 24 2002, 07:56 PM--></span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (Zdrone @ Dec 24 2002, 07:56 PM)</td></tr><tr><td id='QUOTE'><!--QuoteEBegin-->The funny thing is, I have 2, 18 person servers on a line the same speed as the single 30 person server.  It runs quite a bit higher utilization than the 30 person line:  <a href='http://www.h2ocs.com/mrtg/' target='_blank'>http://www.h2ocs.com/mrtg/</a><!--QuoteEnd--></td></tr></table><span class='postcolor'><!--QuoteEEnd-->
    Most boards/forums are hungry little buggers. If you have active boards, that could easily explain the bandwidth use of that box.<!--QuoteEnd--></td></tr></table><span class='postcolor'><!--QuoteEEnd-->
    I can see the spikes on downloads. I was refering more to the averave utilization. The 30 person server (H2O-7), is much lower on average than the 2, 18 person servers (both on the same line).

    I have some better tools to graph with but cant show realtime stats. I will start monitoring with them out of curiosity.
  • verboseverbose Join Date: 2002-11-25 Member: 9968Members, Constellation
    <!--QuoteBegin--Zdrone+Dec 25 2002, 12:55 PM--></span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (Zdrone @ Dec 25 2002, 12:55 PM)</td></tr><tr><td id='QUOTE'><!--QuoteEBegin-->I can see the spikes on downloads. I was refering more to the averave utilization. The 30 person server (H2O-7), is much lower on average than the 2, 18 person servers (both on the same line).<!--QuoteEnd--></td></tr></table><span class='postcolor'><!--QuoteEEnd-->
    Gotcha. It didn't register that the board server was also game server H20-7, so I was reading things wrong.

    Are your servers always full? The bandwidth used is directly proportional to the number of current players, so if the sum of the number of players on the 18-max servers is greater than the number of players on the 30-max server, the usage will be higher. Graphs with a longer timespan would help as well. Need more data to analyze further.
  • ZdroneZdrone Join Date: 2002-11-02 Member: 3914Members, Constellation
    <!--QuoteBegin--verbose+Dec 25 2002, 07:24 PM--></span><table border='0' align='center' width='95%' cellpadding='3' cellspacing='1'><tr><td><b>QUOTE</b> (verbose @ Dec 25 2002, 07:24 PM)</td></tr><tr><td id='QUOTE'><!--QuoteEBegin-->Are your servers always full? The bandwidth used is directly proportional to the number of current players, so if the sum of the number of players on the 18-max servers is greater than the number of players on the 30-max server, the usage will be higher. Graphs with a longer timespan would help as well. Need more data to analyze further.<!--QuoteEnd--></td></tr></table><span class='postcolor'><!--QuoteEEnd-->
    Typically they are always full except between 12-4am or so. Kind of funny, you can see the drop for Christmas this morning on the dual 18 person line.

    Proportionately, the 2 lines don't balance out. The 32 players on the dual 18 person server have a much higher bandwidth utilization than the single 30 person server.
    32 players look like they average around 105k while the 30 person is around 80-90k.

    Unfortunately, MRTG is rather limited in its abilities to view small timelines. I will put Solarwinds on one of the servers and monitor using that. I will have to post the utilization pictures instead of viewing it realtime though. The nice thing is, I will be able to monitor CPU over time as well.

    For a longer timeline, click on the graph to drill down. Still limited though.
Sign In or Register to comment.