what results do you have for threads 2? to compare the difference between original and the changes
On Mon, Aug 2, 2010 at 9:53 AM, Steven Hartland <[email protected]>wrote: > Our results show threads 1 produces slightly lower load than threads 3 > but that may be a sudo effect. The figures for those interested show:- > > threads 3 cpu usage is 52% with an stdev or 15% resulting in 95th pctl of > 67% > threads 1 cpu usage is 48% with an stdev of 15% resulting in 95th pctl of > 62% > > Both measured using over 1,700 samples at 24 players on 2fort of a L5520 > > Regards > Steve > > ----- Original Message ----- From: "M33CROB" <[email protected]> > To: "Half-Life dedicated Win32 server mailing list" < > [email protected]> > Sent: Thursday, July 29, 2010 9:17 PM > > Subject: Re: [hlds] Massive TF2 server cpu hike caused by 2010-07-20 > update/ -threads results part 1 > > > Have you been able to complete your analysis with this Steven? I have had >> my >> GSP add "-threads 1" to the command start up line and things seem to be >> weird. Since adding this I have noticed that my server fps (according to >> net >> graph 4) is always at 30. Previous to this command my server fps would >> fluctuate wildly between 64 and 512. Before this TF2 update my fps would >> fluctuate between 400 & 512 but never ever drop below 400. >> >> Since I dont host my own server (due to lack of upload bandwidth) I cant >> be >> certain if this is an issue with my GSP or with the game itself. It seemed >> to coincide with everyone else having issues. >> >> On Mon, Jul 26, 2010 at 7:23 AM, Steven Hartland <[email protected] >> >wrote: >> >> The results for Windows are in and using both -threads 1 and -threads 3 >>> has produced a definite improvement see the following graphs:- >>> http://www.multiplaygameservers.com/dropzone/13295_week-threads1.png >>> http://www.multiplaygameservers.com/dropzone/12609_week-thread3.png >>> >>> I've now switched one of our popular 2fort servers to -threads 3 from >>> -threads 2 so we can verify which is better threads 1 or threads 3 as >>> comparing two different maps cant confirm that. >>> >>> Our Linux results unfortunately have been invalidated by a change in >>> measurement method, so can't comment there :( >>> >>> Regards >>> Steve >>> >>> ================================================ >>> This e.mail is private and confidential between Multiplay (UK) Ltd. and >>> the >>> person or entity to whom it is addressed. In the event of misdirection, >>> the >>> recipient is prohibited from using, copying, printing or otherwise >>> disseminating it or any information contained in it. >>> In the event of misdirection, illegible or incomplete transmission please >>> telephone +44 845 868 1337 >>> or return the E.mail to [email protected]. >>> >>> >>> _______________________________________________ >>> To unsubscribe, edit your list preferences, or view the list archives, >>> please visit: >>> http://list.valvesoftware.com/mailman/listinfo/hlds >>> >>> _______________________________________________ >> To unsubscribe, edit your list preferences, or view the list archives, >> please visit: >> http://list.valvesoftware.com/mailman/listinfo/hlds >> >> > ================================================ > This e.mail is private and confidential between Multiplay (UK) Ltd. and the > person or entity to whom it is addressed. In the event of misdirection, the > recipient is prohibited from using, copying, printing or otherwise > disseminating it or any information contained in it. > In the event of misdirection, illegible or incomplete transmission please > telephone +44 845 868 1337 > or return the E.mail to [email protected]. > > > _______________________________________________ > To unsubscribe, edit your list preferences, or view the list archives, > please visit: > http://list.valvesoftware.com/mailman/listinfo/hlds > _______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds
