[Ejziponken]

Originally Posted by Maxximou5 Tested on the following running ThePwnt's kernelv2: Processor: Intel® Core™ i7-4790K Processor @ 4.40 GHz OS/Distro: Debian 10 (Buster) Tickrate: 128 Variance: 0.002 @ 24 humans | 0.003 @ 1 human + 23 bots with sv_stressbots 1 Processor: Intel® Core™ i7-6700K Processor @ 4.20 GHz OS/Distro: Debian 10 (Buster) Tickrate: 128 Variance: 0.002 @ 24 humans | 0.003 @ 1 human + 23 bots with sv_stressbots 1 Processor: Intel® Core™ i7-8700K Processor @ 4.70 GHz OS/Distro: Debian 10 (Buster) Tickrate: 128 Variance: 0.001 @ 24 humans | 0.001 @ 1 human + 23 bots with sv_stressbots 1 All servers are running new installations of Debian 10 with the following: Tuned (version 2.10.0) - Profile: latency-performance Docker (version 19.03.2) - CS:GO Debian 10.0 image Kernel (version 5.2.10-rt5) All docker CS:GO images are set for 24 slots @ 128 tick. SourceMod Version: 1.10.0.6441 Metamod:Source Version: 1.11.0-dev+1130 Server rates: fps_max "0" // Sets the servers max fps sv_minrate "128000" // Min bandwidth rate allowed on server, 0 == unlimited sv_maxrate "0" // Max bandwidth rate allowed on server, 0 == unlimited sv_mincmdrate "128" // This sets the minimum value for cl_cmdrate. 0 == unlimited sv_maxupdaterate "128" // Mmaximum updates per second that the server will allow sv_minupdaterate "128" // Minimum updates per second that the server will allow sv_parallel_sendsnapshot "1" // Controls whether the sending of network data is done on the main thread Note: There is a high variance for 2 seconds at round start with 23 bots. If anyone has suggestions for a good monitoring service for CS:GO in terms of slots, tickrate, and variance, I'll be able to measure this much easier.

Is this with or without HT? Do you assign cores/threads?