Am 23.02.2013 um 04:15 schrieb fritz: > I've been playing catch up in the archives here, and I thought I might > weigh in with my motherboard latency results. > > This is a motherboard that I initially tried a latency test when I > bought it, and it was very disappointing. However, my reading gave me a > few configuration tweaks that I had not previously tried. > > In the UEFI: > - disable c6 state > - disable 'cool-n-quiet' > - no overclocking, 'auto' on cpu and memory speed > > The hardware: > Asus E35M1-M Pro motherboard, with E-350 APU @ 1.6GHz > 4GB DDR3 1066 RAM > > I used the Ubuntu 10.04 LTS install I had in my other computer, booting > with "isol_cpus" enabled. I simply unplugged the drive and moved it over. > > The latency results, performed while doing simultaneous full-screen > glxgears and `find /`: > servo thread (1ms) 6684ns > base thread (25us) 4288ns
Hi Fritz, thanks for the data point! It confirms my theory that undeterministic recovery times from powersave states are a major (if not the) primary source of latency, and that factor is largely operating-system independent since it has no control over it any chance of permuting the 'C6 state' and 'cool-n-quiet' settings and redo measurements? - Michael > > ------------------------------------------------------------------------------ > Everyone hates slow websites. So do we. > Make your web apps faster with AppDynamics > Download AppDynamics Lite for free today: > http://p.sf.net/sfu/appdyn_d2d_feb > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://p.sf.net/sfu/appdyn_d2d_feb _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
