In message <[EMAIL PROTECTED]> you write: >gnu not unix wrote: >> [EMAIL PROTECTED] ntpq -c lpe > remote refid st t when poll reach delay offset jitter >============================================================================== > *GPS_NMEA(0) .GPS. 0 l 1 64 377 0.000 0.002 0.001 > +wraith.wraith.s .GPS. 1 u 239 1024 377 0.183 0.007 0.001 > -thrall.wraith.s .GPS. 1 u 214 1024 377 0.181 0.013 0.004 > +smidge.wraith.s .GPS. 1 u 182 1024 377 0.357 0.000 0.005
>That's pretty good... fast processor I guess (SMP too, which has gotta >help). Actually, from top to bottom, they are 1000Mhz, 266Mhz PII, 400Mhz K6/2, 400Mhz K6/2. The kernel is smp but the hardware is uniprocessor (1Ghz Athlon). >So the lesson is if you want any accuracy Linux 2.4 kernel or use >FreeBSD... although the parallel port stuff your site mentions sounds >interesting. The 2.6.3 kernel with the "ppskit lite" worked pretty well here. I don't have one running at the moment, but I recall it was routinely in the 10's of microseconds offset. A proud member of the herd (grin). The fanout box is the Big Win here. I've got to replace the differential driver on the roof, though--its gotten badly corroded after a couple years exposure to the elements. The network here is 100mb switched ethernet. This speed results in 100 microsecond delays. A 10 megabit half duplex isa card will show up with 1 millisecond delay when added to this setup. It is interesting to see that the freebsd 4.7 unit (smidge) seems to have 100 extra microseconds of udp stack latency, compared to the other two linux 2.4.20/ppskit udp stacks. This may be due to the freebsd kernel having some kind of kernel packet filtering code compiled in, although all the rules are simply wide open. None of the linux boxen have firewall code. ../Steven _______________________________________________ timekeepers mailing list [email protected] https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
