On Mon, Feb 20, 2012 at 10:52, <[email protected]> wrote: > I found an error, I did the measurement on a different computer, > not on that computer were the DCF77 clock was attached. Now > I did the measurement on the NTP server directly. I got somehow > better results. I recorded the difference of offsets between the > DCF77 clock and ptbtime2.ptb.de. The avverage is less than > 0.5ms and no single difference is higher than +/- 2ms. I had to > adjust the time1 parameter to reach this. The graph at the pool > shows me now a very constant line at 5-6ms. Since now the > values are less than the oscillation before, I think I can't get it any > better without a GPS clock. > The values for the time1 parameter are > default of the driver: 258 > Best value now: 247 (lowest average difference to local offical time) > Value for a zero line at the pool: 241
I assume all those values are actually divided by 1000 in ntp.conf as time1 is in seconds. For the purposes of giving the best time service to pool clients, I think 0.241 is the winner. For best local timekeeping, 0.247. I see a similar situation on non-pool server ntp.davehart.net connected to a business-class ADSL with 768 kbps up, 3 Mbps down. If I sync to GPS PPS via serial port, I see high-quality GPS-synched sources connected via symmetric access links as 6.7 msec off. If I fudge my PPS so I follow network sources very closely (and see very little jitter), I'm fudging it 6.7 msec which I know is far more than the error attributable to propagation delay due to the speed of light down the cable from the receiver plus the computer's latency getting into the serial interrupt routine when DCD changes on the PPS leading edge, so I know I'm running local clocks milliseconds off UTC. Since for that machine I care more about providing accurate local time, I leave the time1 at its default 0. If it were in the pool, I'd be tempted to use -0.0067. Cheers, Dave Hart _______________________________________________ pool mailing list [email protected] http://lists.ntp.org/listinfo/pool
