i just put together a test time server to see how much better freebsd handles gps pulses compared to linux/gpsd...
i followed the online document at http://www.david-taylor.myby.co.uk/ntp/FreeBSD-GPS-PPS.htm ntpq does not show any attempts at polling the pps: GPS_NMEA(1) .PPS. 0 l - 16 0 0.000 0.000 4000.00 yet ntpdc -c kern appears to be using the pulse: pll offset: -3.766e-06 s pll frequency: 54.280 ppm maximum error: 1.08071 s estimated error: 0.001464 s status: 2107 pll ppsfreq ppstime ppssignal nano pll time constant: 6 precision: 1e-09 s frequency tolerance: 496 ppm pps frequency: 54.280 ppm pps stability: 0.052 ppm pps jitter: 8.03e-07 s calibration interval: 64 s calibration cycles: 89 jitter exceeded: 97 stability exceeded: 0 calibration errors: 7 this machine has not been running long so there will be considerable errors/offsets/jitter i would think. in david taylor's document it shows sample ntpq -p outputs and his polling counts up. any clues what i may be doing wrong? using the utility that is for the garmin gps18 on a windoze machine, it shows satellites, 3d lock and is functional. the unit is configured to deliver pps, pps auto off is on so it wont send pulses with no satellite lock. pulse width is 100ms, baud is the default 4800, and the pg* statements are turned off. -- Chuck _______________________________________________ timekeepers mailing list [email protected] https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
