[ntp:questions] Off by a second glitch from Garmin GPS 18 LVC

Hal Murray hal-usenet at ip-64-139-1-69.sjc.megapath.net
Fri Aug 31 20:43:27 UTC 2007


Has anybody else noticed a glitch like this?

This is running on a system without PPS so I'm looking at the
data from only the NMEA text.  There is no fudge time1 correction.
The 167 ms offset is normal.

Note that the NMEA driver is collecting a sample each second.
They get filterd and averaged.  So this quirk was going on
for most of a minute (the polling interval) if it got through
the filter.


 From peerstats:

54343 31590.170 127.127.20.1 90f4 -0.167075278 0.000000000 0.001188301 0.001607286
54343 31656.194 127.127.20.1 90f4 -0.167529694 0.000000000 0.001241714 0.001890186
54343 31719.193 127.127.20.1 90f4 -0.188947750 0.000000000 0.001745779 0.022509378
54343 31784.170 127.127.20.1 90f4 -1.159567833 0.000000000 0.001544575 0.989659101
54343 31848.196 127.127.20.1 90f4 -0.188667889 0.000000000 0.001562625 0.367421872
54343 31913.166 127.127.20.1 90f4 -0.166658083 0.000000000 0.001562850 0.375471815
54343 31976.175 127.127.20.1 90f4 -0.167776972 0.000000000 0.001392490 0.375030582


 From clockstats

54343 31590.170 127.127.20.1 $GPRMC,084630,A,3726.0818,N,12212.2616,W,000.0,320.1,310807,015.0,E*64
54343 31656.194 127.127.20.1 $GPRMC,084736,A,3726.0819,N,12212.2617,W,000.0,320.1,310807,015.0,E*63
54343 31719.193 127.127.20.1 $GPRMC,084839,A,3726.0819,N,12212.2617,W,000.0,320.1,310807,015.0,E*63
54343 31784.170 127.127.20.1 $GPRMC,084944,A,3726.0870,N,12212.2646,W,000.0,320.1,310807,015.0,E*63
54343 31848.196 127.127.20.1 $GPRMC,085048,A,3726.0876,N,12212.2638,W,000.0,320.1,310807,015.0,E*68
54343 31913.166 127.127.20.1 $GPRMC,085153,A,3726.0863,N,12212.2633,W,000.0,320.1,310807,015.0,E*6C
54343 31976.175 127.127.20.1 $GPRMC,085256,A,3726.0805,N,12212.2604,W,000.0,320.1,310807,015.0,E*6E

Those are just one sample per polling interval.  I haven't checked
the numbers (yet) to see if this one is off.

-- 
These are my opinions, not necessarily my employer's.  I hate spam.




More information about the questions mailing list