On Sat, Feb 11, 2006 at 09:59:00PM +0000, Tony Hoyle has written: > Hal Murray wrote: > >>Recent experience has taught me that the GPS signal is inherently > >>*much* more accurate than 'goodish' servers. > > > >That tells you that your system is stable. It doesn't tell you it's > >accurate. There could be systematic errors. > > > Compared to what you're getting off a network - it's the most accurate > source you're going to get by far. The point being there's no point in > trying to find something to calibrate it off unless you happen to have a > handy atomic clock somewhere.
Thanks for the insights so far, guys. The graphing of offset vs. delay for each server gives a good indication of quality of my 'goodish' servers. Most are not so good, but I need more data with the gps. I have a large diurnal temperature variation, but resulting offsets are dealt with well by the pll. Systematic errors: I assume the gps pps sets out synchronized to UTC, suffers cable delays (well, a few ns), hits the DCD pin, goes thorough a gate or two, is picked up by the os, then the pps driver finds out about it, and it is only then that a (pps time, system time) stamp is made. Just my simple way of looking at it. Has anyone any feel for these delays on a modern pc running linux? Would the error be swallowed up in the inherent noise of the ntp process over the internet? -- Richard A Leach A Centre of Excellence for Domestic Information Technology Solutions 5344.9851,N,00201.2071,W _______________________________________________ timekeepers mailing list [email protected] https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
