[ntp:questions] Loopstats updated less frequently than expected

Dave Hart davehart at gmail.com
Sat May 23 17:23:13 UTC 2009

> Thanks for that, Bill.  With the servers with a GPS directly attached, I
> typically get 5400 lines in the log file, for a 16s polling interval.
> That's one line per poll, not one line per approximately eight polls.

With a refclock, the delay is always zero, so the most recent sample
is always the winner of the minimum delay clock filter (because the
dispersion is higher for the older samples).  Over a network, things
are less deterministic.  The clock filter algorithm is a Very Good
Thing, in my opinion.

Dave Hart

More information about the questions mailing list