[LinuxPPS] a fluke...?

[gnu not unix]

In message <200701231830.l0NIUHN0006385 at ring.wraith.sf.ca.us> you write:

>The plot smoothed right out!

That appears to have been a fluke. I removed the atom refclock,
which left only the nmea (and wwv audio), and the plot went
back to the +/- 20 microsecond swings. Adding the atom driver
back, did not smooth the plot out again (as much). The plot
is now showing +/- 10 microsecond offsets. 


http://www.wraith.sf.ca.us/ntp/rrd/index.html


Is anyone else running  plots and seeing any interesting
trends? 

../Steven