Yo Eric! On Tue, 11 Oct 2016 23:28:26 -0400 "Eric S. Raymond" <e...@thyrsus.com> wrote:
> Gary E. Miller <g...@rellim.com>: > > Yo All! > > > > I like the code direction in today's git head, but something went > > wrong. > > > > See attached. You can see that when I restarted with git head at > > 23:25 11 Oct UTC, that the loss of precision, which was at 0.954 µs > > now appears to be doubled to about 1.908 µs. > > > > You can see that plot live here: > > > > https://pi4.rellim.com/day/#local_rms_time_jitter > > > > BTW, the attached png also shows off the in-progress --clip mode to > > ntpviz. The new mode throws away data less than 1% and more than > > 99%. Good for seeing the data after restarts as it ignoers the > > huge spikes. > > > > RGDS > > GARY > > --------------------------------------------------------------------------- > > Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR > > 97703 g...@rellim.com Tel:+1 541 382 8588 > > I don't have a baseline - I wouldn't know bad numbers if I saw them. The left of 23:25 11 Oct UTC is 'before' the right is 'after'. After is clearly worse. Notice all the white at the bottom? That is bad. More white at the bottom is worse. Good data being thrown away. > You'll have to bisect on this. Maybe something will jump out at us > when we look at the diff. Yeah, for Wednesday... Until then best to not add to the complexity. RGDS GARY --------------------------------------------------------------------------- Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703 g...@rellim.com Tel:+1 541 382 8588
pgpdt0uPqvjkb.pgp
Description: OpenPGP digital signature
_______________________________________________ devel mailing list devel@ntpsec.org http://lists.ntpsec.org/mailman/listinfo/devel