Yo Daniel! On Fri, 6 May 2016 15:34:29 -0400 Daniel Franke <dfoxfra...@gmail.com> wrote:
> On 5/6/16, Gary E. Miller <g...@rellim.com> wrote:
> >> Do we have to live with long convergence times? Do you have any
> >> theory about what causes this and how it can be fixed?
> >
> > I have not gone deep into the PLL, grouping, and selection layers.
> > Daniel Franke's talk at Penguicon leads me to believe he is
> > starting to see these issues.
>
> Depends what you mean by "see".
Well, two steps below grok. No longer blind to the issues.
> In the literal, empirical sense, no, I
> haven't observed ntpd performance closely enough to possibly notice
> this. But I think I understand from first principles why it happens,
> and I'm researching Bayesian approaches to clock synchronization that
> I think will improve both speed-of-convergence and asymptotic
> precision.
Great! Also reseaerch chronyd. ntpd and chronyd start with very
different assumptions and use cases, which lead to different solutions.
When you are ready, we'll bike-shed it for you. :-)
RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
g...@rellim.com Tel:+1 541 382 8588
pgpR8f1Q9mnno.pgp
Description: OpenPGP digital signature
_______________________________________________ devel mailing list devel@ntpsec.org http://lists.ntpsec.org/mailman/listinfo/devel
