[ntp:questions] Re: Behavior when one server is erroneous

David Schwartz davids at webmaster.com
Sat Jan 17 03:15:21 UTC 2004


"Christopher J. Burian" <cburian at ll.mit.edu> wrote in message 
news:mailman.35.1074288033.1757.questions at ntp.org...


> And when jitter is bad, shouldn't
> the polling time be more frequent?


    Quite the reverse. More jitter demands a less frequent polling interval. 
Consider a clock that's only accurate to the nearest second versus a clock 
that's accurate to the nearest hundredth of a second. Which clock do you 
think it makes sense to check more often? High jitter requires a larger 
polling interval to permit the jitter to be averaged out.

    DS







More information about the questions mailing list