[ntp:questions] Re: Improving ACTS performance
jcoombs at gwi.net
Tue Apr 11 18:33:11 UTC 2006
"David L. Mills" <mills at udel.edu> wrote in message
news:e1f46t$3o5$1 at dewey.udel.edu...
> My experience with ACTS over a period of many months has been quite
> positive; however, at a poll interval of 36 hours you must expect
> normal room temperature variations to induce errors in the order of
> 10-50 ms. There is not much you can do about that other than to
> reduce the call interval.
> At the moment I am using USNO for test. Expect errors in the 50 ms
> range, somewhat worse than NIST. It seems the primary reason for the
> increased jitter is the variation in call routing from call to call.
The problem I was having was even if I clamped the call interval to
2048 seconds, it seemed like the time returned from NIST was all over
the board. Disabling error correction and compression in my dial
string seems to have done the trick. The call interval walks up to 36
hours, and I might swing 20ms off at most between calls.
More information about the questions