[time-nuts] NIST UT1 NTP server results

[Bob Camp]

Hi

There have been a number of “unusual” NTP servers running over the years. The whole
“how do we handle leap seconds” thing resulted in a lot of variation. Simple answer is still
the same: Only connect to servers that you have checked out. 

Bob

> On Jul 22, 2016, at 6:23 PM, John Hawkinson <jhawk at MIT.EDU> wrote:
> 
> I have to wonder if it's really such a great idea to have this
> as an open NTP server without huge red flags that it is not UTC.
> One could imagine it leading to big problems if some people started
> syncing to it without undersatnding that it was.
> 
> Has there been thought to at least setting the reference ID to 'UT1'
> instead of 'NIST' (or maybe 'NUT1' since 'NIST-UT1' is too long?).
> 
> 
> With respect to interpolation and soforth, it seems like a lot of NTP
> cares more about frequency than offset, and all this stepping presumably
> wreaks havoc with the frequency? Maybe I'm wrong though...
> 
> --jhawk at mit.edu
>  John Hawkinson
> 
> Tom Van Baak <tvb at LeapSecond.com> wrote on Fri, 22 Jul 2016
> at 15:14:26 -0700 in <60BA6696E49A4C4FA9F6B5792176F81A at pc52>:
> 
> 
>>>      The current algorithm on the server uses the UT1 offset from
>>> Circular A with no interpolation. The value changes at 0 UTC every day.
>>> I did not use any interpolation because the difference in the dUT1
>>> value  from one day to the next is on the order of 1-2 ms, and I
>>> considered that it was likely that the jitter and asymmetry of the
>>> network connection to a typical user would limit the accuracy to a
>>> larger value anyway so that interpolation would not actually improve
>>> anything. However, I will certainly consider changing this. It would not
>>> be a big deal to add interpolation if there were some good reason for
>>> doing so.
>>> 
>>> Best wishes,
>>> 
>>> Judah Levine
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.