[LEAPSECS] BBC radio Crowd Science
zefram at fysh.org
Thu Feb 2 04:10:06 EST 2017
Steve Allen wrote:
>On Wed 2017-02-01T14:27:03 -0800, Tom Van Baak hath writ:
>> 6) clock sends TAI and -N, user wants UTC
>> - 6a) clock must send N1 and N2 so that user can generate UTC in all cases, or
>I recognize this as the model chosen by GPS,
Kind of, but what GPS does isn't quite according to Tom's model.
Tom implies the clock sending DTAI values along with each timestamp.
In GPS the DTAI values are separate from the timestamps, and sent on a
different schedule much less often. It's more like "clock sends TAI, user
has partial DTAI table supplied by something sitting next to the clock".
The two DTAI values and cutover time are the minimum that GPS could send
through the not-quite-real-time channel to enable the user to recover a
DTAI value for each transmitted timestamp. (It actually achieves more
than that, of course.)
More information about the LEAPSECS