[ntp:questions] Re: MAXDISPERSE is too high! was Re: is there a way to "lock" the drift frequency

Terje Mathisen terje.mathisen at hda.hydro.com
Mon Dec 8 13:49:00 UTC 2003


Hal Murray wrote:

>>You're kidding of course, but if we were really trying to synch up a client
>>in "deep space", different rules apply: much of the round-trip delay would
>>be fixed, and easily calclulated, and we could subtract it before feeding
>>the samples to the filter.
> 
> 
> Isn't that delay symmetric?  Won't NTP do the right thing?
> 
> I'd be more interested in relativity.  If I have 2 good atomic clocks,
> one on Earth and the other on the moon or Mars, will they get the same
> answer?  If they are in sync at one point, will they stay in sync?
> 
> Isn't relativity a significant correction term for GPS?

Yes: 4 us/day difference between a sat clock and one at sea level.

I.e. the sat clocks are tuned before launch to be off by those 4 us, so 
that they will be correct while in orbit.

Terje

-- 
- <Terje.Mathisen at hda.hydro.com>
"almost all programming can be viewed as an exercise in caching"




More information about the questions mailing list