[time-nuts] NIST time services

Chris Albertson albertson.chris at gmail.com
Tue Mar 25 19:47:30 EDT 2014


On Tue, Mar 25, 2014 at 5:52 AM, Paul <tic-toc at bodosom.net> wrote:
> On Mon, Mar 24, 2014 at 4:26 AM, Poul-Henning Kamp <phk at phk.freebsd.dk
>wrote:
>
>> In message <CABbxVHuQc0144==21mDa_R8ErKov=
>> EM+9rVRbPggeXNZZTjnAg at mail.gmail.com>
>> , Chris Albertson writes:
>>
>> >Yes.  NTP calls it "root distance" [...]
>>
>> And it is generally useless, because people don't calibrate it.
>>
>
> How do you calibrate root distance assuming that it's "one-half the
> roundtrip root delay plus the root dispersion plus minor error
> contributions not considered here"?


Peole here have very much mis-understood the term "Root Distance".   People
don't calibrate it because It is NOT a user input.  It is an internal
variable that NTP uses. It is not something you can input.    Perhaps
people are confusing NTP's use of this term from the same term used in a
different context?

"root distance is equal to the root dispersion plus half the root delay"
This is a DEFINTION.  One does not calibrate a definition.  These are
measured values done in real-time.   It has nothing to do with how one
server connects to the others or how many source of time are used.  It has
noting to do wight eh number of network "hops" or if GPS or an uncalibrated
source is used for time.    All NTP cares about are the delay in round trip
ping time and the variance in those times.   I have seen NTP reject a local
GPS receiver in favor on an Internet connected pool server because the very
old handheld Garmin GPS had such poor timing.   This is not uncommon with
older nav receivers.


Read how it is used here
http://www.eecis.udel.edu/~mills/ntp/html/cluster.html
Read another definition of "root distance" here:
http://www.eecis.udel.edu/~mills/ntp/html/select.html




-- 

Chris Albertson
Redondo Beach, California


More information about the time-nuts mailing list