[ntp:questions] Re: NTP precision

David J Taylor david-taylor at blueyonder.co.not-this-bit.nor-this-part.uk.invalid
Sat Sep 24 07:40:53 UTC 2005


Tom Smith wrote:
> Richard B. Gilbert wrote:
>> If I understand the definition of jitter, it has units of frequency
>> (1/T) and cannot meaningfully be added to or subtracted from time
>> (T).
>
> Regrettably, even that's not clear. See the definitions that Brad
> pointed out:
>
>  Jitter
>     Short-term variations in Frequency with components greater than
>     10 Hz. The estimated time error of the system clock measured as
>     an exponential average of RMS time differences.
>
> and the 4.2.0 documentation for ntpq:
>
> jitter
>     The estimated time error of the system clock measured as an
>     exponential average of RMS time differences.
>
> -Tom

Perhaps the second definition should read (with an explanation in 
brackets):

---------------------------------------
jitter
    The estimated time error of the system clock measured as an 
exponential average of RMS time differences.  (Jitter errors are caused by 
short-term variations in frequency with components greater than 10Hz).
---------------------------------------

Jitter would (in this case) be measured in time, not frequency, so it's 
unfortunate that the first definition mentioned frequency first, 
potentially confusing folk.


BTW: what are the errors caused by longer term variations called?

David 





More information about the questions mailing list