[ntp:questions] SNTP accuracy

David Taylor david-taylor at blueyonder.co.uk.invalid
Thu Dec 27 09:21:38 UTC 2012

On 24/12/2012 20:17, Ran Shalit wrote:
> Hello,
> I read in http://www.sematech.org/docubase/document/4736aeng.pdf that "SNTP
> provides accuracy typically within 100 ms".
> I would like to ask if this means that on a specific system combining two
> machines connected by ethernet (LAN), it can be any number *changing in
> run-time* within 100 msec or that we will reach more or less a constant
> delay number. This is very important because in case it is a constant
> number, even 98msec delay, we can always delete this delay time (98msec)
> and get a time synchronization around zero. Is that right ?
> Best Regards,
> Ran

Ran, as has already been suggested, why not use full NTP directly 
between the two systems?  If you are running Windows, you can get down 
to "few milliseconds" offset levels with care.  For example, see the PCs 
named "Hydra" and below on this page:


Exactly how close depends on which PC and which version of Windows. 
UNIX like systems, and Windows-8, can get much closer (see Portable PC 
Bergen which is even running over Wi-Fi).

Using a local GPS/PPS source can be even better - even on a low-cost 
computer like a Raspberry Pi:


Getting Windows to return a time closer than about 15 milliseconds 
requires extra programming.
Web: http://www.satsignal.eu

More information about the questions mailing list