[ntp:questions] Local Time NTP Server

Dan Geist dan at polter.net
Mon Aug 24 16:37:47 UTC 2020


----- On Aug 24, 2020, at 11:46 AM, Jakob Bohm jb-usenet at wisemo.com.invalid wrote:

> On 2020-08-24 16:07, William Unruh wrote:
>> On 2020-08-24, Jakob Bohm <jb-usenet at wisemo.com.invalid> wrote:
>>> On 2020-08-24 12:51, Beth Connell wrote:
>>>> Hi,
>>>> I'm struggling to find any information on where the free NTP servers are
>>>> geographically based. In particular, I'm wondering where Facebook, Google,
>>>> Microsoft, etc are based within the UK. Just for curiosity, I'm wondering how
>>>> this affects any interference to my location.
>>>>
>>>> Thanks
>>>>
>>>
>>> All NTP time is GMT (now named UTC, after HM government dropped the
>>> ball).  The only geographic factor is the "ping time" to the time
>>> server, the servers that are the most distant (longest ping) are the
>>> GPS and Galileo satellites, that are hundreds or thousands of km from
>>> the receiver, and use their own (non-NTP) protocols to correct for the
>>> delay.
>> 
> 
> I used "ping time" as the popular name of the round trip time (as
> actually measured by NTP implementations), and thus as a somewhat
> vague reference to the equivalent satellite-to-ground time delay.

FYI, GNSS satellite to ground delay compensation has very little/anything to do with NTP. When a timing GNSS receiver comes online, it trains to each visible bird for a period of time (an hour or more, ideally), allowing that signature to move through the entire field of sky. This allows the timing algorithm to compensate for mild doppler shift as the sources are coming/going (since they're not Geo-stationary). Each source is, however, each at fixed (and published) orbital elevation. Triangulation allows a receiver to determine its relative elevation. Time delay from each source is derived from that and normalized. That timestamp (and phase) are provided to downstream electronics, including NTP server software.

There are definitely differences in GNSS performance (chipsets, algorithms, constellations), but NTP isn't really accurate enough to make many of those deltas relevant. As long as the network path between any given NTP source and your NTP client is fairly symmetrical and absent of much contention (jitter), the time will be quite accurate. Raw network distance/latency isn't even the biggest determining factor in its accuracy.

Thanks.
Dan   


More information about the questions mailing list