[ntp:questions] What level of timesynch error is typical on Win XP?

Joseph Gwinn joegwinn at comcast.net
Fri Oct 22 02:43:46 UTC 2010


In article <i9pkvb$dc6$1 at news.eternal-september.org>,
 "David J Taylor" <david-taylor at blueyonder.co.uk.invalid> wrote:

> > The platforms in question are running Windows XP, not Vista or Windows
> > 7.  How does this change the answer?
> 
> NTP will switch on the multi-media timers in Windows XP and achieve the 
> best performance I have seen on Windows systems.  Temperature (or perhaps 
> CPU load causing temperature change in your case) seems to be the limiting 
> factor in the peak offset.
> 
> > By the way, the hardware is a collection of 8-core HP Z800 workstations
> > connected together by copper gigabit ethernet links and local hub, with
> > one link going via the company network to the GPS network time server.
> >
> >
> > Joe Gwinn
> 
> Beware the 1Gb/s adapter settings, as Dave Hart mentioned.  You may find 
> you get better timekeeping performance setting them to 100Mb/s.

Not really practical, as the sim folk copy large files back and forth.  

Nor does the time offset error show any diurnal cycle, whereas the sim 
folk work only during the day.

 
> You might consider providing a local, more precise NTP server with 
> something like a small, fan-less Intel Atom system running FreeBSD and 
> synched across the network to your GPS time server.  You might be able to 
> keep a small box like that in a more temperature controlled environment, 
> but even without it might provide a way of smoothing out any jitter due to 
> your remote connection to the GPS server.

I'm not convinced that this would help.  NTP reports a round trip time 
of slightly more than 2 mS, which is very close to the two milliseconds 
that ping sees, so it seems unlikely that the time server or intervening 
network is the root cause.


> Another thought is to send 1PPS across the network (an old RS-232 link, 
> perhaps, just used as a direct connection, not for serial data), and 
> duplicate the PPS feed to your remote server locally.  Then make an RS-232 
> level driver to send PPS to all your workstations.  They get the precise 
> seconds from the PPS, and the nearest-second time from the network.

We do exactly this kind of thing in the big radars, but this simulator 
is not going to get the hardware, even though IRIG would work.


Joe Gwinn




More information about the questions mailing list