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

David J Taylor david-taylor at blueyonder.co.uk.invalid
Wed Oct 20 13:22:27 UTC 2010

> I have a small network of Windows XP (64 bit) running simulations, with
> NTPv4 running on all the boxes and using a GPS-based timeserver on the
> company network.  The ping time to the server is 2 milliseconds from my
> desk, but I'm seeing random time errors of order plus/minus 5 to 10
> milliseconds, based on loopstats data.
> This level of timesynch error is OK for the simulation, but still that's
> a lot of error.  I get far better on big UNIX boxes.
> The question is if this level of error is reasonable, given the setup.
> I know that timekeeping under Windows is not optimum, but cannot change
> the OS, so the question is if I have gotten things as good as they can
> be, or should I dig deeper.  One thing that comes to mind is to raise
> the priority of the NTP daemon to exceed that of the simulation 
> software.
> Thanks in advance,
> Joe Gwinn


This is the performance I see:


The XP systems are:

  Feenix: GPS-synched
  Narvik: LAN-synced to Pixie (FreeBSD with GPS source)

Your best bet would be to add a GPS source to your Windows PC, when you 
might expect errors of less than 250 microseconds under stable running 
(i.e. leave the PC on 24 x 7).  If you can't do that, PC Narvik suggests 
you might get within +/- 1.5ms.  That's with a configuration file like:

server A  iburst  maxpoll 5
server B  iburst  maxpoll 5
server C  iburst  maxpoll 5

where A, B and C all have a GPS source.  All PCs on the same switch, so a 
much better ping than 2ms.  You could reduce the maxpoll further to 4 (if 
the server operator agrees) and get somewhat better performance, and 
keeping the PCs in a stable temperature environment would also be likely 
to help.  The bumps at 05:00 are when the heating comes on.

In my experience, changing the priority of NTP doesn't help a lot, but 
most of my PCs are not CPU-bound.  But I have given the account the rights 
to do that.


More information about the questions mailing list