[ntp:questions] refclock on Windows graphs

Unruh unruh-spam at physics.ubc.ca
Wed Mar 18 00:52:51 UTC 2009


"David J Taylor" <david-taylor at blueyonder.neither-this-bit.nor-this.co.uk> writes:

>Unruh wrote:
>[]
>> Maybe when someone finally put refclock support into chrony you will
>> have
>> your wish.

>Not until it runs under Windows.

Fair enough.


>> It sounds like your temp changes causing rate changes on a
>> scale
>> that ntp cannot properly cope with. (ntp is terrible at coping with
>> changes-- a purposeful design decision by Mills). I get about 3-5
>> usec as
>> average offset, but the freq changes are only about .1PPM , not 3PPM.
>> The
>> room is not particularly air conditioned, but the temp variations
>> over the
>> day are probably less than 5C not 20C.

>Here it's probably a variation of at least 10C.  The computer CPU use 
>itself is fairly constant throughout the day, so it may well track ambient 
>temperature:

>  http://www.satsignal.eu/mrtg/performance_feenix.php

>Another PC in the same room running a FreeBSD system has about half the 
>jitter, but far less offset - PC Pixie at the top:

>  http://www.satsignal.eu/mrtg/performance_ntp.php

That looks much more like the kind of timing I get on my Linux system.
driven by a PPS from a Garmin 18LVC.

I assume that the rate is also far more constant on that system. 



>I recall trying some of the motherboard temperature monitoring software, 
>but probably not for that PC.  I'll look again.

>David 




More information about the questions mailing list