[ntp:questions] NMEA ref.clock better than my ISP's timeserver?

Miroslav Lichvar mlichvar at redhat.com
Fri Jun 19 12:22:49 UTC 2009


On Sat, Jun 13, 2009 at 07:46:45AM +0000, Unruh wrote:
> Note that he finds that chrony is about a factor 20 better than ntp at
> poll level 4 (I think there is something wrong there) and about 2-3
> times better at poll level 3. I suspect this depends crucially on the
> thermal behaviour of the computer. 

It certainly does. I'm doing the tests in my office where on a sunny
day the temperature rises by several degrees in less than hour. Over
the day the frequency varies in about 0.8ppm range. This may be a bit
on the extreme side.
 
I did another round of tests, including NTP daemon loop configured to
use FLL. It performed really well, still not as good as chrony, but
much better than kernel loop using only PLL.

Offset RMS:

chrony poll 3		2.0 us
chrony poll 4		2.1 us
NTP poll 4 FLL		6.4 us
NTP poll 3		22 us
NTP poll 4		43 us

Distribution plot:
http://fedorapeople.org/~mlichvar/chrony/chrony_vs_ntp.png

-- 
Miroslav Lichvar



More information about the questions mailing list