Hi--
On Jan 15, 2013, at 10:18 AM, Edward T. Mischanko wrote:
> It means that successive calls to the gettimeofday function return with
> a difference that is not less than 2^-15 sec.
> Thanks for the reply and the information. Is -15 good, bad, or average?
Out of ~100 emails from this group listing precisions in the modern format
(ie, NTP-4.2.x), a histogram [1] of results is:
46 precision=-19
16 precision=-20
12 precision=-14
8 precision=-18
4 precision=-8
3 precision=-15
3 precision=-10
2 precision=-16
1 precision=-21
1 precision=-13
1 precision=-19
1 precision=-10
One might do better using 'ntpq -crv' against published NTP servers or pool
members to collect data.
Regards,
--
-Chuck
[1]: grep -hr 'precision' _IMAP_storage_ | sed 's/3D-/=/g' | sed 's/> //' | sed
's/>//' | sed 's/==/=-/' | awk -F, '{print $1}' | histogram.py | grep
'precision=-'
...and then trimmed a few stray matches, such as the Subject: header of this
thread.
_______________________________________________
questions mailing list
questions@lists.ntp.org
http://lists.ntp.org/listinfo/questions
