[ntp:questions] ATOM falseticker with flag3 enabled
hart at ntp.org
Sat Mar 31 18:18:47 UTC 2012
On Sat, Mar 31, 2012 at 18:12, A C <agcarver+ntp at acarver.net> wrote:
> I understand that the jitter will change but I'm talking about the minimum
> possible jitter that ntpd can report for a given system. Using 4.2.6, the
> minimum possible jitter reported (such as what is shown in the ntpq
> billboard when ntpd first starts up) was 0.061 on my NetBSD system.
> However, with 4.2.7p259 (and some of the earlier versions as well) the
> minimum possible jitter exactly doubled to 0.122 on the same system.
> It would appear that the method used to compute the minimum possible jitter
> changed between the two versions.
That value is stored as a power of 2, explaining the exact doubling.
The method did change to separately determine the minimum apparent
time to read the clock (even when the clock reading does not change
every time) from the minimum change between readings. There is extra
code in the clock reading path to ensure readings never go backward
without notice, as well.
More information about the questions