[ntp:questions] best jitter on qnx always = 0.977

Николай Орехов nowhere at mail.ru
Tue May 10 18:11:41 UTC 2011


Well, I'm using PPS driver which measures ticks with presision more than one
millisecond using ClockCycles and system interrupt.
Even with ref clock jitter less than 1ms I got 0.977 jitter.
I'm agree that system jitter is closely connected to presision of system
clock. But reference clock jitter may be even lower.

Can i somehow convince ntpd that this refclock jitter can be less than 1ms?
Will it influence sync presision? For now i get offsets ranging between
+-0.5ms

2011/5/10 Dave Hart <hart at ntp.org>

> On Tue, May 10, 2011 at 16:22 UTC, Николай Орехов <nowhere at mail.ru> wrote:
> > Why on QNX I always get the same best jitter 0.977 even on PPS ?
>
> I see the same thing on a laptop running Windows Vista.  This likely
> indicates ntpd measured your system clock precision as 2 ** -10
> seconds, or 1/1024th of a second.  ntpd uses the precision of the
> local clock as the minimum jitter, because every local timestamp
> involved in its calculations is trusted only to the level of the clock
> precision.
>
> Reading the clock probably costs less than a millisecond on your
> system, but the value read changes only once per millisecond.  If you
> could convince QNX to provide a higher-precision system clock, you can
> drive the precision down and bring your jitter figures with it.
>
> Cheers,
> Dave Hart
>
>



More information about the questions mailing list