[ntp:questions] Re: precision on different release

FEI WU fxw115 at psu.edu
Tue Jan 20 21:04:22 UTC 2004


Hi, Terje:

I still get the precision of -31 on my running server with ntp 4.1.1b-a. The
other server with exactly same hardware configuration gets the precision of -19
with ntp 4.2.0. 

I guess the developer change the way to calculate the value of precision,
obvious the -18 makes more sense on an old laptop than -31. Just want to make
sure everyone shares the same experience.

Thanks.

Fei  

On Mon, 19 Jan 2004 23:42:01 +0100, Terje Mathisen wrote:

> Fei wrote:
> 
> > hi all:
> > 
> > I notice that the precision ( if you command sysinfo in ntpdc) has decreased
> > after I upgrade the ntp from 4.1.1 to 4.2.0. There is no other changes on
> > hardware and system. However, the precison has dropped form -31 to -18.
> 
> A precision of -31 corresponds to 0.5 ns (or thereabouts), which is 
> still unattainable for nearly everyone except the national timing labs 
> that define what UTC is.
> 
> -18 otoh is about 4 us, which is doable on most unix systems with a good 
> refclock.
> 
> Terje
> 
> -- 
> - <Terje.Mathisen at hda.hydro.com>
> "almost all programming can be viewed as an exercise in caching"
> _______________________________________________
> questions mailing list
> questions at ntp.org
> http://mailman.ntp.org/mailman/listinfo/questions
> 
> 





More information about the questions mailing list