[time-nuts] an interesting problem

[Poul-Henning Kamp]

In message <4D4D6BF6.8070202 at earthlink.net>, jimlux writes:

First, this isn't that different from the analysis used in the NTP
protocol, so you should read that with an open mind.

>Or, given that the interval between ticks is one 
>of 28 or 29 discrete values 

If you can reliably bin your measurements in these discrete values,
do that, and calculate stats for all bins with sufficient data
points to say something meaningful.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.