[ntp:questions] Re: SPECTRACOM (GPS) - no server suitable for synchronization found

David Woolley david at djwhome.demon.co.uk
Thu Sep 21 19:35:59 UTC 2006


In article <uni4h21t5stk689npfchl4j2nve9red3nr at 4ax.com>,
Brian Inglis <Brian.Inglis at SystematicSW.Invalid> wrote:

>  Stratum 0 report says 235, 1 says 236, 2 says 237: perhaps a bug when
> reporting positive precisions -- 2^35, 2^36, 2^37, doubling each
> stratum? 

That doesn't make any sense at all.  Typical values are around 2^-13, 
and the value should not be a function of stratum; it depends only
on the machine that is sending the value.

A correlation also doesn't make sense as printing the raw byte, as both
precision and stratum have a byte to themselves.

As an uninterpreted value, 235 would be -21, which is rather good:
2^-21 = 100ns, but not impossible.  236 and 237 are 1 microsecond and
10 microseconds, respectively.

I think it is printing the uninterpreted, signed, logarithmic value as
though it were an unsigned, linear, number of seconds, and the correlation
with stratum is coincidental.




More information about the questions mailing list