[ntp:questions] best gps receiver for time synchronization

Unruh unruh-spam at physics.ubc.ca
Thu May 28 03:30:39 UTC 2009

"hvengel at astound.net" <hvengel at astound.net> writes:

>On May 13, 10:25=A0am, Unruh <unruh-s... at physics.ubc.ca> wrote:
>> What the structure returns need have nothing to do with the accuracy.
>> Linux system calls can return a structure with nsec but the accuracy of
>> the linux clock is no better than usec.

>This has not been true since 2.6.26.  Linux since that time has had a
>nanosecond kernel and with the LinuxPPS patches and some other tweaks
>and a good reference clock will give you sub-microsecond offsets most
>of the time.  There is still work being done on the kernels time
>related functionality but as it matures it will be as good as anything
>out there now and will be comparable to FreeBSD.

Sorry, I think that the timing of the interrupts themselves is no better
than usec and certainly nowhere near nsec. Similarly branching into the
kernel routines and running them and setting the clocks is again hard to
do to nsec. 
All I was saying was that just because the time routines are capable of
recording time with nsec precision does not mean that that accuracy of
the clock is anywhere near that, and I have great difficulty believing
that the accuracy is any better than usec.

More information about the questions mailing list