[ntp:hackers] hackers] Intel's Galileo as NTP server?

juergen perlinger juergen.perlinger at t-online.de
Sun Oct 6 09:39:02 UTC 2013


On 10/06/2013 08:49 AM, Hal Murray wrote:
> david-taylor at blueyonder.co.uk said:
>> Using "normal" PCs I'm not convinced that you would see any difference
>> between the slightly more accurate 6T and other GPS modules.  We are looking
>>  at the microsecond level in PCs (perhaps even greater in ARM boards), but
>> the differences are in the one-tenth microsecond level.  Having said that,
>> no I haven't played with the other boxes you mention. 
> I think the PPS an ARM based system could be more accurate than a PC.
>
> The trick is that most ARM SOCs have counter/timers that can be used to 
> capture the time without the delays of the interrupt processing.  The delay 
> itself isn't a problem.  It's the noise/jitter in the delay.  The counter 
> avoids that.
>
> I don't know if any code does it that way.  If not, it would be an 
> interesting experiment.
>
I think that was why some SOEKRIS boards made/make excellent time
servers -- they have pins available on the mainboards that could be uses
as capture signal inputs. OK, you could go the full way and even replace
the crystal with an oven stabilized frequency synthesizer, and if you
went far enough, even with a rubidium-stabilized frequency normal. Doing
*that* pushes things to a higher level.

But as Hal stated, the primary precision limit for regular PC hardware
is the interrupt latency time. If you can capture HW counter with a
hardware trigger, the results are bound to be better by at least an
order of magnitude, probably two. If you can even capture the counter
that's used to implement the clock, the results will be awesome. IMHO,
you will reach the level where the difference between timing and
position modules becomes definitely noticeable.

I wish I had more time to tinker...


More information about the hackers mailing list