[nycbug-talk] GPS hardware for OpenNTPd

[Okan Demirmen]

On Wed, Nov 20, 2013 at 7:52 PM, George Rosamond
<george at ceetonetechnology.com> wrote:
> Okan Demirmen:
>> On Wed, Nov 20, 2013 at 6:35 PM, George Rosamond
>> <george at ceetonetechnology.com> wrote:
>>> Nikolai Fetissov:
>>>> Hey folks,
>>>>
>>>> Can anybody share any experiences with GPS receivers under OpenBSD/OpenNTPd?
>>>> I'm having hard time figuring out what  hardware to look for.
>>
>> nmea(4), mtms(4), endrun(4) ...
>>
>>> Not to distract, but . . .
>>>
>>> I'm a bit confused by the idea of OpenNTPD with a GPS receiver.
>>>
>>> I thought OpenNTPD wasn't concerned with precision as much as security
>>> and simplicity.
>>
>> Simple doesn't mean lack of features :)  timedelta sensors have been
>> supported for quite some time.
>>
>
> I'm sure the features are there and the hardware support.
>
> But if the hardware is there for serious accuracy, and openntpd isn't,
> then explain why they make sense as a duo.
>
> You're not getting the latency crap over the internet, but I don't
> understand why you'd then slack when it comes to the daemon then.

I don't understand what you're asking.... read ntpd.conf, noteable of
the below configuration:

sensor device [correction microseconds] [weight weight-value] [refid
string] [stratum stratum-value]

What openntpd might be missing are the features that like a tiny tiny
fraction of people need/use.  The ntp source might be this great GPS
device, but now one has an ntpd server servicing requests from some
random network...go hog wild on the gps side, but one can lose how
much of it once it "on the network"...?