[time] GPS Time Server Survey

Chris Kuethe chris.kuethe
Thu Apr 24 21:57:44 UTC 2008

On Thu, Apr 24, 2008 at 2:19 PM, Kipton Moravec <kip at kdream.com> wrote:
>  The problem with putting it in the PC is that the PC is most often not
>  near the sky. GPS needs some view of the sky.
>  You loose a lot of signal with a long antenna lead, that is why you
>  rarely see more than a 15 foot antenna lead.

Bah! I've got 3 private servers with 65ft of cable: 25-28dB patches,
15ft RG-174, 50ft RG-58/U. Works like gangbusters.

>  Just a 32-bit microcontroller with a good crystal and a GPS. All running
>  linux. No keyboard, no monitor, just ssh to it over the Ethernet. Not
>  that exciting.

Don't forget a serial console. Or have some way to set the IP over the
wire... possibly by sending a ping to it's ethernet address.

>   This is 30 times more accurate than the Garmin, but the PC is no where
>  near that accurate, so it does not matter.

Yes, I see that, I'm just observing that once you make one of these
things talk across a LAN you're going to be competing with all the
other traffic out there, and subject to all kind of variation and
jitter... that being said, I like good clocks.

>  All that will be possible, you could have access to the NEMA data.
>  Is it important?


quite a number of receiver chips have 2 ports, some have up to 4
(sirf). thus, you can gratuitously spew NMEA data and PPS on port 1
and various flavors of "control" traffic on the second port. this
could be handy if you like snooping around inside the chip, if you
want to get raw data, ... that'd be nice.

if that's too much work (and it could very well be, given your target
market) at least some fluffy little web interface would be nice.


GDB has a 'break' feature; why doesn't it have 'fix' too?

More information about the pool mailing list