Re: Synchronising PC clock on the GPS time using GPS receiver
From: Jon Parmet (jon_at_parmetpc.volpe.dot.gov)
Date: 08/30/04
- Next message: Sam Wormley: "MINIATURE ATOMIC CLOCK MAY REVOLUTIONIZE PRECISE NAVIGATION"
- Previous message: dold_at_Synchronis.usenet.us.com: "Re: Synchronising PC clock on the GPS time using GPS receiver"
- In reply to: John Navas: "Re: Synchronising PC clock on the GPS time using GPS receiver"
- Next in thread: Dave Patton: "Re: Synchronising PC clock on the GPS time using GPS receiver"
- Messages sorted by: [ date ] [ thread ]
Date: 30 Aug 2004 09:36:49 -0700
John Navas <spamfilter0@navasgroup.com> wrote in message news:<IoOXc.10265$54.144165@typhoon.sonic.net>...
> [POSTED TO sci.geo.satellite-nav - REPLY ON USENET PLEASE]
>
> In <f0701dba.0408270700.21ea6aea@posting.google.com> on 27 Aug 2004 08:00:10
> -0700, jon@parmetpc.volpe.dot.gov (Jon Parmet) wrote:
>
> >John Navas <spamfilter0@navasgroup.com> wrote in message news:<OYtXc.9906$54.142640@typhoon.sonic.net>...
>
> >> You don't, but that's not NTP -- that's NMEA.
> >
> >My understanding may be wrong, but I thought NTP came with a plethora
> >of device driver options for different GPS devices and different
> >interfaces. So you'd configure it for Garmin whatever on Serial port
> >wherever using NMEA.
>
> NTP per se is the standard time protocol used on the Internet. What's
> confusing is that ntp.org chose to call their reference software "NTP", which
> made about as much sense as calling a browser or webserver "HTTP". It happens
> that NTP-the-software can use NMEA to obtain GPS time information, but that
> has nothing to do with NTP-the-standard. It's also massive overkill if all
> you want to do is set the PC clock using the NMEA output from a GPS.
>
> >Please correct any mistakes I'm making, because I'm interested in how
> >the flow actually works here. NMEA is the protocol used in delivering
> >the data from (in this case) the serial port.
>
> From the GPS over a serial connection to the PC (by means of a serial port).
>
> >But after that, NTP
> >actually takes care of serving out the time.
>
> If serving is what you want. FWIW, I personally run my own NTP network
> server, but I simply sync it from a public Stratum 2 server, rather than using
> a local GPS. I can't say for sure, but I'm guessing that NTP from a Stratum 2
> server might well be more accurate -- there's a fair bit of uncertainty and
> delay in an NMEA data stream.
>
> >Ok, so in the case of a
> >standalone box, the NTP Client would be on the same machine as the NTP
> >Server.
>
> Neither NTP client nor server would actually be used, just clock setting from
> NMEA input. In other words, most of the code would be unused, which is what I
> mean by massive overkill.
So, if I understand you correctly: If I can use NTP to get time from
the receiver and set the system clock. And, if at a later point in
time, I decide I also want to serve out the time, I'm already most of
the way there. I only need one 'application' to worry about.
Thanks ;)
> >I've never worked with a Windows' solution.... nor NMEA for that
> >matter in this type of application ;)
> >
> >We were running under Unix, used a Trimble SVeeSix, TSIP protocol to
> >acquire GPS Time, then simply called settimeofday(2) to effect the
> >change in system clock time. The accuracy requirements we had for time
> >were no where near what NTP provides, so the code was very
> >simple-minded (it didn't account for any latencies).
>
> Which is why I personally use NTP from a Stratum 2 server.
Which works if you have an Internet connection, I guess. This
particular machine was, due to mission critical needs, prevented from
being able to talk to the outside world. The GPS receiver happened to
be connected to the machine for the initial purposes of being a
reliable source of the GPS Almanac. As I implide above, we didn't need
picosecond levels of accuracy, so I threw together a quick solution
that worked given the constraints I had to work with.
Regards,
Jon
- Next message: Sam Wormley: "MINIATURE ATOMIC CLOCK MAY REVOLUTIONIZE PRECISE NAVIGATION"
- Previous message: dold_at_Synchronis.usenet.us.com: "Re: Synchronising PC clock on the GPS time using GPS receiver"
- In reply to: John Navas: "Re: Synchronising PC clock on the GPS time using GPS receiver"
- Next in thread: Dave Patton: "Re: Synchronising PC clock on the GPS time using GPS receiver"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|