Re: GPS Unit for Timekeeping on Linux?

From: W. Watson (wolf_tracks_at_invalid.inv)
Date: 11/15/04 

Date: Mon, 15 Nov 2004 11:08:33 GMT

Seagull wrote:

> W. Watson <wolf_tracks@invalid.inv> wrote:
>
>>Doesn't that imply that I'm in constant communication with some remote
>>device, or on a global network? I need the accurate time 7/24.
>
>
> Not necessarily constant.
>
> I guess it comes down to, exactly what kind of accuracy do you require,
> and for what application? GPS will get you time to within a second,
> but it's more expensive than occasionally syncing your PC to a network
> time server which can do the same. If you have a particularly nasty clock
> drift problem, sync more often (1/day instead of 1/week).
>
> Now, if you can't ever connect to a global network, then something like
> GPS is a good option. But we see a lot of people on here asking to
> use GPS for getting the time for applications where something like NTP
> can do the job for free. So, it's only natural that the regulars here
> question the initial assumption.
>
>
> Cheers,
> -+JLS
>
I have a meteor application that operates a camera from sundown to sunrise, and the
more accurate the times the better. Right now the drift is in the order of several
minutes per month. Synching once a day might do the trick, but I'd like to see it
automated, so that I wouldn't have to dial up the internet. If the application were
near my most frequently used computer, the one in my house, I'd probably be inclined
to manually update the time by the network every day; however, it is a building 100
feet away and going out there every day is not something I think about much.

I know there are manual techniques in Linux to refine the times, but I don't
presently have a lot of confidence in them. In particular, it appears if Linux goes
down unexpectedly, one gets to start all over with the refinement. My system seems to
crash about once every two weeks. I could be wrong about having to start almost from
scratch on the refinement. It's something I should explore. 

-- 
              Wayne T. Watson (Watson Adventures, Prop., Nevada City, CA)
                  (121.015 Deg. W, 39.262 Deg. N) GMT-8 hr std. time)
                   Obz Site:  39° 15' 7" N, 121° 2' 32" W, 2700 feet
             (Formerly Homo habilis, erectus, heidelbergensis and now sapiens)
               "The only title in our democracy superior to that of President
                is the title of citizen." -- Louis D. Brandeis
                         Web Page: <home.earthlink.net/~mtnviews>

Relevant Pages

  • Re: Garmin GPS-18 with prn 42 and 50, no DGPS.
    ... close to the associated reference network. ... there is other work going on in the area of GPS augmentation. ... Airservices Australia (AsA) has begun deploying GBAS and are also ... So why has the SBAS solution been abandoned? ...
    (sci.geo.satellite-nav)
  • Re: GPS Cellphone Tracking - Track Anyone Anytime Anywhere
    ... >GSM mobile phone tracking system via the GPS-TRACK satellite network. ... I am presuming that the phone constantly sends via GSM its GPS location. ... Some phones do have some GPS functionality built in but use the communication with the network to actually calculate the position of the handset. ... That is a form of tagging but mobile phone records can be used to work out roughly where the phone is. ...
    (uk.rec.walking)
  • Re: GPS Data Bandwidth Usage
    ... a Motorola Datatec 900 MHz network. ... transmit time on the field systems. ... GPS Settings within PEN.INI ... data to the UCAD server which creates RF traffic. ...
    (sci.geo.satellite-nav)
  • GPS shutdown pointless, as GPS+GLONASS systems exist...
    ... This kind of GPS shutdown is rather pointless, ... Bush Prepares for Possible GPS Shutdown ... Any shutdown of the network inside the United States would come under only ...
    (sci.geo.satellite-nav)