When my Garmin (eTrex Legend HCx) loses signal, it continues logging as if I
had continued in a straight line for 30 seconds before either regaining
signal or giving up and stopping logging, and I imagine you are seeing
something similar.

I see this all the time when going into tunnels and worked around it by
postprocessing the logs to drop any run of entries with identical or
near-identical latitude and longitude deltas (to within 0.000002 degrees)
leading up either to the end of a log or to a sudden change in speed
(increasing or decreasing by 10mph in a single second).

Eric


On Wed, Apr 29, 2009 at 4:39 PM, Jim Ley <[email protected]> wrote:

> Hi,
>
> Last week, and tuesday I did a cycle circuit race, previously (couple
> of months ago the tracklog has been
> pretty flawless, but both the last two times it was interestingly and
> consistently
> wrong on every lap, the same errors also happened on my girlfriends
> Garmin.  The logs were both recorded on a Garmin 705, you can get a
> gzipped version of the tcx file here
> http://jibbering.com/2009/04/screwy.tcx.gz (or screwy.gpx.gz for a gpx
> version of it)
> I also uploaded it to http://connect.garmin.com/activity/3973718  From
> which you can see that the
> Garmin recorded the distance log right (as the speed graphs are pretty
> much correct) but the GPS positions were all way off.
>
> What happened was essentially that in a number of areas on every lap,
> the log went crazy, and instead of following where I went, it went in
> a weird straight line off course.  In the map view of it
> http://jibbering.com/2009/04/screwy.jpg or on the
> satellite view on the garmin page you can see the circuit clearly, and
> previously the logs have been accurately following it, but in both the
> last two weeks races
> it was just in a couple of parts of the course where it was on course,
> the other times being peculiarly off course, with weird straight lines
> - not just the ones where it suddenly jumped.
>
> Anyone have any insight into why this might've happened?  Particularly
> to two seperate devices.
>
> Also if anyone can think of a relatively simple way to "fix" the log -
> to a reasonable approximation I would be happy.
>
> Cheers,
>
> Jim.
>
> _______________________________________________
> Geowanking mailing list
> [email protected]
> http://geowanking.org/mailman/listinfo/geowanking_geowanking.org
>
_______________________________________________
Geowanking mailing list
[email protected]
http://geowanking.org/mailman/listinfo/geowanking_geowanking.org

Reply via email to