Waypoint time stamp

[Christian]

Hi menion,
i let Locus generate a track with nav commands from Brouter.
After export there are timestamps in the waypoints like this (in bold):

<wpt lat="48.130144" lon="11.449368">
<ele>539.32</ele>
<time>2016-05-19T18:50:27Z</time>
<name>right</name>
<desc> <![CDATA[<font color="#ff000000"> <table width="100%"> <tr> <td> <table width="100%"> <tr> <td align="left" valign="center"> <small> <b>Höhe</b> </small> </td> <td align="center" valign="center">539 m</td> </tr>
<tr> <td align="left" valign="center"> <small> <b>Erstellt</b> </small> </td> <td align="center" valign="center">2016-05-19 23:24:38</td> </tr>
</table> </td> </tr> <tr> <td> <table width="100%"> </table> </td> </tr> </table> </font>]]> </desc>
</wpt>

I guess time tags and values are from Locus ?!?

Other analyses softwares declares that the track took 10 minutes.
Can we get rid of the useless (and in this case wrong) time stamps in waypoints? Like in tracks some times ago?

Thank you,
Christian

[Menion]

Good day Christian,

this won't be so easy. Time values are generated to trackpoints and waypoints to correctly match each other in case of multiple waypoints on a single trackpoint. After quite a lot of testing with @0709, I think that current system works and without serious reason, I do not want to dig into this.

So question is: are these times a serious reason that do not allow to use correctly exported GPX files?

[Christian]

Generated time is wrong. An error remains an error even if it fits in another system.

If you want to mark a waypoint with an identifier please do not use time stamps.

If i try to analyse a track without timestamps in trackpoints and the time overall is displayed as 10min, the further computation will fail due to the time stamps in waypoints.

[poutnikl]

Timestamps are not really needed for proper matching of hints and trackpoints, as they are redundant info even for multipass case. Coordinates and point order should be enough, if Locus uses pointers for both passed trackpoints and navigation points

Edit: If Locus is following the pointers above, no timestamp info is needed. If pointer info is destroyed for whatever reasons,  neither timestamp info helps, if we ended in ambiguous multipass part of the route. For non ambigious part, pointers can be easily reestablished without timestamps.

Timestamps would make sense if used with an added value of estimated point arrival ( relative to start ). For BRouter case, this can be done with the advantage of present elevation info.

[Menion]

Hi Willy,
may you please try what happen, if for your track select "activity type" "Cycling" or "Car", if there will be difference in times?

[Menion]

If I understand this chart correctly, then for car, there is not constant average speed around 15 m/s , but some weird peaks and also values close to 1 m/s.

Ah after test ... you are correct, damn. Tiny issue, will be fixed in next version.

Just to keep in mind: idea is here that generated tracks without a speed or time, will have generate times based on some predefined average speed for an activities (not all activities has currently default speed, then 1 m/s should be used).