Related
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Garmin Connect Course elevation gain underestimated

tauhammerhead
over 2 years ago

I'm having issues with plotting routes in Garmin Connect on a desktop. It is massively undercooking the elevation gains on trail runs meaning that the PacePro strategy I then sync to my Forerunner 945 is wildly off (and quite demoralising!).

In the below screenshot, the left elevation map is from Garmin Connect and the right hand one from OS Maps (I'm UK based). You can see the hills in the middle which Garmin Connect completely neglects. This isn't a barometric vs GPS issue like previous posts - it is just flat out poor contouring on the Garmin online map before I even get out running!

What map and what resolution contours are the elevations based off for course maps?

Top Replies

All Replies

  • 0 over 2 years ago

    UPDATE: After the run, my Forerunner 945 recorded 246m ascent - that is nearly double what Garmin Connect's Course had estimated. I feel the Forerunner is telling the truth as that is mightily close to OS Maps (which estimated 241m).

    Again, the issue is Garmin's maps.

  • +1 over 2 years ago in reply to tauhammerhead
    tauhammerhead said:
    Again, the issue is Garmin's maps.

    The issue is not the map, but the maximum number of Course points, which is set to 200 (250 at some devices). And since the algorithm only calculates the elevation gain between course points, the results will be the less accurate, the longer the course is. If you need a better estimate beforehand, you'd need to split the Course into several shorter ones.

  • 0 over 2 years ago in reply to trux

    The height gain should be c. 45m from bottom to top of one of the hills. My full course (8.5m trail run) says there is only 2m of climb on that hill.

    My route is only 8.5miles which seems small compared to some of the routes I bet cyclists plot - It only has about 10 points. Based on your reply, I tried plotting the course again with some more frequent turns/points at each intersection and peak/trough and the elevation gain is much better - still undercooked but much better.

    This result is surprisingly but I guess I need to plot my routes a bit more detailed!

    Thanks for the help!

  • 0 over 2 years ago in reply to tauhammerhead
    tauhammerhead said:
    It only has about 10 points

    There is a difference between way points and track points. The way points are limited to 50 on most devices, but those do not influence the resolution. It is the track points that do. A track point is every point of the course changing the direction or elevation of the course (even a tiny bit).

  • 0 over 2 years ago in reply to trux

    I think I understand what you are saying but still not sure why adding waypoints increases the accuracy of the elevation gain plot? See below re-plot of my route, this time with about 40 waypoints instead of the 10 used before. If you compare this to my maps above then it is much more like the OS maps version now.

  • 0 over 2 years ago in reply to tauhammerhead
    tauhammerhead said:
    I think I understand what you are saying but still not sure why adding waypoints increases the accuracy of the elevation gain plot?

    Because by adding waypoints you let less space for the optimizing of the trackpoints. They waypoints have to be kept as accurately as possible (unless you have more of them than the limit of your device allows), hence the track and the trackpoints can be optimized only between the waypoints. And of course, the more waypoints there are, the shorter the section between them are too, and the less optimization occurs.