BUG: When reversing a course, the course immediately gets longer by more than 50%

Hello NormalGarminUser and thank you for the details in this post.  Following the steps you outlined, I was unable to reproduce this issue.  Have you been able to duplicate this with multiple courses?

Yes, I tried it with two tracks and it happened with both. The effect is sometimes more and sometimes less extreme.

I just tried with one gpx track of which I made the screenshot above, to reproduce it with the following steps:

1.) Import the gpx track and select Course Type = "Running".

--> It shows as having 11.72 km and ascent 255 m and descent 252 m.

2.) Click "Freehand", and then "Reverse Direction". Wait about 10-15 seconds (it takes some time until the ascent/descent changes).

--> It shows as having 11.78 km and ascent 228 m and descent 230 m.

3.) Click "Reverse Direction" again. Wait about 10-15 seconds.

--> It shows as having 11.78 km and ascent 255 m and descent 253 m.

----------- ----------- -----------

I click "Discard" and repeat step 1 above again, new, and this time I get a different result:

1.) Import the gpx track and select Course Type = "Running".

--> It shows as having 11.72 km and ascent 230 m and descent 227 m.

2.) Click "Freehand", and then "Reverse Direction". Wait about 10-15 seconds.

--> It shows as having 11.78 km and ascent 254 m and descent 256 m.

...

and it goes on like this. Absolutely crazy!

I tried to attach this gpx to this message so you can try it, too, but I think I can only attach images and videos here.

Garmin-Kevin said:

What is the origin of this course?  You say you are importing it, but from where?

First I manually planned a course in Garmin Connect web app (browser) in "Freehand" mode. Then I exported this to my local PC to a gpx file, This is the file that I then (re-)imported as described.

Garmin-Kevin said:

I do not see the distance increasing more that 0.06 km in the latest example.

Yes, that's what I can reproduce all the time. 60 meters from just reversing the track is definitely a bug already and may point to the root cause of the much greater deviations that I also achieved, as described in my original post. By just reversing the course the difference should be very precisely 0.0 meters.

And as I said under "update 4" in my original post: The elevation diagram still shows 10.72 km and is thus inconsistent with the displayed total distance of 10.78 km, so there's something fundamentally wrong. Also, when I export and re-import the track, I get 10.72 km again (and also different ascent/descent values). So I propose to remove the root cause of that bug first, and then maybe the other problem with the sometimes much bigger deviation is also fixed.

Garmin-Kevin said:

Can you reproduce that % increase again using another course?

I did not try it. I tried the reversing only on the two course I described in my original post. I did not try any further reversings of other courses. You can probably reproduce it yourself by planning manually a course with many points and some way points on top. My course has 12 waypoints (8x crossing, 4x rest area) and the course itself has 2562 points.

NormalGarminUser I have not been able to reproduce this on any courses that I have created, though I do not have any courses with 2,562 points.

Instead of exporting and then re-importing (as you described below):

NormalGarminUser said:

First I manually planned a course in Garmin Connect web app (browser) in "Freehand" mode. Then I exported this to my local PC to a gpx file, This is the file that I then (re-)imported as described.

Let's see what happens when you DUPLICATE:

and then REVERSE DIRECTION:

Well, the duplication is what I had done at the very beginning - that's where I discovered the big bugs. I had a route already planned - then I duplicated it in order to reverse the direction.That's where the bug happened. As I described in my initial post.

In your case, the ascent/descent values also changed in an implausible way - albeit not with the huge side-effect that I faced in my original post acc. to the 2nd image after "Update 2".

I am adding a video to demonstrate that - obviously - the "reverse course" operation becomes much more buggy/crazy when the course contains (many) manually set course points. And also the length/number of points that the planned course consists of apparently plays a role:

[Video upload to this post did not work this time, so I uploaded it to my dropbox - link here]:

www.dropbox.com/.../Course-Reversal_Bug-Demo.mp4

Thank you for your participation in this forum thread.  So that I may assist you further, please reply with the following:

1. May we, if necessary, have permission to email you?
2. May we, if necessary, have permission to view/access your Garmin Connect account?
3. In what country do you live?