Ascent, doesn’t work?

1 lurker | 6 watchers
Feb 2022
1:51pm, 3 Feb 2022
16,888 posts
  •  
  • 0
larkim
Not sure if this is right, but this is my assumption about how ascent / elevation works on FE

Elevation graph in log - straight from watch data on a point by point basis, but used to be from map data so older runs (to about Nov last year I think) will use map data instead

Elevation summary in log (i.e. total ascent / descent) - calculated from watch point by point data by a bespoke algorithm written by our good lord himself. I *think* (happy to be corrected) that this new algorithm was retrospectively calculated across all runs, which means where map data is graphed, the elevation numbers may be out of sync with the graph (e.g. see this log of mine where the graph is spot on as it is map-derived, but the total ascent is miles out because the actual run data was also miles out) fetcheveryone.com/t-19003417

Climbs feature - straight from watch data, whether it was nonsensical or not. See above log where I started at about 67m (per graph), rose to about 90m over a 0.8m stretch, but which the nonsense underlying data reported I'd climbed 264m and an eye watering 21.8% climb which I did at 7m36 per mile :-)

I can definitely see an argument for using the .fit figure for total ascent / descent, though that runs into trouble if the run is trimmed or split as the site needs to have some means of calculating the figure independently.
um
Feb 2022
5:11pm, 5 Feb 2022
5,735 posts
  •  
  • 0
um
I'm struggling to understand the climb info derived from elevation.
Is it me, or is the display a bit skewed, eg from today


Why it there a 31m climb at the top? Why do some downhill sections show as climbs?
Feb 2022
9:01pm, 6 Feb 2022
16,905 posts
  •  
  • 0
larkim
That's odd. If you look at the elevation graph on gain connect does the profile match, or are the profiles quite different?
um
Feb 2022
9:28am, 8 Feb 2022
5,745 posts
  •  
  • 0
um
The profile is right for the run, very similar on GC for corrected and source elevation. Equally, I can't find, on the source GPX/Excel a 30.6m climb, well, there's the obvious ascent, but that has frequent dips to disrupt the full climb.
Feb 2022
9:39am, 8 Feb 2022
9,224 posts
  •  
  • 0
Fragile Do Not Bend
I’m getting the same sort of graphs as you um, roughly half of my ‘climbs’ are actually downhill and a lot of what look like really obvious climbs don’t register. My watch is too basic to have an altimeter, so according to Fetch I shouldn’t be getting any climb data anyway, so I’m confused 🤷‍♀️
Feb 2022
9:42am, 8 Feb 2022
9,225 posts
  •  
  • 0
Fragile Do Not Bend
Good example - 4 identical laps with a big climb in each.

Feb 2022
10:14am, 10 Feb 2022
16,976 posts
  •  
  • 0
larkim
Those elevation traces look like they must come from map-data rather than GPS data given how consistent they are. Which I thought Fetch said he had had to stop because of usage?

If you look on Garmin connect and disable the elevation correction (if it's turned on) do the graphs look like they match the figures the watch has collected (or do the climbs look like they do?)?
Feb 2022
10:31am, 10 Feb 2022
27,167 posts
  •  
  • 0
fetcheveryone
Those of you with dodgy data, please can you send me feedbacks, with links to your examples? Ta.
Feb 2022
11:49am, 10 Feb 2022
9,238 posts
  •  
  • 0
Fragile Do Not Bend
I’m fairly sure my watch doesn’t do elevation (FR45S) so must be map data. I can’t find an option in GC to turn off or on elevation correction, maybe that’s watch dependant?

I will send feedback Fetch.
Feb 2022
3:06pm, 10 Feb 2022
16,983 posts
  •  
  • 0
larkim
I think it should be available for all Garmin watches, but that it's only evident on the webpage rather than in the app (I could be wrong!).

On my view of GC it's on the RHS under the image of the watch. A toggle here flips between the data the watch has recorded (either from GPS or from altimeter) or map data.

About This Thread

Maintained by Bowman 🇸🇪
It has been a little wonky for sure, but fetch made som changes a while a go and now nothing seems to be right?

Today I went for a trail run with some hills, I got 1m here, Garmin says 354m.
Same last few runs.
I have changed watch though, don’t know if it’s something to do with that.

I know that the ascent on GC is not the truth either, but it’s always consistent though. With my new watch to.

Anyone else having problems?

And, it would be nice to be able to let the ascent use...

Related Threads

  • ascent
  • website








Back To Top

Tag A User

To tag a user, start typing their name here:
X

Free training & racing tools for runners, cyclists, swimmers & walkers.

Fetcheveryone lets you analyse your training, find races, plot routes, chat in our forum, get advice, play games - and more! Nothing is behind a paywall, and it'll stay that way thanks to our awesome community!
Get Started
Click here to join 113,372 Fetchies!
Already a Fetchie? Sign in here