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

Unacceptable - Elevation Gain Off consistently

Dear Garmin,

I've tried waiting for solution and fixes. It started with my Fenix 6. I bought a 965 hoping it would improve things. It didn't. Every height relevant activity where I wear the watch on my wrist has severly wrong elevation measurements. Fenix does this since SW version 25.xx - Forrunner is *** from the start. Below is just an example of which I have plenty. My guess is that you have messed up algorithms for barometer sensitivty.

Very disappointed in your support. Do not believe that you focus on your for values anymore (function for athletes) but want to cater to every snob with too much money lying around who does not rely on accurate data. The correct data is confirmed with Komoot, Strava, Google. All very close to each other.

Keep up this *** and u will ruin yourself.

Sincerely your beta testers.

  • Add: what did Garmin tell me: 174.

  • Measuring elevation gain is not as simple as you might think, depending on various smoothing modalities you can get different values, and there is often no "right" and "wrong". In my experience (which is in the alps), my 965 does a good job at measuring the elevation, of which I do a lot. Garmin has many shortcoming when it comes to high elevation runs, but measuring the elevation is not one in my experience.

    To show the issue you are having, post the exact route (e.g. gpx), the elevation measured and the elevation expected - and tell us also why you expect this elevation. From your screenshots I guess garmin said 174m and strava said 93m? Looks like strava might be doing more smoothing and, garmin is measuring the small up and down on the route? At least on your garmin plot I see quite some up and down within the plot, not sure how real it is, especially without a gpx to look at it myself.

  • Hey Fredderic - thank you for answering. I am not usually this annoyed/aggressive when writing publicly. But I do not have the feeling that Garmin is really listiening anymore which is really sad. 

    Fully understand that elevation gain can differ sometimes and that smoothing has a lot to do with it. Thing is - this is completely new behavior. And this micro spikes in the hight graph are new as well (you can see how in the first graph there are these micro jitters in there... this is now true for 965 and Fenix and this messes up the calculation). 

    Here you have the plots: 

    New - Messed up elevation: 

    Old_example1: 

    Old_example2: 

  • Interesting. So these "micro spikes" in the graph seem to not reflect reality, as in you are not going up and down on this route? For reference here is a trail run activity I did today (recorded via a 965), I do not see any micro spikes at most places, only at a few that I think are real short up and down parts of my route:

    I also checked a cycle ride that I recorded with the watch (which I usually don't do, as I normally use the edge bike computer for cycling), similarly I don't see bumps like you:

    One idea that I have: The elevation is measured through air pressure (barometric), in combination with some GPS data to correct any offsets (usually only if you stop your activity, not during activity). So if something messes with the pressure sensor, you get weird elevation graphs - for example I put my watch ~50cm into a lake for a few minutes during a hike to measure the water temperature, I got a significant dip in elevation (higher pressure => lower down), and a spike back up when I took the watch back out.

    If you look at the backside towards the bottom of the watch, there is a small hole, which is the entry for the barometric pressure sensor. Maybe you are doing something during your riding that causes alternating pressures towards this hole? E.g. if you mount the watch on your handlebars the hole might sometimes be "in the wind" of your bike ride, causing increased pressure, and sometimes not. I would try and see if you find a well protected place for your watch that keeps the barometric sensor out of any potential "wind" (and also away from water, if that could be on your ride :D), and see if the micro spikes continue.

  • Thank you so much for looking into this. Very interesting! Funny enough - I've tested this with my Fenix strapped to the handle bar. No spikes. I'm almost certain it will be the same with the 965. What brings me to believe: they have messed up the algorithm which takes pressure into the calculation. I've not had these issues before a certain sensor hub update with the fenix (the 965 had it from the start).

    • Is there a way to reduce/ turn off the impact of the barometer during an activity?
  • Sensor ports are placed differently between Fenix and 965…

  • I do not know how to reduce the sensitivity of the barometer. Maybe it wasn't the algorithm that changed but rather the position/angle at which you use the watch on the bike? E.g. if you now have TT bars and now sometimes ride with your hands out front, or even if your preferred position on the "normal" road bike handlebars has changed?

    Since the barometer is at the back it could help to wear the watch more tightly, to avoid and wind going behind the watch into the barometer hole. An alternative would be to mount the watch on the handle bars as you said, then it will at least have a "stable" exposure to any winds, as it is always in the same position during the ride. You could also try wearing the watch on the "inside" instead of the normal "outside" of arm, as some people do. Or you could track your bike rides with a dedicated bike computer, I personally enjoy doing this as the screen is larger and at a better position to read compared to the watch - but I can understand not wanting to throw more money at the problem too ;).

    For a bike computer I can recommend the Edge 530, it is small and works decently, and it is cheap (currently around 200eur).

  • Bikes have remained unchanged. I haven't changed my position on the bike either. Reason why I don't use my Edge 530 is that this is my quick and ready to go training route for which the watch always has been enough (until software messed this up).

    I will try your suggestion with wearing it a bit differently although I really like to wear it as it was supposed to be (shocker :D).

    There is an option in the sensors menu where you can choose "only height" for baro and height meter. Die this only impact the weather warnings?

  • One thing you can do is to enable the "elevation corrections" for the watch. If you do that, then the barometric data will not be used, but instead it will just look up elevation information for your route (based on gps position coordinates) and determine it this way after uploading the route. Note this will not happen *during* activity, so if you care about the number it says while you ride, then this will not help you there. Normally, for watches with barometer, this is the inferior method, as it can lose small details (that are beyond the resolution or the digital elevation map), but it sounds as it might be the right thing to make you happy :). You can find it if you open an activity in the web version of garmin connect, on the right side you should see your watch with a drop down: