Climbed meters: Which system can I trust the most? How accurate is the fenix?

today i had a mountain bike tour with some of my friends. at the end one of them said, that we climbed a little more than 800 meters (he uses an iphone with runtastic). my fenix 5 told me it was 621 meters.

after the sync from garmin connect to runtastic i was told more than 800 meters as well (in runtastic). then i exported the gpx file from garmin to komoot and i was told 520 meters...????

is there a way to evaluate which might be the most accurate value?


i must say, that after the experience within the past weeks with my fenix 5, i trust the iphone (or any other device) more than the fenix. with the fenix i get so many sketchy and random values for different things, that i lost faith in the fenix.

  • Was this a few, large climbs or a lot of small climbs?

    There are two ways of measuring altitude and climbs: GPS altitude or a barometric altimeter.

    Your Fenix uses a barometric altimeter.
    An older Iphone uses GPS altitude.
    Newer Iphones (from 6S and up, I think) have a pressure sensor, but I don't know if Runtastic will use it for elevation, or it will still use GPS altitude.

    A barometric altimeter is very accurate for small, fast climbs. But for long duration climbs, or climbs with a large elevation difference it is less accurate.

    A GPS is not good for small, fast climbs, since they will be difficult to discern from the noise which is is always present in GPS altitude. But the absolute altitude will always be in the correct ballpark, which makes it more suitable for long durations or large elevation difference.
  • The Garmin calculation for climbing is not 'add up all the elevation increases' - it is more like 'do we seem to be going up hill for a sustained period? If so, work out how much climbing we are doing'. Strava has a similar but not identical method.
  • thank you for your explanations, but to be honest, i still don't get it.

    we were riding trails, so there were quite a lot of small up and downs. but i did not imagine that the climbed meters would differ by 25% - this is a huge difference and makes the numbers nearly useless!
    so the climbed meters shown in a device like iphone or garmin are more dependant of the calculation of each device/company than the reality?

    as i already wrote, gamin synced to my runtastic account and there i can get two different graphs for climbed meters. the graph with gps data is smaller than the other (i guess barometric altimeter). so if garmin uses the barometric altimeter, i would assume that garmin would show a higher value for the climbed meters than runtastic/iphone with gps.

  • As I wrote: GPS altitude has noise.

    If you ride at a constant altitude of 100 meter, your GPS altitude will have a lot of variation. One minute, it will show 90 meter, and the next minute it will show 110 meter.

    So if you climb a lot of 20 meter high hills, the watch/phone can't really know if you were climbing, or it was just GPS noise looking like a climb. So the software in the watch/phone will have to do some filtering of the data to cancel out some of the altitude noise. With no filtering or too light filtering, you will get a total climb which is higher than your true climb. With too heavy filtering, you will get a total climb which is lower than your true climb.

    So you can't really say that a device which uses GPS altitude will always show higher or lower climb. It depends on the choices made in the software which is running on that device.

    Anyway, on a trip like the one you described, I would trust the barometric altimeter over the GPS. (Of course assuming that you don't have a faulty pressure sensor - but that will usually result in errors of several km of altitude, so you will not be in doubt when that happens.)
  • AllanOlsen67, thank you again for your explanation. this sounds good in theory, but after all the strange behaviours and data on the fenix 5 i have seen in the past weeks, i have my doubts, that the fenix is showing correct values for the climbed meters. perhaps the fenix does it right in theory but ends up with wrong results.

    i feel, that my fenix is more like lucifer_fi described his observations.
  • Well, the climb on my Fenix 3 is far more accurate than anything I have ever seen from GPS. Your Fenix 5 may be different.

    But it is easy to test. You just need to ride a path with known altitudes of each high point and low point along the path. Then you will know the actual climbed meters and can compare them to whatever device you want to test.
  • you are right, i already thought about finding a path, where i can test it with known altimeters. i will see if i can find one.

    normally i am not to crazy about knowing absolute exact values about climbed meters, exact speed and so on... but if the fenix is unreliable in so many things and shows sketchy values, i can go on using my iphone with runtastic and gps (as before). then i don't want a quiet expansive device just for the fun of it and will try to give back the fenix.