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

GPS Accuracy

So it begins.

I will have mine Fenix 5 on Saturday and will start doing comparisons to an Ambit 3 Peak. I don't have an F3 to directly compare to as of now.

Anyone have an F3 and F5 to compare?
  • Looks like some spots where Pace and Cadence dropped out, or you stopped. Did you have any devices connected during the run / walk? There have been instances where dropouts from Ant+ devices (particularly Stryd, but also HRM-Tri in at least one case) have affected the distance accrual on the fenix 5 that I have seen.

    If not then my explanation would be that on a tortuous path at lower speeds in forest conditions it is a difference in the algorithms used by GC and MGPSF.


    Thanks for taking a look. I did stop a few times, once to watch some wildlife and a couple of other times to stretch a healing hamstring. No other devices, just the two watches.

    I figured this would put the 5x to the test since it was a really slow pace on twisting trails, which was particularly problematic for the F3 as well. If it's Garmin's algo, it's not very good in these conditions and is really frustrating as the actual GPS track distance from the watch is good. To report 4/10ths of a mile off over 7 trail moderate trail miles is pretty sad. Once my hammy heals up, and while I can still return the 5x, I'll hit a true mountain run that includes some canyon portions as well as slow power hiking steep climbs. Given this run's results, i'm not too hopeful.

    Maybe I'll try with GLONASS off to see if that helps somehow.
  • No more GPS accuracy comparisons from me as I sent mine back after two firmware updates that have not addressed the issues Inoticed vs my FR235 and the similarly featured FR935 doesn't have them so it's hard not to conclude that it's not a hardware problem...
  • fenix 5 Hiking

    Interestingly I took a look at a hike I did yesterday, so slow pace and challenging environment with a lot of switchbacks in some sections. What I found is that MyGPSFiles is overestimating the distance, possibly due to noise in the track. The way I assessed this was I plotted my exact route in Strava and imported that GPX file as well as my Garmin Connect file in to MyGPS Files:

    http://www.mygpsfiles.com/app/#FtWqVxpu

    Here is the Garmin Connect link: https://connect.garmin.com/modern/activity/1669929296

    and the Strava Route: https://www.strava.com/routes/8241917

    so to summarize:

    Garmin Connect and fenix 5 Distance: 6.43 miles
    Strava Route Distance: 6.3 miles
    fenix 5 GPX loaded in to MyGPSFiles: 7.0 miles
    Strafva Route GPX file loaded in to MyGPSFiles: 6.5 miles

    Looking at this comparison I am confident that the fenix 5 and Garmin Connect distance are accurate and Garmin is filtering this properly to arrive at the correct distance accrual.

    Thinking again about your data, it actually may be that the fenix 5x is correct and the filtering on the Suunto as well as MyGPSFiles on this type of data is giving more distance than actually covered. Might be worth plotting you exact route in you system of choice and see what distance that gives.
  • No more GPS accuracy comparisons from me as I sent mine back after two firmware updates that have not addressed the issues Inoticed vs my FR235 and the similarly featured FR935 doesn't have them so it's hard not to conclude that it's not a hardware problem...


    See you on the 935 forum :) Mine arrives in 2 days and if the issues with Stryd are not present on the 935 (as I assume they won't be) I will be keeping the 935 and sending my f5 back as well. Too bad, for me I can now say GPS works quite well at least as far as what I am most interested in, but cannot continue to use without getting Stryd data as I manage all my training now with running power.
  • No more GPS accuracy comparisons from me as I sent mine back after two firmware updates that have not addressed the issues Inoticed vs my FR235 and the similarly featured FR935 doesn't have them so it's hard not to conclude that it's not a hardware problem...


    I too am actually considering sending it back. The GPS track is generally acceptable, but I somehow can't convince myself to pay more for same HW, same functions but worse GPS performace compared to FR935. Also I'm quite annoyed that Garmin refuses officially comment on this issue.
  • Wish I didn't really prefer the look of the F5 over the 935 :/

    Otherwise I'd be going with the 935 without a doubt as well, but for now gps seems good enough for me for the most part so I guess I'll keep it for now. Oh yeah, also the fact that I can't even buy the 935 here yet plays a part ;)
  • Interestingly I took a look at a hike I did yesterday, so slow pace and challenging environment with a lot of switchbacks in some sections. What I found is that MyGPSFiles is overestimating the distance, possibly due to noise in the track. The way I assessed this was I plotted my exact route in Strava and imported that GPX file as well as my Garmin Connect file in to MyGPS Files:

    http://www.mygpsfiles.com/app/#FtWqVxpu

    Here is the Garmin Connect link: https://connect.garmin.com/modern/activity/1669929296

    and the Strava Route: https://www.strava.com/routes/8241917

    so to summarize:

    Garmin Connect and fenix 5 Distance: 6.43 miles
    Strava Route Distance: 6.3 miles
    fenix 5 GPX loaded in to MyGPSFiles: 7.0 miles
    Strafva Route GPX file loaded in to MyGPSFiles: 6.5 miles

    Looking at this comparison I am confident that the fenix 5 and Garmin Connect distance are accurate and Garmin is filtering this properly to arrive at the correct distance accrual.

    Thinking again about your data, it actually may be that the fenix 5x is correct and the filtering on the Suunto as well as MyGPSFiles on this type of data is giving more distance than actually covered. Might be worth plotting you exact route in you system of choice and see what distance that gives.


    That is interesting. I've got another trail run tomorrow, so I'll try do another test and also plot it and compare. Thanks!
  • I was quite happy with GPS from my F5, a little better than F3 which was ok. but my advice is : Use Smart recording instead of 1 sec and GPS will be better!

    My 2c experience, but from now on I am going to use that.
  • I was quite happy with GPS from my F5, a little better than F3 which was ok. but my advice is : Use Smart recording instead of 1 sec and GPS will be better!

    My 2c experience, but from now on I am going to use that.


    To each their own, but I would recommend against this as while it may result in what looks like a smoother track, you run the risk of throwing out data that might be good data for no reason and this may end up in under accumulating distance.
  • Ah well, just to keep the tradition going... Did 9.2k run this evening with both the 735 and the F5 (1sec, gps+glonass).

    Mygpsfiles:
    http://www.mygpsfiles.com/app/#Fe1XajiZ

    F5 GC:
    https://connect.garmin.com/modern/activity/1671813948

    735XT GC:
    https://connect.garmin.com/modern/activity/1671824099

    The 735 was most correct most of the time but overall they changed places quite a bit... Think the F5 did slightly better in the first half while the 735 was better in the 2nd. Note, 6k-7k I just walked and the F5 did pretty great in that spot actually, was surprised about that one.

    End distances are identical even though 1k laps were not (max diff was about 30m or so).