Related

Awful GPS track, any fix incoming?

ncelik
over 4 years ago

Based on a few weeks of observations, GPS track on the Fenix 6 Pro is absolutely horrendous, probably the worst I have had on any device. Yet, surprisingly the total distance displayed is about right.

Case in point, I recently I ran a marathon in mostly open/forested area in Maine. Most other GPS devices had no issues, it was easy reception. Check out the traces, I picked a dozen or so other runners around me in Strava flyby; the 6 Pro is the black line

https://imgur.com/a/Lkw6ooH

It's not just these three areas, during the entire 26 miles it was almost always the outlier. It is BY FAR the WORST GPS track of all of them. The guy I was running with had an ancient TomTom and his trace was pretty much the entire time on the road, spot on. Yet, this brand new, premium priced device from the market leader MISERABLY FAILS in the one thing it should do well, GPS tracking. It’s sad and I feel like a fool for purchasing it.

Are we aware of any incoming fixes for this? I have a few more days before I have to return this to REI.

  • 0 over 4 years ago

    These threads are always so interesting... First of all, what do you use the track for afterwards? To email Google, Here, Microsoft or which ever map vendor you render the track on an complain about the accuracy of the map or to get a clue of where you actually ran? I want to know where I went within reasonable limits.
    The other argument would be absolute accuracy on the pace, which of course would require 100% GPS tracking. If pace is that important at a single point of time I would suggest running with your eyes on the shortest and most effective route instead of keeping the eyes on the watch. Pace is something that can be analyzed afterwards for reference runs that is done frequently for a training purpose.

    To get back to the GPS accuracy, I have two done two trail run competitions with my F6X Sapphire. Both with good accuracy:

    https://connect.garmin.com/modern/activity/4081333616
    https://connect.garmin.com/modern/activity/4056991105

    Ofcourse we can do the strava flyby example on both, I'm the black...

  • 0 over 4 years ago in reply to AndreiDracul

    Garmin has to use software AI simulators to test, as almost all real word testing is nowhere near what Garmin sees.  I mean, even their software engineers who i've been in contact with seem to have no idea why the problems we see are happening.   A lot of companies use Software AI to test their programming, but for a fitness device, they need data from users watches that are out and using them.

  • 0 over 4 years ago in reply to AndreiDracul

    What do you expect, the accuracy of GPS is ~3m

  • 0 over 4 years ago in reply to TobiasLj

    The saving grace is that the distance (and hence the pace) was reasonably on track. The first mile was 0.1 mile too long and then it slowly creeped up another 0.1 over 25 miles as expected (you never do perfect tangents).

    That said, I did a run yesterday on a route that I have dozens and dozens of runs with various GPS devices, generally all within 0.05 miles. Quite open area. Compare the run yesterday with my Fenix 6 Pro https://www.strava.com/activities/2779047515 and then the same run done with my older Forerunner 935 a few weeks ago https://www.strava.com/activities/2658735030.

    The Fenix 6 is a complete outlier. It missed the startpoint by a 1/4 mile (I did let it soak in for 5 sec or so before the run, maybe too short) and you can see the track to-and-from diverges although there is only one sidewalk and the run was done on it. The FR935 dealt with it fine. The Fenix 6 is beyond horrible.

    In fact, since it is so much off, would strava even recognize all segments?

  • 0 over 4 years ago in reply to Loen82

    3m would be phenomenal. This is in parts off by 5x.

    I expect it to be close to my old FR935. I dream it to be like my old Suunto Ambit 2R.

  • 0 over 4 years ago in reply to DCeremuga

    Hey, let me give you some encouragement.

    I am almost 50, 6'0' and wore the same size pants since high school. Last winter I was forced to size up, and the scale tipped 193. Not that far from where you are. I said enough is enough, started watching what I eat for the first time in my life, cut out drinking, and in April I started running regularly. Although I had done marathons years ago, I was not able to complete 2 miles at a 10:20 min/mile pace on a treadmill. 

    By July I was down 30lbs and some weeks was running 50 miles. I have run 950+ miles since April and I completed the marathon at a 8:10 pace. I am also a pant size down since high school.

    It was hard. It was very very hard. But if you had told me in January that I'd achieve 1/2 what I did, I would have not believed you. It is not easy, but it can be done.

  • 0 over 4 years ago in reply to PatrickNSF

    I wonder if it is because the metal case is attenuating signal more than the plastic case.

  • 0 over 4 years ago

    So funny, I opened exactly the same topic in adjacent F5+ forum few days ago. Nothing changed but Garmin GPS :)

  • 0 over 4 years ago

    Here's my 2 cents worth.

    I have noticed that when cycling the tracks from my Fenix 6S are PERFECT. I mean, I could literally tell from the map precisely what side of the road I was on and how I navigated the roundabouts.

    Switch to running, and the tracks are very hit and miss. Then again, so was my previous 645m.

    Something tells me that the Fenix 6S GPS chipset has great potential, but picks up a much stronger satellite fix when the watch is stable and always facing the sky, as it does when you're on a bike. When you are running your arm is swinging and the watch faces sidewards, making it harder for the watch to maintain a stable satellite fix. My theory is that with cheaper plastic watches there is no blockage effect of the casing degrading the signal between the antenna and the satellites. And a weak antenna will show itself most when running next to buildings or trees. But I am not an electronics engineer so this is only my personal theory.

    It's worth noting that despite these new watches costing up to £700, GPS technology has not advanced at all in about 5 years or more, and you are not going to get better tracks on a Fenix 6 than a 5 year old device. So Garmin tries to convince you that their latest high end watch is worth it by marketing the added features - mainly software improvements and more luxurious build quality (albeit at the expense of performance) - in order to justify the higher asking price. The fundamental functionality required of a GPS watch never improves.

    I have also noted that some days are much better than others. I think the track accuracy has a lot to do with how many satellites are overhead at any given time and place, and whether you have enabled Glonass or Galileo will determine how good your tracks will be. It would be good if the watch could automatically change modes depending on how many satellites it picks up.

  • 0 over 4 years ago in reply to scotthunter2

    No, accuracy is mostly related to speed due to the inaccracy of the measurement. Example: when running 10km/h, you run 2.8 m/s which is less than the GPS accuracy (~3m).

    GPS accuracy for cycling >> running >> hiking