Related
Options

Problem in walking activities

davidgtc
2 months ago

I have a problem with my epix 2 pro. It measures the distance traveled incorrectly, but this only happens during walking and hiking activities.
I compare it with my wife's fenix 8 and it always shows me about 100-200 meters less for each kilometer. On some 5 or 6 kilometer routes I lose 1 km.

I have tried to compare both routes, and you can see how the speed of my epix changes suddenly and makes strange waves, as you can see in the capture.

Could it be a problem with the accelerometer? Since the distance measurement works perfectly during running activities, because I suppose the GPS is used more than during walking activities.

I have already tried to reset the watch to factory settings several times, use the data recording per second, use the multiband GPS, etc.

  • 0 2 months ago

    It doesn't work for me either, I already wrote in the beta forum.

    https://forums.garmin.com/beta-program/fenix-7-series/public-beta-reports/i/public-beta-20-xx/distance-measurement-algorithm-failure-epix-2-pro-beta-20-xx

    "On February 28th, I already reported that since version 20.xx, the distance measurement when walking activity became inaccurate. I copied several fit files from the clock and I converted it into an Excel spreadsheet for better readability. The distance measurement randomly stops, it seems that the coordinates change, but the distance traveled remains the same, i.e. it does not increase. This happens many times during activity, it doesn't count the distance traveled for 3-4-5-10-20 or even 30 seconds. It's probably related to speed, because it doesn't happen when I run, only when I walk. It happens even if the data recording is intelligent and even if it is per second. I found the error in many fit files. I wanted to attach several files where the error is clearly visible (marked in red), but the interface doesn't allow it, so I'm attaching a OneDrive link from which you can download it.

    https://1drv.ms/f/c/ab132c39247715c6/EmHH84BxmQxKu8IF7JEgpA4B1NxDY1Ze8MOI9EaLU3v5zw?e=aCyeC9"

    the coordinates change, I move forward, I have speed, but the distance doesn't change

  • 0 2 months ago in reply to aribacsi
    aribacsi said:
    It's probably related to speed

    More precisely to the cadence. The watch uses the combination of GPS and accelerometer data (especially the cadence) for estimating your speed and distance. When the GPS signal is of lower quality, or dropping completely, the pace and distance are derived from the accelerometer data. So far so good, and perfectly comprehensible.

    Unfortunately, the implementation is far from perfect, and not only is the estimation highly depending on the cadence, often overestimated at high cadence, and underestimated at low cadence, but at cadence 60 spm or lower it often stops counting  completely (unless the GPS signal is perfect).

    It is most noticeable at a slow indoor walk (where no GPS is used at all). I am able to replicate an hour long indoor walk where the watch measures barely 20m, although it detects over 3000 steps. I have sent the data to Garmin several times. Unfortunately the report was not taken seriously, and it was not even assigned a tracking number.

  • 0 2 months ago in reply to trux

     Will this bug be taken seriously? Can you do something about it?

  • 0 2 months ago

    I think it has to do with firmware 20.16. There are more affected people in this thread:

    forums.garmin.com/.../fenix-7x-erratic-pace-on-20-16-firmware

    Yesterday I downgraded to version 19.41 which seems to be the one they say works well, but I haven't tried any activities yet.

    Today, in my garmin ticket, they told me to update to version 20.22 which they released yesterday, but I want to try 19.41 first to confirm that the problem is with the firmware and not a broken watch.

  • 0 2 months ago in reply to davidgtc
    davidgtc said:
    Today, in my garmin ticket, they told me to update to version 20.22

    I doubt it will help at all. They changed the algorithms since version 20.XX and they are counting on the latest version to magically fix a problem they haven't even looked into.

    Version 20.XX is a total disaster...

  • 0 2 months ago

    why is this thread marked as solved? It doesn't look like it's solved 

  • 0 2 months ago in reply to gaterobs

    I have backtracked poor accuracy in walking activity to exactly the update to 20.16. I walk the same route every weekday morning of 3.4km, and 19.41 was exact every time. In 20.16 the distance fell short of roughly 200m, despite the route plot being identical to the plots from 19.41. This indicates that the walking activity may rather be mapped to motion than GPS. This exact issue was what caused me to update from Fenix 5S to Fenix 7X, where they messed up height measurements in walking activity in the same way (motion based and not GPS). Now again someone made a FW release without proper regression testing as you do for every SW release, known to every SW engineer outside Garmin apparently. in 20.22 it seems they did attend to the issue, but from the looks of the first walk I made with this release it is still short 80m.

  • 0 2 months ago in reply to Aasimonsolar
    Aasimonsolar said:
    Now again someone made a FW release without proper regression testing

    Garmin doesn't do things like that. What's more, they don't pay attention to bugs reported by users. The company's income charts would go down if they bothered about such trivialities (sarcasm): the5krunner.com/.../

  • 0 2 months ago

    I accidentally clicked on Verify Answer, my fault. I've now clicked Reject Answer, but I think the comment has been deleted. Sorry, it seems more difficult than this forum seems.

    If it helps, I did an activity with 19.41 and it seems the pace and speed are back to normal. I'm not going to update until I'm assured they've fixed the problem. I've already done three official hiking activities with the issue, and it really bothers me to see the graphics, lap times, and incorrect distances. Perhaps Garmin could have told me about the firmware problem from the start and not have spent almost two weeks offering me solutions like restarting the watch, deleting certain files, etc.

  • 0 2 months ago in reply to davidgtc
    davidgtc said:
    Perhaps Garmin could have told me about the firmware problem from the start and not have spent almost two weeks offering me solutions like restarting the watch, deleting certain files, etc.

    Support doesn't care about solving the problem, only closing the thread, so they send some ready-made "problem solutions", like restarting the watch.

    And why is that? We can only guess...