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

Forerunner 955 stopped recording climbed floors

Today my FR955 stopped counting climbed floors, I thought it was a one off error, so 1 hour ago, I took another 10 floors, to no avail.

It appears thar it got my descent, with a feeble 80% accuracy.

I actually wanted to write up a small review of the FR955, because I have been pretty happy with it so far, with one of the few issues being accuracy in floors climbed in general.

When it works, and it did, for the previous 6 weeks, it is usually at 80-90% accuracy, which may or may not be good enough (no, it isn't good enough).

On most days, I do my 10 floors in a closed indoor fire staircase, with sealed doors and no ventilation. I would argue, that this ought to be ideal conditions for measuring ascent and descent. And yes, I read the guides about accuracy on measuring floors, and I rarely touch the rails. I do sometimes walk two steps at a time, just to get more motion in my legs, much to my surprise, this doesn't appear to affect the count, contrary to Garmins advice:

https://support.garmin.com/en-US/?faq=LJ6vYF2phv7JUAOlsRRjC8&identifier=7EL016492&tab=topics&topicTag=region_dailywellness

Neither does the movement of my hand affect it much, sometimes I hold the dogs leash in the hand where I wear the watch, sometimes not. There does not appear to be any significant relation to this either.

Rather, it seems random whether the accuracy is close to 100% or down towards the 60%, and mostly it falls short when climbing. Typical accuracy is 80~90%.

Well, it's a lot of free exercise, when you really want to hit the 10 floors, but today I did 20 floors, with a 0 count. Not gonna do any more... Just have to live with that streak being blown.

But the watch is not only inaccurate on indoor stairs, it also has massive problems in the open.

On some days, when walking in the open (and near / in between some 10 storage buildings), it is often way off due to the wind.

Some days it easily gives me 10 floors for "free", due the wind and the subsequent fluctuations in air pressure. While I, to some degree, could understand this between the buildings, I can't see the logic in this issue in the open.

Such a high tech device, as the FR955, which incorporates super precise satellite positioning, should be able to correct this when I walk in the open. And really, where I walk, there is very little elevation, the real elevation differences of a 2K walk is typically less than 2 meters. And a 20K bike ride in the same area gives me 24m ascent/descent.

Also, you could do some backend post processing, of such walks, just to make a sanity check of the elevation.

I don't know if todays 0 count is due to the recent SW upgrade I accepted one or two days ago.

Regardless, with the capabilities (and price tag) of the FR955, this ought to be much much better, and except from holding the rail, I don't see much reason why the device shouldn't be a whole lot more accurate, especially in a a closed stair case with no notable fluctuations in air pressure.

Had it been in the mall, with it's large open doors and heavy ventilation, I could understand some inaccuracy, surprisingly, it doesn't seem to affect the FR955 much, I'd estimate the largest inaccuracy I saw in the mall to be 15%.

In all fairness, this feature is far from the most important in the FR955, however, when such a basic thing is so inaccurate, it does make me doubt some of the other more sophisticated things, which was the real reason I upgraded my Vivoactive S2 to FR955 after a few months.

And well, my mood got no better, after my new Index S2 Index scale threw away the reading this morning and for the second time in 5 weeks. Frankly, spending that amount of money on a scale, just to get data transferred automatically to my phone, and then it throws away approx. 2 out of 80 readings, that's just poor.

It's the _one_ thing I paid for. yeah, it gives other metrics, but I have to admit, that I find those to be a bit of hogwash, I seriously doubt that my bone mass fluctuates with almost 300 grams in a day. Well, I never expected those advanced metrics to be very accurate, I had a Tabita, the one with the SD card for years, and yes, it's fun, but that's it. The most valuable thing in those devices is the day to day stats and lovely graphs you can make.

Those automated data transfers was ONLY reason I wanted the Garmin Index S2, and then it isn't even 100% reliable.

I even put the Index S2 tp the test, because I know I sometimes have some DNS issue (yes, I should change the router config), but the funny thing is, if I break my network somehow, and jump on the scale, it stores it, and as soon as I reconnect my network or turn on the wifi, it will sync, next time it is activated.

Since it can store it and sync later, how can it miss any readings at all?

Garmin, please, creating a 100% reliable protocol and transferring mechanism for such simple and infrequent data, that isn't hard....

And why, why, does it take minutes before I see the numbers in Garmin Connect? The scale obviously finishes with the sync seconds after it stopped calculating the data. Yes, it's silly, but jumping on the scale is the last thing I do before I go to bed, and for some weird reason, I like to see that graph before I go to bed. The same in the morning, I jump on it, before I start making the morning brew, and would like to just catch a glimpse of it on the phone, but I'm not gonna wait minutes, I need to make lunch box for the kids and brew coffee. And no, I'm not going to stand and look at the scale either, and I would waste a second reading the weather report on the scale either, I just want it to sync my data, fast and reliably.

It reminds me of the old days, when the first iterations of Garmin's unreliable chest straps, broken fit files in the 910XT, sync and connectivity issues of all those clunky old ANT devices. Yeah, I have quite the pile of ancient ANT dongles, extenders, long USB cables, and broken chest straps, reminding me about how poor this all used to be. The old TacX direct drive trainer is long gone.

I didn't really want Garmin this time, but DC Rainmaker convinced me to try again, and granted, there is no comparison to how poor things were 7-10 years ago. Back then I understood it, you truly pushed those tiny devices to the edge and it was all brand new.

In 2022, however, there is no excuse for poor connectivity, losing readings, or a feature like floor climb counting stopping to work.

Anyway, Garmin, I had planned to acquire the Tacx NEO Bike Plus in spring, but, if these issues aren't fixed before, then I think I'll go with the Wahoo instead.

  • Today it was 100% accurate. Despite wearing the watch on the right arm, where I held the dogs leash, and taking double steps on the three last floors.

    Oh, and after I posted yesterday, the watch counted 1 climbed floor, when I climbed none.

    Well surely, it isn't random, but sometimes it feels like.

  • Today:

    0% accuracy down

    90% accuracy up

    Still in the sealed fire staircase

  • I actually have the exact opposite problem since the lateste update to 13.22: my watch counts floors even when I'm on an elevator or escalato. Even on days where I do not actively walk upstairs at all I get a nice floor count of most floors I ride upstairs in an elevator or on an escalator.

    It's a shame to see what Garmin does with this watch, especially as it is basically a Fenix 7 in a different case. I don't understand why they didn't simply use the Fenix 7 software in the first place - the developer team seems to be much more sophisticated, too. Also, even my 3 year old (and much cheaper) Samsung watch manages to get a decently accurate floor count.

    I already opened a ticket with Garmin support (and hope that they will fix it soon without messing up something else in the next SW update), but I am wondering if anybody else has got a similar issue?? I'm pretty sure we can't be the only 2??

  • I noticed some other FRx55 users also have issues. So we are not alone, but I guess this isn't the most important feature for most FR955 users.

    But it is really off.

    Yesterday it managed a 140% inaccuracy on climbing!

    I walked up the usual 10 floors, got credit for 0.

    Then I walked 4K or so in open flat terrain, and got credit for 4 up.

    Today I got credit for 13 up, but only climbed 10. And I descended 10, but only got credit for 9.

    Garmin should disable this app, until someone got time to fix it.

    As I recall it, my VivoActive was somewhat more accurate, sure, far from 100%, but in the 80% range, and never off by 100% like FR955.

    So surely, if they wanted to, they could fix this.

  • Have you found any improvement, since this original posting?  I'm having a terribly rough time with my elevation gain on trail runs (3-11 miles) always showing hundreds of feet more than on the Garmin Connect elevation correction.  The Garmin Connect elevation correction is way more accurate than the altimeter.  I've tried so many different settings and calibrations with no luck.  Sounds like floors climbed would be related to this issue.