Related
Options

Fenix 7 step counter compared to 920XT

glenas
4 months ago

I posted about the Fenix step counter being markedly different than the 920XT awhile back, in that it counted way less steps when my arms were stationary, ie. while mowing lawn etc. 

Someone suggested taking watch off and fastening to belt loop, which I tried but that did not make a lot of difference. 

Anyway, to do a comparison I put the 920XT on just as I was beginning to cut my lawn one day last summer to get a real life comparison, here are the results. Note that the bottom right image is later in the evening as I forgot to take a picture right after I was done. 

I think that the 920XT must use accelerometer when it counts steps, which I actually prefer because lots of times I walk if I am carrying something my arm motion will not coincide with my steps and I think the Fenix under counts. 

Top Replies

All Replies

  • 0 4 months ago
    glenas said:
    I think that the 920XT must use accelerometer when it counts steps,

    All Garmin watches use the accelerometer to count the steps, the F7 including.

    Unfortunately we cannot make any conclusion from your photos - it is also well possible that the steps from F7 are correct, and the 920XT counts excessively high because of the mower vibrations not being filtered.

    In order to see it better, I recommend recording the mowing as a Run activity on both watches, and posting then the links to the activities here. The run activity file contains also the cadence data, from which we can see how the steps are counted. I recommend recording two sessions - one with both devices on the same wrist, and another one with both devices in the pocket.

  • 0 4 months ago

    ... looking at your photos again, it is in fact clear that your 920XT indeed detects false steps (most likely the vibrations of the mower). Within the first 27 minutes (between 1:43 and 2:10) it detected 4573 steps. That corresponds to the average cadence of 170 steps per minute, which is a cadence of a runner at a decent pace. At a rapid walking, the cadence is rather between 80-100, and at casual walking even lower, and at mowing it can be even much less. So unless you really mow while running behind the mower at full gas, I do not think it is possible to do 4573 steps in 27 minutes.

    Well, the 400 steps of Fenix 7 are likely incorrect too, but that's normal - if you hold the mower, the watch cannot detect your motion, it detects the motion (and in some cases also the vibrations) of the mower. Keep the watch in the pocket, or stop holding the mower with the hand wearing the watch.

  • 0 4 months ago in reply to trux

    Thanks for great feedback, I will do the comparison again as you suggest and record it as a run activity, unfortunately it will have to wait until May or possibly April for the first lawn cutting where I live. I suppose I could do a session walking behind the snowblower but that is a very slow activity so I don't think it would be accurate. 

    I do however, walk my dogs nightly on bush trails, so tonight if I remember I will put on my 920XT and record as a walk activity on both watches and see how the comparison looks. 

    I actually did record a lawn cutting as an activity a couple of times to see how far I actually traveled, but this was with my original 405 watch which did not record steps, but I'll attach an image of the rest of the data that was recorded so you get an idea of distance covered etc 

  • 0 4 months ago
    glenas said:
    Someone suggested taking watch off and fastening to belt loop

    It needs to be on a moving body part that would be similar to arm movement when walking. During mowing, I typically place the watch in a pants/shorts pocket, so as to detect leg movement.