Counting steps in various situations

[5ggggg]

Counting steps in various situations that are wrong from the customer perspective:

I experienced the following situations myself:

- while cycling

- driving (it is not only on bumpy roads 😉

- lawn mowing

Others experienced similar situations:

cycling less than 20 minutes not recognized or detected as sport

Versa steps inaccurate

Versa counting too many steps

Mowing the lawn or riding bike?

Step counting - cycling/paddling/swimming/driving, where is the option to turn it off.

Still counting steps while cycling

Step counting whilst cycling

Counts steps while driving

Versa counting so many steps while driving

Steps counting while not wearing the Versa

Steps overnight

Step Sensitivity

....

I just took some samples of the last 2 month 😞

I have no idea why Fitbit is not addressing this issue which it reads in this forum since many month. Some of those problems also happen on the Charge 2 I wear, but way less often and a lot like charging, fans, humidifiers or fridges are Versa specific.
Interesting though that some of those posts are marked solved, even when customers say that they will return the watch or when the cause was identified as fridge, humidifier,... of course vibrations should not be dedected as steps by a 180$ device 😉

For me being a technician I think even if it cant be solved completely, at least an improvement could be reached by changing the firmware algorithm and filter settings of the G-sensor.

And it would also be a simple programming task to just not count steps when an exercise like cycling is dedected.

Bottom line:
1 Those like me where it is counting steps caused by bumps (vibrations) = G-force (not related to steps) might wait for a firmware update

2 Those of us having it count near electrical devices may be able to avoid this or return it as the electrical device (electric field) and/or vibrations influence the sensors.

3 Those having it count while charging should return it. There the charging circuit seems to influence the sensors.

4 Those having wrong auto recognized exercises like biking when mawing the lawn might also hope for an new firmware addressing point 1.

5 Those like me having steps count during execises like biking could also hope for an easy firmware fix. But sadly this could have been done since the Charge2 came out and never was fixed so far.

[Rich_Laue]

Part of this is because the Versa does not count steps by looking at leg motions, but looks at arm motions. 

If you look at the common point with your links, they all include a vibration or up/down movement of the wrist .

Yes with any non mechanical tracker, from any manufacturer the user will occsisionly see false steps during these periods of excessive movements of the arm.

One way to minimize this is to try different Fitbit settings for the arm. 

Most of your links already explain this. 

[5ggggg]

This is why I addressed 3 different situations:

1 Vibrations that engineers can correct rel simple. Technically spoken Versa works like Charge2, but its software algorithms and filter settings for the sensors must be improved to reach the level of Charge2.

2 Can be split up into the prob with
a) vibrations (fan,...) which could be addressed when solving 1 and

b) a prob with electrical fields (which has never been proofen so far) that might corresponds with point 3 and is more a hardware design based issue that not fully can be compensated by software and might need a new hardware revision which is a lot more expensive for Fitbit.

3 Is a quality or hardware design based issue (2b).

@Rich_Laue: THX for your reply. Please explain what you mean with "different Fitbit settings for the arm".
If you mean its position or the tightness of the strap .... I dont think so, because both have to be optimised for the heartrate sensor (and not the G-sensor). 

[Rich_Laue]

I simply said to check and play with your arm/wrist settings for the tracker .

Is the traxoer on your dominant wrist? 

• Does the wrist I wear my device on affect accuracy?

[5ggggg]

I wear it on my left and so have set it to "non dominant" .. I will try dominant now. So THX for this info.

• The dominant wrist setting decreases the sensitivity of step counting and should reduce any over counting of steps when your body is not moving.
• The non-dominant wrist setting increases the sensitivity of step counting and should reduce any under counting of steps. Non-dominant is the default.

@Rich_Laue: On my Charge2 there also is a setting on which hand you wear it. This way it knows on which hand turn it has to switch on the display. Do You know how Versa solves this (hopefully not with the dominant setting).

[5ggggg]

Just to say: The dominant setting did not help to solve my issues 😞 and those many threads complaining about wrong step counts in various situations speak for itself.

For me I will wait until one of the next firmware updates and hope that Fitbit improves the treatment of the sensors which obviously is worse compared to the Charge2.