Connected GPS Pace and Distance is Inaccurate

[DanielMena9]

Went on a run today with my Charge 2 on one wrist and Charge HR on my other wrist. Also had RunKeeper to measure my distance. I started my run with about the same number of steps on both FIT devices. 

After the run, my Charge HR had almost 2K more steps than my Charge 2. In addition, my GPS map on FitBit matched my Runkeeper map which is about a 3.5 mile run BUT FitBit pegged the same route as 1.73 miles >.> 

Why doesn't this device work as advertised? I really like the design and feel of Charge 2 but if it can't do its job accurately, all is for not and I will be shorting FIT as it appears to be a widespread issue (I have maps and photos if someone from FIT wants the data).

Moderator Edit: Edited thread title.

[CTT2615]

I've done some more tests and regrettably have to agree that the problem still exists. Yesterday I did a 5km run, over an independently measured distance and it read out as 5.5km on connected-GPS. Just the same as last time.

I have sent the TCX files to @Kobalto by PM - he may be able to confirm that the error that I continue to see is the same as everyone else. (By the way: I appreciate all the analysis you're putting into this, Kobalto!)

What a pity - early signs were promising!

[CTT2615]

... and I agree @Rollerskier - it would be good to see a moderator involved so that we could be confident that Fitbit acknowledges the problem!

[hawkeye1952]

Here is something I just learned. It takes 3 satellites to give you accurate position. There is another side to this. It call noise ratio which has to do with the signal strength. In my house right now I have 3 to 4 satellites which my program accesses. They have a very low noise level or very poor to poor signal strength. When I put it up to the window I get 5 or more satellites that read from poor to very good depending on the satellite number. I think the main problem with fitbit is that it can not lock on to more than 1 or 2 satellites here a picture of my satellitesthis the strengtht of my gps singnals at my window

[Rollerskier]

I did some testing of accuracy of GPS for Apple technical support before.  If you like to check your phone GPS accuracy position yourself on a spot you can easy recognize on the map.  Like a street corner or road crossing. Open a map that plots your position and compare this position with were you are physical standing. If the phone signals are weak the position will move around. 

[hawkeye1952]

I took my morning walk today. Using both the fitbit app and my regular program. Both where started at the same point. Then I check my satellites strength. When I start I had 7 that were very good and several other listed as good. I also check them along the route. Total distance was 1.7 miles. fitbit app 1.48 miles. Again it failed to give the right distance traveled. 

[Colinm39]

@hawkeye1952 Thanks for keeping us up to date. What App are you using because I'm testing my Ionic and it can take up to 2 minutes to connect. My satellite App only shows the strength which is always good.

[Rollerskier]

Would be interested to find out if you get less deviation if you test with 2 phones - Fitbit App on one and your regular Program on the other one.

I can try to start 2 different GPS Apps on one of my phones too to see if I get more deviations. And start Fitbit on the other phone too as a reference.  

[Rollerskier]

So I tested this just out.

iphone6+ with both Fitbit App and Nike Run Club and Iphone7 with Fitbit Charge 2

Results:  

iPhone 6+ ; Fitbit 3,43 km and NRC 3,25 km

iphone7 ; Fitbit Charge2 3,25 km

Conclusion: The deviation increased for Fitbit when the same phone run another GPS tracking app simultaneously.

I think that’s part of your problem with deviation. Fitbit gets less accurate when it has to compete with another app for the GPS signals on the phone. 

[hawkeye1952]

My main fitness program is endomondo. There are 2 features that are wonderful. 1st one tell you when you lose your gps signal and when it is restored. 2nd one is when hit the start button it tells you to go. Which means it as locked on to the gps signal. Takes anywhere from 1 second to 5 seconds..

The one that gives me the gps satellite signal strength is called gps waypoints. It also has a mapping feature. The time I have compared the 2 they been right on.

[Colinm39]

@hawkeye1952. The App your suggested was a great help today.. I had 7-8 satellites and the Ionic connected immediately as did MapMyRun and the tracking was within acceptable limits and no wobbly paths on each.  But I used Run instead of Walk for my Walk.. as recommended by some moderators in similar posts about GPS...

More to be done at my end....

---

@hawkeye1952 wrote:

My main fitness program is endomondo. There are 2 features that are wonderful. 1st one tell you when you lose your gps signal and when it is restored. 2nd one is when hit the start button it tells you to go. Which means it as locked on to the gps signal. Takes anywhere from 1 second to 5 seconds..

The one that gives me the gps satellite signal strength is called gps waypoints. It also has a mapping feature. The time I have compared the 2 they been right on.

---

[SunsetRunner]

Hello Colinm39,

If you look on the thread for the Ionic  https://community.fitbit.com/t5/Ionic/Problem-with-Ionic-GPS/td-p/2229217/jump-to/first-unread-messa...

you will see there is already a number of posts about the problems with GPS tracking. ( Just like the problem with the Charge 2 Connected GPS )

You are just wasting your time, Fitbit are not listening.

[Kobalto]

Hi everyone,

Here is a PowerPoint a few TCX analysis (thanks to the files submited by @Colinm39 in Melbourne and @CTT2615 in Camberra):

https://1drv.ms/p/s!AkHt0GIzDqM5hoEf_I4p_e35OFHrQw

Each graph in the ppt contains 3 lines:

- blue is the distance calculated with the GPS coordinates (this is the "correct" calculation)

- gray is the distance a calculated by Fitbit.

- orange is the distance calculated with an induced error by intentionally omiting the latitude information in distance calculation.

i.e., if gray line is close to blue line, fitbit is doing the right calculation. If gray line is close the orange one, fitbit is wrongly translating coordinates into distance. Main conclusions are:

1. Melbourne (TCXs from @Colinm39😞 using the Fitbit Android App, and Surge, the distance is correct. But using the Blaze, distance is wrong.

2. Camberra (TCXs from @CTT2615😞 There are two TCX, both using Fitbit Charge 2, but one in "Sport:Running" mode and the other in "Sport:Other" mode. Strangely, the Running mode is wrong, but the Other mode, seems to be accurate. This is really puzzling!! 

3. Santiago - Chile (my TCX): Distance is wrong.

Please notice that in the examples, the gray line is either close to the blue line or close to the orange one. This means that the error is not random (random error would be cause by poor GPS reception, bad connection with the cellphone, etc). The error is caused by fitbit incorrectly translating GPS coordinates into distance.

Here is a link to the Excel I use, feel free to use it, just be careful with adding the right number in the yellow cell, which is different for each latitude. 

https://1drv.ms/x/s!AkHt0GIzDqM5hoEip_KnY_59SXODLw

[Kobalto]

---

@Rollerskier wrote:

The internal life of Fitbit is a black box.

I only run one App using GPS at the time on the same phone.  I use always 2 phones to check and compare the results simultaneously.

"Kobalto" - Are you using 1 or 2 phones when you compare the results done simultaneously?

You mentioned you have changed to iphone7 lately but do you still have your iphone6s around to check this out with 2 phones simultaneously?

Previously I had 2 Fitbit trackers set up at the same phone. This was also a mess if both where active connected at the same time( the other tracker in the same room ). I was also trying to compare the HR with 1 tracker on each wrist connected to the same phone...did not work out too.

---

Hi @Rollerskier

I use just one one cellphone: I turn on Connected GPS with the Fitbit, and at the same time I start tracking with the Fitbit App. So the source of information is the same one (GPS of the Phone). Both tracks look alike, none of them with any points showing inaccuracies due to bad satellite reception, but still the distance of both tracks is different.

On the weekend I tried using just the Connected GPS, with other apps closed (with the exception of the Fitibit App that needs to be opened), to see if there was any interference, but it is still measuring incorrectly. This time I didn't have a second track recorded with the App to compare, but I made a track with a known distance to measure the error.   

[CTT2615]

Hi Kobalto,

Both of the activities that were recorded via Connected GPS to produce those two TCX files were recorded in 'running' mode. After recording I changed the shorter of the two to 'Walk' for my own record keeping purposes. I wonder if that resulted in the mode being recorded differently in the TCX file itself?

The only other difference between the two is the overall length and the fact that one was walking and the other running. And that is definitely puzzling!

[Rollerskier]

I made another test running 2 apps this time on iPhone 7 and 1 app on iphone6+.

iphone 7. -  Fitbit 3,27 km and NRC 3,06 km

But both apps completely lost the path along the road  in the valley and made a shortcut trough the forrest and some gardens. This for a distance of 2-300 m.  GPS signals transmitter apparently far off to the apps. 

iphone 6+ Fitbit 3,27 km.  

Whats more normal. With 2 apps yesterday Fitbit calculated 3,43 km on the same track.

Its looks to me that the phone with 2 GPS Apps ends up with more deviations than the one with only the Fitbit tracking. Lost path and wobbling pattern are symptoms with 2 apps. 

[Rollerskier]

Well, you had a good start. Will be interesting to see the stability over time.  Keep us posted.  Given your location in Canada/Ontario you are at least not affected by the Southern hemisphere issue. What kind of phone/model are you using? 

[hawkeye1952]

The first btest I did with my new phone and the distance was way off.

George Howell Jr

[SunsetRunner]

Hi agetori77,

Hope you got a good deal on your Charge 2 because if you want it to record accurate data for training purposes, I wouldn't bank on it.

I read your post and you seem to have got good results, but have you checked them against another source for accuracy?

How did you count your steps without error ?

Why did you start and stop the app for each new road?

I have found that the farther I walk the bigger the error in the data.

If you read the posts about these problems on other threads you will see what people are talking 

about.

Look here https://community.fitbit.com/t5/Windows-10-App/GPS-Tracking-unreliable/m-p/2295864#M12686

If you just bought the fitness tracker for a guide to your activity then you will not be disappointed with the results, but do not rely on it for accuracy.

And don't expect any help from Fitbit, they've got your money. 

Best of luck,

Clivep.

[SunsetRunner]

First, I wanted straight line tests to avoid any 90° turns that would have the fitbit fill in the gaps (should gps be lost) with steps instead of gps distance from the phone. I noticed people that had that happen had incorrect distance and mapping that cut corners and joined the last gps position with an incorrect slashing line to the new gps position. 

Second, I can count to a hundred 12 times. It's cold out. Hands are in my pockets. Count 1-100 then stick 1 finger out straight.  Count 1-100 again... I hope you can fill in the rest. By passing grade school with some decent counting skills, anyone should also be able to do this. I wasn't counting to 1200 and hoping I didn't skip any numbers or lose count.

(Edit: 1200 was the "long" side of the route and the highest step count.)

As for the rest of your message, you can ease that negativity  of yours by a few levels. It's not adding to the discussion at all and frankly you're lucky I'm even responding to you. 

[SunsetRunner]

Well I am "honoured by your response" ( twice.)

I did not mean to infer that you could not count, I just thought you might have used another device or app to count the steps. Using your method you must have 12 fingers 100 x 12 = 1200 ?

After using my Charge 2 for the last 12 months and talking with the Fitbit support team for advice I have never been told to start and stop the Connected GPS tracking at any 90 degree corner to increase accuracy.

And "as for the rest of your message" where do get off telling me what to do or say ?

It sounds to me that you're the one with the attitude.

Don't bother responding, I have unchecked the "Email me response box"