GPS distance totally wrong on Charge 3

[BertilDator]

Was doing cross country skiing, and set activity to "run" (as there is very limited activity list). Had phone connected, and got perfect GPS route on Fitbit activity. However, all the splits are way off. Total distance in Fitbit shows 4.8km while real distance was 7.5 km (I know the route, and even double checked via ontegomap.com).

How can Fitbit calculate distance so wrongly even though the GPS route is perfect on the activity? I repeat, the tracked route is 100% correct on the Fitbit activity in question, so there is no disconnections, or trees, or anything between the phone and GPS satellite, nor between phone and Charge 3.

Is Fitbit using GPS route for distance calculation when assisted GPS is enabled during "Run" activity, or does it STILL use stride length?

Should I use "Biking" activity for XC skiing to get accurate distance based on the GPS route?

[t.parker]

@socal13 you have just answered your own question. Fitbit won't fix it because according to Fitbit it is not broken. From the user's perspective the problem is quite serious (at least for those who care about the data correctness and use it for training). If the live distance for bike or running isn't correct then all other metrics are not reliable and useless for serious training.

[socal13]

For anyone interested in actual fitness data, the Charge 3, as currently being sold, is pretty worthless.  I've had several discussions, and emails, with fitbit support and have no confidence that they even care about their problem.  Too bad, with a simple fix to it's GPS distance calculation, the charge 3 would be an ok device.

[t.parker]

@socal13 this is something that didn't work well for Charge 2 and Ionic which both I owned so this isn't really the new issue and it isn't specific for Charge 3 only. As much as I can understand why walking/running distance is wrong (using stride and steps) I don't quite get how is it possible the cycling distance is also wrong. There are two things I can think of:

- Bike exercise doesn't block step detection - maybe that detected steps automatically contribute to the distance calculation.

- To make the battery last longer, the BT communication between the watch and the phone may be limited and the GPS updates come with greater intervals (for example every 10sec). That would skip lots of waypoints and could affect the distance computation. This theory, however, should not apply to devices like Ionic, with built-in GPS but even Ionic gets the cycling distance wrong on multiple occasions. Looking now at one of my old rides, the 30.8km (distance from the captured map) has been saved as 29.8km - it's 1km of difference. This is a lot.

Whatever it is, I wouldn't have any hopes that it's going to be fixed in the near future.

[Springy.BitFit]

For Cycling, maybe they use the number of cycle you take per minute and have a figure that one rotation = x amount of distance. The fitbit should be able to detect that... But again that is rubbish as it depends on the gear you are in on your bike.

Also is the distance generally less? If so that may point to when one does not pedal and you free wheeling. But maybe fitbit has a bit of code that says if person x moves without pedalling then use the GPS distance to calculate...

But who knows... FitBit use some complicated stuff to work out distance when they already have GPS.. They making it so much harder for themselves. They need a code re-write 🙂

[t.parker]

@Springy.BitFit it isn't possible to detect revolutions by wearing the wristwatch as it isn't so simple. The majority of sports watches require an external cadence and speed sensors attached to the bike. Something like this: 

https://www.amazon.co.uk/Wahoo-Cadence-Sensor-Android-Computers/dp/B01E0YRQIA

You attach the cadence sensor near pedals (or even to your shoe) and the speed sensors on the wheel, then you pair the watch with those sensors to obtain data. A proper sports watch should be able to combine data coming from sensors and GPS to improve overall accuracy.

[AdamH0]

@Procter101 how are you recording the bike rides? 

When I was using my Charge3 I would start recording from the watch, by swiping over to Exercise and then selecting Bike. It'd wait for it to connect to phone, and then start the activity.

I find it strange that you are seeing steps recorded for a bike activity, this was never the case for me. I've logged into my fitbit dashboard on the website, and found some past bike rides. They don't list steps nor floors, you see elevation gain instead. Down at the bottom of the webpage is an 'impact' section, this says +0 of 9000 Steps - which I interpret as this activity added 0 steps of my 9000 steps for the day.

For Bike rides recorded with Connected GPS, then GPS is the only possible option for distance measurement.  

[t.parker]

@AdamH0 this is interesting what you are saying about the Impact. From the past bike rides I remember, there was always something in the Impact field, however, now it's also shown as 0. So here's a small breakdown of one of such rides (I combined all as one image):

This ride was tracked with Ionic. One thing - the distance is wrong. The same TCX uploaded to the MyGPSFiles gives a distance of 22.96km while Fitbit shows less 22.26km. Now, the Impact field indeed shows 0 however if you look at the overall steps graph, there are plenty of steps during that ride registered and they do contribute to the daily steps count. Now, there may be two valid explanations.

1) Bug on the dashboard and always showing 0

2) Fitbit stopped counting steps during a bike ride (very quiet update?) but past bike rides would not change data (something to be verified)

Until verified, I vote for option number one - a bug in the dashboard.

Edit: just noticed that even time doesn't match and looking at timestamps, MyGPSFiles is right.

[t.parker]

@AdamH0 This is the bug on the dashboard. I took recent ride (about 2 weeks ago) recorded with Charge 2 (Connected GPS, same as Charge 3). Results below:

This test in fact is way better as I didn't wear the device at all except the time I rode the bike. My bike ride added 4,5k steps to the daily total. Showing "Impact" as 0 must be a bug then. The good thing is that this time the time matches.

[ukAdamR]

I got an answer from Fitbit support, but it's such a non-answer obviously crafted to buy time and send me on my way:

We'd like to apologize for the delay response. Let's try our best to find the promptest resolution to your Charge 3's distance accuracy issue.

After checking in our system, we were able to confirm that the last time your Charge 3 synced was today with 87% of battery.

Fitbit devices calculate distance based on the device’s accelerometer data and stride length setting. Default stride length is based on your height and gender. If available, Fitbit devices use GPS data to calculate more precise distances.When you sync your device, our algorithm interprets the data from the device, and you see your distance traveled in the Fitbit app.

Please take a moment to look at How accurate are Fitbit devices? and How do I measure and adjust stride length for my Fitbit device?

We hope the information provided is helpful for you.

Let us know if there's anything else we can do for you and we'll do our best to take care of it.

---

I have checked the stride length setting, I have it configured to use GPS to determine this. (Surely the most accurate option, and recommended in the "How do I measure and adjust stride length for my Fitbit device?" article I was linked to.) This doesn't seem relevant though, as I don't understand is why Fitbit needs to use its accelerometer at all when GPS is available throughout the whole exercise. (Google Maps has no problem with this.)

Looking at the walk I did today Fitbit claims I walked for 2.66 miles. Google Maps timeline however shows this same walk at 3.1 miles. Even compared to an identical walk yesterday Fitbit claimed this to be 3.03 miles, but with less than 100 steps in difference. -- These inconsistencies are just so distant making Fitbit useless.

I know this walk is actually 3.25 miles when drawing it out perfectly, but a 5% margin of error for GPS is acceptable. -- For example the Google map does show some slight (impossible) shortcuts being taken, but these are all small. Google is also very consistently measuring my identical daily walk at 3.1 miles with the same GPS device.

Fitbit is just so far in the wrong.

[ukAdamR]

Another answer from support on Friday:

---

Hi Adam,

Thank you for getting back to us along with the information and screenshot that you've provided.

All devices with GPS sensors require a direct path to GPS satellites to calculate location. A weak GPS signal might affect the accuracy of your route and other activity data. Your device uses your step count to calculate distance until it finds a signal, so the total distance calculated for a workout may be slightly less accurate when GPS isn't available for the entire time. 

Environmental factors, including tall buildings, dense forest, steep hills, and even thick cloud cover, can interfere with your phone’s ability to connect to GPS satellites.

If you still think that this is not the factor affecting your GPS, please try to restart your device by doing the steps in this link: https://help.fitbit.com/articles/en_US/Help_article/1186We hope that this helps. Should you have any other questions, feel free to let us know

---

I can only strongly disagree with Fitbit on this.

On Friday I rebooted my Charge 3 (the smile appeared on screen) and have gone on 3 walks since. -- Fitbit is still inconsistently recording my distance, and recording it way short of the truth.

Meanwhile Google Maps timeline shows me still consistently walking 3.1 miles. Not only is Google consistent, it's significantly closer to the truth.

I don't think I'm being listened to when I've repeatedly stated that I am using an identical GPS hardware and environment for both Fitbit and Google. Many of you in this thread with the same issue are experiencing an identical situation, with enough of you supplying evidence of it.

Both services are using my phone as the only means of GPS source, my OnePlus 5T. Both of these must use the Android location services API to read the position from the GPS sensor. The phone's location settings are configured to use high accuracy mode, and I'm ensuring the compass is calibrated to medium or high before every journey. -- Fitbit is simply can't be reading the GPS data available to it via Android location services API correctly.

Come on Fitbit. You need to address this issue. Even a message of progress on the matter is better than silence.

[tom43614]

That is a Bullsxxt answer from Fitbit. Go Garmin. I HAVE!!!!

[Procter101]

Just keep banging on and demand a replacement
If you get through on the phone they are easier to pin down. I am awaiting
my 3rd one but will demand money back if this one does not work.

[PRCstuart]

Wow, seems this problem hasn't gotten any better.  I ran the exact same trail yesterday as I did a couple of days ago; yesterday it was 12.14 kms, last time it was 9.97 kms.  This is one of the largest differences I've had but my Charge 3 rarely measures a trail the same distance twice.  I'd kind of like to know the actual distances...

[Ste247]

I am also having this issue when compering my running distance with my charge 3 and samsung health run on a s10 5g there is about 20% less on my tracker....this is affecting my overall cardio score too..

My girlfriends fitbit seems to be tracking normal 

This was all checked using active gps 

[PRCstuart]

I finally found a likely explanation for this: 

"...All devices with GPS sensors require a direct path to GPS satellites to calculate location. A weak GPS signal might affect the accuracy of your route and other activity data. Your device uses your step count to calculate distance until it finds a signal, so the total distance calculated for a workout may be slightly less accurate when GPS isn't available for the entire time.

Environmental factors, including tall buildings, dense forest, steep hills, and even thick cloud cover, can interfere with your phone’s ability to connect to GPS satellites...."

I run in a heavily wooded area so it makes sense...

[ukAdamR]

Latest support response is now blaming my phone being unsupported, despite it fully meeting the product specification requirements at the point of sale.

---

It would be a pleasure for us to assist you in regard to the GPS issue you are experiencing with your Fitbit Charge 3 using the OnePlus 5T.

After checking in our system, we noticed that due to the large number of Android and devices, Fitbit has not yet tested all models including the OnePlus 5T. Sometimes non tested Android devices may meet all the requirements by the system allowing them to sync and received notifications intermittently even when they are not listed. But, since it's not tested yet by our development team, this connectivity may not always be successful. We recommend to keep an eye on the Fitbit supported devices in order to see if an update has been set.

---

A seller can't play the requirements or "supported devices" card on a sale retrospectively. My phone reaches the "Android 7.0 or higher" requirement being 9.0 (unmodified), and is not listed in the device exceptions. Even if it were to be added now, it wasn't during the point of sale, thus would be irrelevant.

As of 12th June this issue has flipped direction. At the end of the same walk Fitbit now overstating the walking distance at 3.38 miles. Prior to my walk on 14th June I made an adjustment to the stride length (turning off automatic GPS mode) and calculated it as:

205,920 inches (3.25 miles) ÷ 5,400 steps (an approximate average for me walking the same route) = 38 inch stride (rounded)

The subsequent two walks were clocked in at 3.37 or 3.39 miles, so I don't see this change having any impact compared with automatic stride length.

Meanwhile Google is still clocking an ever consistent 3.1 miles. -- It's not for me to tell Fitbit how to do their job, but perhaps asking Google can yield an answer? (They are the parent company now after all.)

I'm going to draw a line under the issue. A product under warranty entitles the buyer to a reasonable resolution should it not function as described at the point of sale. If an answer can't be achieved the most appropriate route is a replacement, or return the product for a full refund.

---

@PRCstuart wrote:

I finally found a likely explanation for this: 

"...All devices with GPS sensors require a direct path to GPS satellites to calculate location. A weak GPS signal might affect the accuracy of your route and other activity data. Your device uses your step count to calculate distance until it finds a signal, so the total distance calculated for a workout may be slightly less accurate when GPS isn't available for the entire time.

Environmental factors, including tall buildings, dense forest, steep hills, and even thick cloud cover, can interfere with your phone’s ability to connect to GPS satellites...."

I run in a heavily wooded area so it makes sense...

---

That doesn't answer the issue. Identical hardware, identical environment, identical route -- Google get an incredibly consistent figure each time. The problem fairly and squarely lies with Fitbit. Enough of us have shown our data to prove this to be the case.

[Nneicho]

Sadly i gave up with.this a year.ago. after contacting the fitbit.ceo team i finally got a refund as customer service would only replace.it which didnt help as they are all the same. So i got a refund onto paypal as they originally only offered me a virtual credit.card which i refused.

If the item. Is not fit for purpose then you iare entitled to a full refund. Which i then spent on a garmin vivoactive 3. Amd i love.it  works perfectly tracking my runs and so far hasnt let me down

[DavidHermann]

My Charge3 counts almost half of my actual distance. This is extremely frustrating, as the counted steps are way too many compared to the "measured" distance. How can a Software be so stupid? Two steps for one yard? Who do the good people at Fitbit believe is running here?: A Hobbit? I must say - I dumped my Garmin because of their ever unstable Software, then my Suunto as their device was not readable in broad daylight (!). Now Fitbit is letting me down too ... Any advice?

Rgds from Asia

H.

[Sandles]

Confession... After 4 months trying to join the Fitbit clan I gave up and
threw (gave) away my Fitbit Charge yesterday. The inaccuracy of distance
travelled being the one and only reason. Why do they have GPS if it's
useless? I wish they didn't claim GPS functionality and would never have
purchased in the first place, since it's no better than a free pedometer
you get registering for annual charity 10,000 step challenge..

Concession... have gone back to my Samsung devices (Fit 2 Pro & S3
Frontier) for the time being. Sure their battery life is poor given their
age, but accuracy of GPS is mint and I can go back to my 5, 8, 10k runs
i've been doing for years that Fitbit was 10-20% less, despite the same map
posted afterwards.

Awaiting... the next Samsung Watch 3, due circa August 2020 and if it's
anywhere near $900AUD, maybe just maybe i'll hit Garmin - anything but
Fitbit, forever.

[RobaJett]

Yeah I've now bought a Garmin Forerunner 245 and I'm absolutely not looking back. This thing is incredible, it even gets run distances correct!