Fenix 6X : critical underestimation of the distance

O.Ret. said:

But see below some screenshots with the GPX

Unfortunately those are not helpfull for the diagnostic, at all. Please post the screenshots from Garmin Connect Web, following the steps I described in the previous post:

trux said:

a screenshot of the elevation graph with superimposed pace on it - on the respective activity screen of GC Web, click the double-arrow symbol in the upper right corner of the elevation graph, and then, on the zoomed version of the graph, select the Pace Overlay in the uper right corner.

.

O.Ret. said:

3d speed & distance are deactivated on the F6X (for 3 & 4 but not 2) but activated on the F5X for all of them.

Why do you want to compare apples with oranges? If you want to compare data from multiple devices, make sure they use identical settings. 

Since you are displaying the chart by distance on the X axis, it's normal to see a difference. They can't be synchronous if the F6 watches are recording less distance. My F5x was also too short, and I don't have the feeling that my F6 is worse. I should try to run with both to compare. 

What is interesting is that the F5x is recording more distance suddenly around km 2.5 for one run. What happened there ? 

philipshambrook said:

Data is ALWAYS recorded at 1s intervals IRRESPECTIVE of whether Smart Recording or 1s is selected. Those settings do not affect the any distance or pace measures. The difference between the two settings relates to the track display on the map - it might not follow paths as closely as 1s recording - and the size of the resulting data file. Smart recording discards data points if there is no change in some of the measurements for x number of seconds however, all the data points are used for calculating distance etc.

I'm not sure if I believe this. If you are going around a bend such as on an oval 400m track, the smart recording track will cut the corners as it will joining the dots, whereas the higher resolution 1 sec recording will provide a longer, more accurate track.

Believe it or not, it will stay a fact 

https://support.garmin.com/en-US/?faq=s4w6kZmbmK0P6l20SgpW28

Bonjour

Thanks for your replies.

here the screenshot. The place is pretty flat (35 to 50 m)

Will retry with 3d stuffs activated later indeed. However it hasn't made any big difference when I had deactivated them.

@Risbac : en effet, merci pour le rappel ...

Using time as X Axis there is still a ~15 seconds delay on the F6X (blue) vs F5X (green) GPX

The speed is also very volatile on F6X (blue) vs F5X (green), esp on the middle area and on the acceleration there is also 15 sec delay

O.Ret. said:

here the screenshot.

Yes, it is exactly what I assumed. The pace graph contains 7 gaps with 0 pace. If you zoom closely on the gap (selecting a small area around it) and move the mouse pointer over the edges, you will see that the time stamps of the edges differ. When you then open the same graph in the Distance mode instead of the default Time mode (see the buttons over the graphs on the Activity page), and zoom on the same gap, the edges will show identical distance. This happens even if you never stop during your run, so it clearly proves that the distance algorithm is faulty - it excludes those gaps from the total distance. 

In your case, by an estimate, the gaps would make around 3% of difference in the total distance. You could tell it better if you zoomed on all the gaps, and noted the times of all their edges, summing up their total duration and comparing with the total time of the activity.

This is clearly a bug in the algorithm calculating the total distance of an activity, on many recent Garmin models, so we can only hope it will be recognized and fixed in one of the future firmware updates.

The importance is to send this kind of data's to Garmin development team and not only speak about it here. 

To have simple datas could be great to compare a Garmin using GPS and a Garmin using Stryd. 

Well, I really hope Garmin will do something about this. Today I ran a 14.8 km trail and it only recorded 13.9 km. That makes a difference of 20 seconds in my average pace (5:50 instead of 5:30). For me, this is relevant. What bothers me is that I know they could do better. I used a Forerunner 220 for years and it was much more accurate than the Fenix 6! Unfortunately I sold it. I've got no time to collect data and send it to them, I did it many times with other bugs and I am tired of it. They can look at all data in my garmin-account (hundreds of trail runs) and easily find out themselves that there is a inaccuracy.

I really think that you need to do a controlled test. All you know at the moment is that your watches give different measurements, not that the Fenix 6 is wrong. Any distance measured on a map will not translate well to that over the ground on a GPS device unless you move in a straight line for a great distance.

I suggest you either try the watch on a measured (with a wheel, not online), paved course such as a known 10K route, or a track.