Fenix 6X Pro - why do my GPS tracks have straight line sections?

Wow, very interesting, good effort there.

I've had similar thoughts and issues with planning bike trips abroad. In the UK Strava predictions for total height gain tend to match the actual recorded ride stats afterwards. When riding on the continent the Strava distance is about right, but total climbing is always too much. We did a ride from Chamonix to Monte Zoncolan over 2 weeks sometime ago and were expecting days over over 4000m of climbing, they turned out to be much less than this when recorded by the Edge GPS. I wasn't sure to be relieved or disappointed!! Turns out actually trying to measure total height gain is a bit of a black art whatever system you use, to predict or actualy measure. A bit like a BBC Coast TV programme that explored how to measure the length of the UK coastline ... total length seemed to depend on the size of your ruler! Point being, there isn't one definite measure.

Same kind of applies to height gain, what is the gap between the contours your digital map model uses. The UK is different to the continent I think.

Roads in the Alps are another example in a way. The Km markers on big cols / hills reel off each Km travelled, I assume tyres over tarmac distance. But the 'height gain' is simply the difference in actual altitude between the 2 points in space. No account is taken of any undulations in-between. So the 'average gradient' stamped on the Km marker you read as you are sweating up the hill, while mathematically correct, isn't really much help, or indication of how hard it is going to be, unless it's a perfectly smooth incline with no dips. You become quite good at mental maths watching these things go by slowly, in the heat! 

The relevant point for me here in this conversation thread is I tend not to fuss the actual figures. (different for runners perhaps who like to detail things like pace?). Hard to know with one device if the numbers are 'correct'. But consistency between days out is kind of important.

I still don't like the straight-line bits on my recorded tracks though... I wish Garmin would do something to stop this happening...

With all its faults I persevere with my wrist GPS for walking and mountaineering as it's very convenient to carry/wear as opposed to anything hand-held. When in extremis it's easy to glance at the wrist to help with navigation rather than fussing something hand held. Though we can and do stop to look at proper paper maps and larger phone screens when the need arises.

That is a perfect illustration of the problem. Can you just confirm which version of the firmware you're on please?  Data recording 'every second', auto-pause off etc? I think this is worth sharing with Garmin 'cos it's a bit crap! It's certainly spoiled most of my tracks. I spend a lot of time in the mountains going very slowly uphill.

Were you watching the map at the same time?  I assume the GNSS receiver always showed you in the correct place and it's the algorithm controlling the tracklog file that's causing the problem?

Hi Keith. Another mountain fan here. I love the convenience of having the highlighted route on my wrist. However I was expecting GNSS-accurate tracks. I spend a little time contributing to OpenStreetMap and an accurate track is of course a valuable aid to plotting an unmarked path. The straight-lining the watch does makes it unsuitable. I'm disappointed with that aspect of its performance. Anders seems to have sussed out the problem.  I'm going to make some similar tests myself and complain to Garmin if I can reproduce the behaviour.  'Every second' should mean 'Every second'.

The tests was done 2021-10-27 with firmware 19.75 but I don't think anything regarding this have changed since then. Data recording every second and auto pause was off. I didn't look at the map when the tests was done so I don't know how that behaved. 

Yes, AndersB’s test seems a really good one to replicate here too. Once the straight line thing is reproducible consistently we can all challenge Garmin to do something about it, and we can check to see if any attempts at a fix are effective too. Now, where can I find a suitable place to walk very slowly!!

I like to update OSM data everywhere I go too. Amazing how in Scotland the OS maps show paths that no one walks any more and are abandoned or obsolete. My wife and I have fallen into the habit of checking the Strava heat map (make sure to select walk or biking as appropriate, they are different!) when planning a new route to walk or climb the next day, it’s so handy to see where actual people have been walking. A solid blue line means the path is obvious. Braided faint blue lines all over the place usually means it’s a bog!! Often we don't even bother to draw a .gpx route, just seeing the blue lines on-screen the night before is enough, can even do a quick screen capture to refer to when out on the hill. She even asks me to research new obscure places here at home in Snowdonia before leading a small group of friends out on the hill, quite a handy resource I think.

I often refer to my created and uploaded track on a phone screen when actually walking as well as the Fenix 6X. Though the 7X with touch screen may be great in this respect to be able to scroll around and zoom the maps a lot easier. I never bother with the button scrolling on the 6X.

One really good feature is ClimbPro screen to display a profile of the actual hill being climbed at the time (on foot or bike). You can only see this when following a pre-defined course. It's very powerful, but I don't use it al that much as you have to follow a course you've drawn beforehand.

I can edit actual OSM data from my uploaded ‘traces’ (as OSM call the .gpx tracks) at some point later on to help fellow walkers see the best route. I also have GoMap! on my iPhone which is a really good full OSM editor. Means it’s possible to add features and even edit paths in real time, while out on the hill . Though it can be a bit anti-social to try and do this too often when out with partners and friends . Quite spooky to see the edits appear either straight away on some mapping products, or a while later on with others. My OSM edits even show up on Strava maps eventually. Fixing bad road / paths junctions that previously weren't routable. 

Also … I have found the ‘TalkyToaster’ mapping looks really good on the Fenix 6X, better for me than the Garmin Topo it comes with. OS 1:50 mapping isn’t very good on the small screen. Though, on the road-bike, I stick to the Garmin Topo on the Edge 830 as all I want is roads, keep it simple! Especially when riding the bike in the Alps.

Brilliant, thanks Anders. I will try this slow walking test out with my 6X as soon as I can… I'll look forward to running this reproducible test with the 6X and my new 7X when it arrives to see if it's any different.

Any advice on how slow is slow? You have some numbers in your post. Curious how that relates to 'on the ground' so to speak. Something like, one step, pause, 2nd step pause, or whatever!

Looking forward to badgering Garmin with the data / tracks they can't wriggle out of ...

Another OSM fan and volunteer here. :) I try to map frequently my area on OSM. And actually, there are not a lot of mappers around. Also if I'm going to attend some race or activity, I map the area a few months ago, so it can be displayed on all OSM based sources. However, I usually use OpenTopoMap and its updates are weekly.

Strava heat maps imagery on JOSM is helpful as on satellite imagery some tracks are not present or some are but can't be used easily. Heat maps helps choosing the highway to tag them.

Also Fenix GPS accuracy is good enough to use for mapping as even the satellite imagery is not perfectly aligned to objects. The highways I draw slide to a different side after an update. Older drawings with older imagery is worse as they're goinfg out of road...

When I did my tests I walked like in slow motion without stops. I think the speed threshold is somewhere around 0.3-0.5 m/s so you need to walk slower than that. And with a cadence less than 60 per minute (less than 1 step a second).

I first discovered the problem when I was mushroom picking and then I walked really slow (and with low cadence) with occasional stops when I found mushrooms. 

There might be algorithm differences between different activity types. But not necessary. I used the "Walk" activity for my tests.