Tread Changes Exisiting Routes - Shaping Point Issues

[proofresistant]

Trying to figure out why all this is happening, and exactly where the problem(s) are, is an interesting puzzle, but the bottom line is that we users have only a single choice - either let the Tread App sync or don't.

Only Garmin can fix the The Tread App's shortcomings and the issues it causes.
...

@smfollen As you know, I largely agree with you.
To be honest, I don't want to try anymore, even though I had so much hope.
Wireless is and remains my dearest wish.

So, I've been testing a lot over the last two days, again.
With my zumo XT2, with my Tread 2 also.
All results were comparable, with one exception: the Tread 2 can also synchronize directly with the Explore cloud via Wi-Fi, and the zumo XT2 cannot.
The rest is more (or less) comparable.

So here are my findings edit: for the moment.

• Tread 2 and Explore Cloud
  No problem, after device-cloud synchronization, shapes remain as planned, in Explore cloud also

• zumo XT2 and Tread 2 synchronization with the Tread app offline (Wi-Fi and mobile data disabled)
  No problem, after Tread app-device synchronization, shapes remain as planned
  Note: the Tread app cannot calculate this.

• GPX import with the Tread app offline (Wi-Fi and mobile data disabled edit: , bluetooth disabled also)
  No problem, after synchronization between the Tread app and edit: (not the device), shapes remain as planned
  Note: the Tread app cannot calculate this way

• edit:
  GPX import with the Tread app offline (Wi-Fi and mobile data disabled, bluetooth enabled, zumo XT2 is connected)
  Well-known problems, shapes are moved
  Not Always, in my tests with the Tread 2 and Chatham Loop GPX i didn't saw those issues in such test case
  But yet, with my WK 3B Route for the zumo XT2 and tread 2 also, Shapings are shiftet like before with tread app
  The Tread app cannot calculate this way, but the zumo XT2 can

• Use of routes with only VIA points
  No problems, regardless of the configuration, VIAS remain as planned

• Use of routes with VIA points and shaping points, but the routing algorithm must be direct (no faster, no shorter, and no ADV).
  No problems, regardless of the synchronization configuration,
  VIAs and shapers remain as planned
  These shapers are then also final in the Explore cloud, with all shapes unchanged

• But then,
  Tread app online (Wi-Fi and/or mobile data enabled)
  Well-known problems, shapes are moved
  I don't think it's the synchronization. I think it's the route calculation method of the Tread app itself.

Conclusion:
If the Tread app can optimize the shapes, it does so, but it does it the way it thinks is best, which unfortunately isn't really optimal.
edit:
May zumo XT2 and or Tread 2 does changes also, sometimes? if routes comming from Tread app?
It's a nightmare. Whenever I think I've understood something, I start all over again shortly afterwards.


PS
@smfollen
by the way is that your noted deviations have roughly the same effect as my own examples.
What confused me at first was that it looked like an 8 to me at the beginning, but after the Tread app “optimization,” it no longer did. But that was probably due to the route selection, and if the northeastern section is avoided, it looks a little less like an eight.
I also think that the Tread app has improved at least a little. Not significantly, but marginally.

[proofresistant]

So here are my findings edit: for the moment.
...

It's a nightmare. Whenever I think I've understood something, I start all over again shortly afterwards.

I have revised the upper section “findings...” again.
The reason is that I have also managed to get the zumo XT2 or the Tread 2 to shift shapes when the Smartphone Tread app was offline and only the connection to the navigation device was enabled via Bluetooth.

[Oop North John]

It's a nightmare. Whenever I think I've understood something, I start all over again shortly afterwards.

I have revised the upper section “findings...” again.
The reason is that I have also managed to get the zumo XT2 or the Tread 2 to shift shapes when the Smartphone Tread app was offline and only the connection to the navigation device was enabled via Bluetooth.

Was your XT2 still connected to wifi, as then the XT2 / Tread would still be able to connect to the Garmin cloud?

Congratulations on your patience and work with this!

[jfheath]

Thanks @proofresistant very similar to my own results on the XT2 Making it Work for You pages.
Mine was a very targetted series of tests on a single route that runs close to a motorway. So it illustrates behaviour rather than proves it. So its good to have other examples of a similar nature. In general Via points seem to be pretty robust - with/without Tread being Active. Shaping points are always possible candidates for being relocated with tread active. The use of Waypoints to form the shaping or via points does not alter this behaviour - but it does prevent the XT2 from changing the name if Tread is not synching.

Was your XT2 still connected to wifi, as then the XT2 / Tread would still be able to connect to the Garmin cloud?

I've not checked this - but I assumed that wifi connectivity on the XT2 was purely for downloading maps. Are you saying that without the phone connected, but with wifi active on the XT2 - the route is synched back and forth to the Explore datbase / software at explore.garmin.com ??

That hadn't occurred to me - I always have wifi turned off. I may have made a rather stupid assumption if that is the case.
(Not beyond the bounds of possibilty).

[proofresistant]

Was your XT2 still connected to wifi, as then the XT2 / Tread would still be able to connect to the Garmin cloud?

I believed I have described my scenarios in detail above. However, I will take another (deeper) look and add this facts to the list.
But my tests covered virtually everything, including with and without a Wi-Fi connection to the device.

Tread 2 is able to connect to the Explore cloud via Wi-Fi and without the Tread app.

zumo XT2 is not able to connect to the Explore cloud via Wi-Fi on its own; it needs the Tread app to do so. That's what I've observed..

In general Via points seem to be pretty robust - with/without Tread being Active.

Thats true in my opinion also.

Shaping points are always possible candidates for being relocated with tread active.

That's mostly correct, i observed,
but then I found the test with the sectioned direct route calculation, where the shapings aren't adjusted either because nothing is being calculated in my opinion

Was your XT2 still connected to wifi, as then the XT2 / Tread would still be able to connect to the Garmin cloud?

... Are you saying that without the phone connected, but with wifi active on the XT2 - the route is synched back and forth to the Explore datbase / software at explore.garmin.com ?...

As I just wrote to Oop North John, Tread 2 can operate independently, but zumo XT2 cannot synchronize with Explore Cloud on its own.




To me, it seems that it's not the synchronization with Explore Cloud,
but rather the calculations of the Tread app and/or the devices that are shifting the shapes.

I'll do another test step by step
Testing with a changed Explore Cloud password.
I already tried that and, as I remember, it also led to the problem, but I'll test it again carefully.

[proofresistant]

I'll do another test step by step
Testing with a changed Explore Cloud password.
I already tried that and, as I remember, it also led to the problem, but I'll test it again carefully.

Quick update
By changing the password on the Explore website only, I can't bypass the Tread app Explore Cloud synchronization, even if the Tread app doesn't know the new password.
Similarly, I tested independently, that once a Tread 2 device is set up, changed passwords don't really matter.
I have also noticed that synchronization did not occur at times, but this was really only rare, probably due to other circumstances.
Ergo, changing the password does not help in my cases

[jfheath]

My testing deserves an extra bit of information. I chose that section or road becasue I know that the Tread App will mess it up all by itself.
I know that some of the issue will be to do with faster time and the presence of a motorway
And I know that I have a series of either 6 shaping points or 6 via points between start and end.

In other words - I know that the test route would fail to stick to the route given the opportunity.
What I had never documented before was exactly when it failed and when it didn't.

I have seen circumstance when I have used a combination of Via and Shaping points on the same route and slthough shaping points have moved, they have not affected the outcome of the route.

I agree about the suggestion that it is the Tread/Explore/Cloud where significant alteration occurs - although you can see the position move if you use the XT2 screen to change a via to a shaping point. I also note that I sometimes have to force a recalculation to make it happen. (it will do it by itself eventually). Edit the name to add a full stop. Also the synch has to be in both directions - the orange Tread symbol with the arrow and the blue circle. I wait for this to happen twice before observing results.

[proofresistant]

... Tread/Explore/Cloud ...

Even if I am repeating myself please try to consider my observations in your thoughts
namely that synchronization with the Explore cloud may has no negative effect.
Shapings are shifted, more or less randomly, at nearly the same time that the Tread app calculates the routes, while “optimizing” them in a "suboptimal way", and after that also synchronizes them with the Explore cloud.
The fact that everything works so quickly probably led me to believe, may wrongly, that zumo XT2 “optimizes” also suboptimally. I'm not completely convinced of that anymore, but I've largely moved away from that belief.

[proofresistant]

Perhaps there is still hope, at least for my wishes, to become reality.

After unsuccessfully attempting to block access to Explore Cloud by changing the password, even though I didn't believe it would have any positive effect, I made two more attempts to find a workaround.

Attempt 1 failed
I tried to block the Explore pages in my Wi-Fi router firewall, but this was unsuccessful, probably due to a lack of in-depth knowledge.

Attempt 2 is promising at least for now and according to initial observations
The goal again is to block the Tread app's network connection.
For Android “NetGuard - no-root firewall” seems suitable for this: https://play.google.com/store/apps/deta ... e.netguard

At the moment, I can't receive any weather or traffic data, but I never need or use it , partly for reasons that are well known.
What's much better is that the Tread app can't calculate without Internet (nor can it synchronize with the Explore cloud).
Anyway, after almost two years, I seem to be able to get routes with shapings onto the navigation system for the first time without any problems
I'm going to try it now with both navigation devices (zumo XT2 and Tread 2 at the same time), both connected to a Tread app. Wish me luck, I'll let you know how it goes.

Who knows, maybe at least my dream will finally come true.

[jfheath]

Even if I am repeating myself please try to consider my observations in your thoughts
namely that synchronization with the Explore cloud may has no negative effect.

I did, I did - and I agreed:

I agree about the suggestion that it is the Tread/Explore/Cloud where significant alteration occurs - although you can see the position move if you use the XT2 screen to change a via to a shaping point.

I’d come to the same thought myself, but am aware that route points move without Tread or synch being involved too. Nowhere nearly as dramatically though.

If it is any help in your observations - I have just been trying to create a demo of the route altering when using the Tread App on my ipad. I have a video of me doing just that a while ago, but not a very good one. Now, I cannot do it. Tread is installed on my ipad. It is connected to wifi, it is connected to the XT2 as if it were a phone, by BT - but the ipad has no SIM card in it - and now the Tread app on the ipad just spends ages calculating with no sign of a route to join the points. Do it on the phone with a Sim card in, and it produces a route.

So I could easily conclude that the Explore/Tread/Cloud is the thing that is doing the calculation in this instance. Certainly not the Tread App on the ipad. I'd need to put the Sim card from the phone into the ipad to confirm this.