Zumo User Forums

Using an SD card is probably not the most convenient method.
However, based on extensive experience, not using the Tread app, or at least not using tread app data synchronization, and instead using USB or SD card for transfare routes is the most reliable way to transfer routes from other sources to the navigation device.

PS
I would need to test myself whether it is now no longer possible to install the Tread app without an Explore connection, but I prefer to avoid the Tread app altogether.

By all means reset everything and install without Explore sync (that’s if you can still do that). But….to me, that’s giving in and I wouldn’t let it defeat me. We don’t get lessons on this technology which might help so have to work it out ourselves.

I also use MRA having previously used Mapsource, Basecamp and Tyre. MRA works and does everything I need. It’s then very easy to export my plan (1.1 - route, track, POI) into Tread on my iPhone. I use enough shaping points to keep it on the roads I want to ride. On Tread, I can then compare the calculated route with my track. If there any significant differences, I then add another shaping point or two in MRA then do it again. It doesn’t take long. Then turn on the Zumo and after a sync, make sure the route calculates which if it does on Tread, it usually will. If it doesn’t, it’s almost certainly a planning error, a motorway or faster road near your route that that Garmin thinks you should be on (more shaping points needed otherwise Tread will move them). MRA has a habit of breaking up a route if you have been a bit slapdash with your waypoint (shaping, via, POI or waypoint) placing so also check that.

As has been mentioned, all the platforms have different ways of working so it can be a challenge to get everything right. On previous satnavs, it was always hit and miss if your ridden route was exactly what you planned and sometimes hard to tell. If you ride in groups, how many times did we all end up with different routes from the same plan? Even with the same avoidances set and the same satnav, it could be different. Then we learned how to display the track as well as the route. GPX 1.2 was invented by MRA in order to ‘nail’ your route to the track. Great, we could all do that and get the same route……until you deviate then disaster, an exploding route between via points. Solved by turning off auto- recalculate but then you have to navigate yourself back to the route (using 1.1, we get guided back on to it - if we copy the route to avoid RUT - yes, another issue!)

I’m sure there’s something I’ll have missed there but I seem to have got MRA, Tread and the XT2 working for me. As the tech has advanced along with our knowledge of what we can do with it and our planning has got more intricate, the more we expect it to just work. This kit needs effort I’m afraid.

Nailing the route down to the track while keeping viapoints sounds good enough to me and was what I was trying to do with shaping points. But those are influenced by tread. So far the nailed down route stays in place even after tread syncing back and forth. Unfortunately the one day spring has turned into fall already in the Netherlands so will try some more 1.2 routes when the weather turns a bit better. I think I can live with some flexibility from my side in case of a closed road and keep the recalculating on after approval, at least my actions are more or less predictable . And the navigation picks up again when back on route. Another option I guess is to stop the navigation after the detour, restart the original route and then start at closest point, that worked the other day. I expected it would go for the next viapoint but that didn’t happen. It calculated a route back (forward) to the planned route.

It took a while but I have come to accept that Tread/XT2 don’t play as Garmin says. And by now there are several options to make one or the other work acceptable. Off course that also means several opinions and different preferences.

I believe the Tread app also received an update recently. I now have a message that shaping points are limited to 10 between viapoints FYI

Regain wrote: Fri Feb 27, 2026 11:22 pm ...
GPX 1.2 was invented by MRA in order to ‘nail’ your route to the track. Great, ...

MRA: Please tell me why?
It shouldn't be the case that everyone builds their own GPX format interpretation

As far as I know, www.topografix.com/gpx.asp is the source of the GPX specification and no one else, not even MRA .

Using the route point extensions, it has been possible for a long time, since the official GPX 1.1 schema, is part of the GPX standard, to “nail” the track to the route (using “ghost points ”)

Regain wrote: Fri Feb 27, 2026 11:22 pm ... but I seem to have got MRA, Tread and the XT2 working for me. ...

It's nice to read that it works for you

But unfortunately, it doesn't change the fact that there are a few frustrated users for whom it simply doesn't work

When the Tread app is involved in synchronizing routes, unwanted things can (will) happen
Shaping points can be moved unintentionally, that's just the way it is, unfortunately
Synchronization can create seemingly endless loops, severely impairing the operation of the device and even causing it to restart
Synchronization may also result in duplicate trips being created in the background, or routes may be unable to be calculated (temporarily)
...
And last but not least, the problems can also occur even if the routes have already been correctly calculated and were accessible on the device.


But we have been criticizing all of this for years now, and nothing significant has improved to date.

Your suggestion work flow is still worth a try if it helps you, and maybe others too,
but don't expect success too much

restart the original route and then start at closest point, that worked the other day. I expected it would go for the next viapoint but that didn’t happen. It calculated a route back (forward) to the planned route.

Closest Entry Point (CEP) seems to find the closest point on the existing route then calculate a route from current location to that point. See this post viewtopic.php?p=20805#p20805 for details if interested.

I believe the Tread app also received an update recently. I now have a message that shaping points are limited to 10 between viapoints FYI

I could easily live with a limit of 10 shaping points between via points if the shaping points stayed where they were placed. Maybe I'll get brave enough to put tread back on my phone and test the new version.

@BasPost where did this message about 10 shaping points between via points originate? I have some routes planned for a tour in June and each one of them calculates both in Tread and the XT2 with many more than that. I’ve just tried one of them that has 37 shaping points between vias and it runs just fine.

My Tread and XT2 are both recently and fully updated.

@proofresistant I don’t profess to be a GPX expert but I don’t think you can build a GPX 1.2 route outside of MRA and I believe it was their method to enable exactly what I described earlier. With 1.1, the Zumo can still decide its own routing between via and shaping points. Not so with 1.2, SPs are stripped but not really needed with auto-recalculate turned off. It’s a similar way of building a trip from track but keeping your via points. Please correct me.

Regain wrote: Sat Feb 28, 2026 5:58 pm @BasPost where did this message about 10 shaping points between via points originate? I have some routes planned for a tour in June and each one of them calculates both in Tread and the XT2 with many more than that. I’ve just tried one of them that has 37 shaping points between vias and it runs just fine. ...

Maybe it's not 10 shaping points in total in whole route, but 10 shaping points between every two waypoints?

@Francis that still doesn’t seem right, as stated I have 37 between two via points (start and finish of the 186 mile route). I haven’t used any specific ‘waypoints’ in the Garmin meaning of the word. As an aside, I wish there was a renaming convention to clean up all our understanding of the different route points across all platforms.

Regain wrote: Sat Feb 28, 2026 5:58 pm @proofresistant I don’t profess to be a GPX expert but I don’t think you can build a GPX 1.2 route outside of MRA and I believe it was their method to enable exactly what I described earlier. With 1.1, the Zumo can still decide its own routing between via and shaping points. Not so with 1.2, SPs are stripped but not really needed with auto-recalculate turned off. It’s a similar way of building a trip from track but keeping your via points. Please correct me.

I think the use of the MRA naming convention GPX 1.2 is or was somewhat misleading.

I just took a look at one of these files.
At first glance, the MRA GPX 1.2 function also uses parts of the GPX 1.1 standard and exports the GPX file as <gpx version="1.1".
Compared to Garmin, however, MRA “only” uses a small part of the GPX 1.1 standard, but has probably not invented anything new.

The big difference, however, is that no shaping points are actually output with what they want to call 1.2.

It may be true that you don't need shaping points if you turn off recalculation.
But keep in mind that some people really thought about it and decided that both shaping and recalculation should exist.
If someone can work well without using both of these valuable functions, then that's a big advantage, at least for him personally.

But seriously, MRA, what are you thinking with GPX 1.2 and why are you leaving out valuable shaping points?
And why don't you call it what it is when you export it? For example: Export GPX 1.1, routes without shaping points with track details in the route point extensions, or something like that.

PS
When I have plenty of time, I'll try out for myself how well these so-called 1.2 files can be used, but for now I'm putting that plan far into the future.

@proofresistant just to point out, I don’t use GPX 1.2 anymore since buying an XT2. I use 1.1 (track, route & POI), I have Tread installed and syncing. I build routes with many shaping points, occasional (rare) via points and leave auto-recalculation on.

I’ve merely been trying to explain where I believe 1.2 came from, what it was intended for and how it worked. It can still be used with the aforementioned conditions. I’m not its defender.

I hope that clears it up.