function EXPAND

[Karel76]

I did not create it directly in OSM, but in MRA through the OSM layer, see picture. Yes in the settings it was zigzag, out of town, off highway. I liked the route and that's why I wanted to add more waypoints, so that the navigation (MRA) would navigate me exactly along this route, because I know that MRA follows HERE and it has a different algorithm for creating routes and would guide me differently. Unfortunately after adding 50 points it created a completely different route. I kindly assumed that it would add the points to my original route. Now I hastily made a second route and the same result.

[Karel76]

there is link on second route
https://www.myrouteapp.com/en/social/route/11665201?mode=share

[Karel76]

there is video for example how created route
https://k00.fr/kvxz2d3t

[Marinus van Deudekom]

2 possibilitys
When you’re done like in the first video, try to export that as a gpx track
Then upload that gpx to MRA and open it while Here is your standard map
2 put a lot shapingpoints, I mean a lot, by hand in that first route and then change the map to Here

[Jörgen]

I am a bit confused? I usually use the same process with open street maps. I have no idea why you get a totally different route when you expand with more waypoints.

[Jörgen]

@Marinus-van-Deudekom said in function EXPAND:

2 possibilitys
When you’re done like in the first video, try to export that as a gpx track
Then upload that gpx to MRA and open it while Here is your standard map
2 put a lot shapingpoints, I mean a lot, by hand in that first route and then change the map to Here

I tried this as well and was surprised that the gpx file (the track) is different when you export (save) it from the original.

Whats wrong here?

I made routes with open street maps by using sections and curvy roads and filled it as well with waypoints to use it for navigation with Here map. I can't remember having this problem.

[Karel76]

So probably a bug, because I used this feature once and it worked as it should. Now it's unusable. Thank you for your efforts. Is there any way to send this bug to the developers?

[Jörgen]

Yes, contact the support.

[Con Hennekens]

@Jörgen said in function EXPAND:

I am a bit confused? I usually use the same process with open street maps. I have no idea why you get a totally different route when you expand with more waypoints.

I am not so sure. This is not my method of creating routes, but isn't it so that Curvy options have some kind of randomness in it, exactly to prevent always getting the same roads? I am not surprised that this happens, I would be surprised about how it worked before when it did not change the route

[Marinus van Deudekom]

@Con-Hennekens I've created a route using that tool in the TomTom map and couldn't get rid of the algorithms when I extended the route with a piece of highway. The way out was then to export the curvy part as a gpx track with I then uploaded in MRA. Yep a bit complicated

[Karel76]

So I found a bug. when I perform the same action on the base of TOM TOM maps, the "EXPAND" function works as I want. On the OSM base it generates a different route. I attach a video of the creation on TOM TOM

video TOM TOM (there is OK):
https://k00.fr/n1vrs605

video OSM (wrong route):
https://k00.fr/kvxz2d3t

[Jörgen]

@Con-Hennekens that is also not my usually method.
But when I create a curvy route (with what ever map) it should be possible to use them in MRA navigation as created.
So, why is the exported gpx file already total different to what was seen on the screen?
And why do I get a totally different route, when I expand what I see on the screen?

[Erdna]

@Karel76 said in function EXPAND:

So I found a bug. when I perform the same action on the base of TOM TOM maps, the "EXPAND" function works as I want. On the OSM base it generates a different route. I attach a video of the creation on TOM TOM

video TOM TOM (there is OK):
https://k00.fr/n1vrs605

video OSM (wrong route):
https://k00.fr/kvxz2d3t

That is not a bug. TomTom and OSM use different card material and different algorythms for calculation of the route. Compared with different setting like cury route, restrictions like don´t use highways, etc. that behaviour is absolutely normal.

[Jörgen]

@erdna, you may be right. But this is not the point. Why is MRA making a totally different route when waypoints are added (with openstreet map)?
WP should be added to the route that was created with OPM as well as it is done with TT map.

[Con Hennekens]

@Jörgen said in function EXPAND:

Why is MRA making a totally different route when waypoints are added (with openstreet map)?

Seemingly that is due to a randomize function in the automatic creation of scenic routes. A curvy route between 2 points can be different from a curvy route between three points, even when the middle point is on the original 2-point route. That does not sound completely illogical to me.

WP should be added to the route that was created with OPM as well as it is done with TT map.

Remember that routes are calculated by the map-providers, not by MRA. Each map provider does things differently. That is nothing new.

[Rainer Treichel]

OK,
I'll leave MRA and a software problem aside for now.
Adding loads of shaping points seems impractical to me.
There is <Compare route calculation with>.
Then you set shaping points to make the routes congruent. Incidentally, that was one of the parts involved in the creation of MRA
Or you can display the route as an overlay (on the left side of the menu). This would also work with a track log. Or simply the desired route. And on this basis, you then create a new route with the route engine of your choice.
I find the first alternative nicer and faster.
Interesting topic, maybe I'll add it to my to-do list for the next video conference.
RT

Translated with DeepL.com (free version)

[Jörgen]

@Con-Hennekens said in function EXPAND:

@Jörgen said in function EXPAND:

Why is MRA making a totally different route when waypoints are added (with openstreet map)?

Seemingly that is due to a randomize function in the automatic creation of scenic routes. A curvy route between 2 points can be different from a curvy route between three points, even when the middle point is on the original 2-point route. That does not sound completely illogical to me.

WP should be added to the route that was created with OPM as well as it is done with TT map.

Remember that routes are calculated by the map-providers, not by MRA. Each map provider does things differently. That is nothing new.

Ok, I understand that different maps generate different routes. The curvy function in OSM and TT are fine. And with TT generated routes, I can take them nearly 1:1 when I add WP to use with HERE (and navigation).
But why does it not work with OSM (nearly like in TT)?
If I can not use the generated route, what sens does this function make?
Or, how to solve this issue easily? (ok, you could add WP manually

[Lex.Kloet.RX]

@Con-Hennekens said in function EXPAND:

@Jörgen said in function EXPAND:

Why is MRA making a totally different route when waypoints are added (with openstreet map)?

Seemingly that is due to a randomize function in the automatic creation of scenic routes. A curvy route between 2 points can be different from a curvy route between three points, even when the middle point is on the original 2-point route. That does not sound completely illogical to me.

Sorry Con, but it is totally illogical. With this tool, one expects MRA maintains the route. The route is already calculated, why not put the new added points on the route "as is", that would be logical. There is no use of a "new calculation" because the parameters for calculating the route didn't change.

[Lex.Kloet.RX]

By the way, I was there in that area, it is very nice!
https://routes.lexwill.cyou/cabriotour-tsjechie-2025/

[Con Hennekens]

@Lex.Kloet.RX said in function EXPAND:

With this tool, one expects MRA maintains the route.

Sorry Lex, but MRA maintains nothing except a web planner tool. Calculations are done by mapproviders. Each adding of a waypoint triggers recalculations. Rest assured: I really don't know whether OSM has randomize functions in it's curvy routes calculations or not. But what happens here makes me think so. And it would have a purpose. No one wants the exact same route each time they create a roundtour from home.

But I agree completely that the difference shown by in the first post are pretty dramatic.