Weird Routing Behaviour (2)

[FrankB]

Indeed @jfheath I have been looking into the subclass field. Until now unable to find out more than already reported to you.

Just to make sure what is meant by 'Recalculating on the XT':

A) A route recalculated already loaded on the XT. For example by changing to 'Straight Lines' and then back to 'Faster Time' , or
B) A route that is recalculated when loading into the XT. For example when it was calculated in Basecamp with a different Map from the XT.

I ask because the latter (Option B) seems to be the case when you import from MRA, or when the 'subclass is patched'.
I expect to do a few tests on Wed./Thursday. Not on the Bike, but by car! I have been called 'pussy biker' for a reason....

[jfheath]

That's a good point to make @FrankB .

Recalculation occurs on transfer :
a) If the Basecamp map used differes from the XT map loaded
b) If settings in Basecamp's Options ->Device Transfer are selected.

It Recalculates on load if
c) The vehicle / transportation mode is set differently in the XT from that what is set in the BC route (the profile)
d) If an inappropriate profile is set in Basecamp - (it defaults to using motorcycle)
e) if route preferences or avoidances are changed onthe XT
if any of the last 3, the XT will prompt

It calculates when a route is active .
f) If you deviate from the plotted route. I believe it recalculates the whole route. The 590/595 recalculated just the current section. This imcludes losing all of the ghost points that hold the Basecamp route in place.
g) If you Skip a route point - the entire route gets recaclulated

For most of my tests, I force the route to recalculate as soon as I have transferred it - usually by changing vehicle and back or by changing routeing to straight lines and then back to Faster. Doing this ensures that each test is comparable with the previous.

So far my tests for the reapeated U turns hav involved.
Basecamp Route, all planned with Flag saved Waypoints.
All Waypoints set as Via Points
Transferred to XT.
Route forced to recalculate.
Start of route a good way from my start point.
Extra Via Points near 1-3 miles from the start so that I can skip a point.
Later, deviating from the route which the XT has calculated as faster, but which I want to ignore.

That has always created the RUT scenario.

But if I repeat the test using the same identical Waypoints (now in Favourites) to create a route on the XT Trip Planner. The behaviour is as it should be.

B) A route that is recalculated when loading into the XT. For example when it was calculated in Basecamp with a different Map from the XT.

I ask because the latter (Option B) seems to be the case when you import from MRA, or when the 'subclass is patched'.

MRA gpx v1.1 (Route, Track, POI)
Contains no routing information (ie ghost points or gpxx: route point extansion). It is like sending the XT a route joined with straight lines. The XT has to recalculate it (or it may need to be forced to recalculate it).

MRA gpx v1.2 (Route, Track, POI)
Contains the ghost points / route point extensions - so it retains the original plot of the route that was created in Basecamp. But there are no shaping points - only Via Points.

Also - the last time I looked, which was just before Christmas, MRA's POIs were similar to Garmin's Waypoints - but they did not show up in the XT's 'Favourites' (or 'Saved' in USA).

Garmin's POIs are implemented slightly differently from Garmin's Waypoints. For Garmin, a POI is a saved location (like a favourite, and like most POIs), but GArmin's POIs will alert you as you get within a specified radius - whether or not it is part of a route. In fact you didn't need a route to be running. The alert can be a bit of music, a voice recording, or as in the case of speed cameras a ping.

Great fun to play with - although I haven't done this since I had the 550 - I programmed in the location of all of the Andy Galsworthy stone sculptures and the Henry Moore sculptures that are scattered in various surprise places around the countryside. Not wanting to make a special trip to see one, as some have been removed over time, the satnav alerted me if we ever got to within a 5 mile radius of one. 'Henry Moore 5 miles away' in my own voice.

But I digress. Again.

This could be useful for tour guides.

[Oop North John]

Has there been any update from Garmin on this problem?

[jfheath]

Has there been any update from Garmin on this problem?

No. They have gone quiet. I thought I'd leave it for a while. They replaced my XT Zumo and I asked if they wanted me to feed back whether or not it had solved the problem. Yes- they wanted the feedback. It sometimes takes a while to pick up from where they left off, but it has been 2 months. It has taken that ong before - My intention was to wait until March and then chase it up.

That is what I intend to do. It has just been a busy week.

[FrankB]

@Oop North John @jfheath
I can add that I have done some testing also to help John. In total 8 tests. These tests show clearly (I think) that the problem exists. Indeed only when the route is created on the XT it behaves correctly. Very surprising to me was that a route initially created on the XT, where it behaves correctly, then transfered to the PC, and then sent unmodified again to the XT does NOT behave correctly.
All the data collected, tracklogs, screenshots etc, have been sent to John together with my summary. Sure hope it helps.

[pcdabbler]

So is this a problem with the latest Garmin route planning software, i.e. Basecamp ? The feature has happened to me twice when touring abroad, including last year, when, due to a road closure in the pouring rain, the GPS went nuts trying to take me back to the closed road and refused to offer me an alternative. In the end I just cancelled the route and drove blindly for around twenty miles, the restarted the route with a single destination point. Totally unacceptable, Garmin should be ashamed of themselves. I wonder if the problem occurs on using a mapsource or MRA created route ?

[jfheath]

No. Its not to do with Basecamp otherwise it would do it with the other Zumos. Frank and I were doing different bu parallel tests and both came to the same conclusions.

Just got to re-start the conversation with garmin if I can. It seems as though someone higher up has blocked the thread. This is the second time it has happened with the same topic, but we have more detail this time. We seem to know exactly what it does and when it happens.

Thanks for the comment and confirmation that you have had the same issue in the same circumstances.

[Stu]

It seems as though someone higher up has blocked the thread. This is the second time it has happened with the same topic

That to me says they aren't interested in fixing it

[Oop North John]

Don't know if anyone else has seen this, in case not, here's a possible way of fixing the repeated u-turn misery for imported routes with a via / way point from BaseCamp:

Today I set off on a route that had a via point, but local advice suggested going a slightly different way so I added a via point into the BaseCamp generated route in the XT by adding a "where to" point before the planned one. Due to a mix up I went off route, but the XT didn't keep on trying to make me do a u-turn

So maybe if we work out our routes in BaseCamp download them to the XT, and then add a way / via point that we already know about, ie it's in as a favourite, then the routing will be happy? I'm reversing the route tomorrow, and am going to add another point after the last waypoint in the Basecamp pre-planned route to see if this works as well.

[jfheath]

I'd be interested in seeing the outcome of that John.

What we have discovered so far is that (seemingly):

A route that is created on the XT (with or without using favourites, works OK. Even if you deviate from the route.
Possibly (not 100% sure) If you stop a route add in your current position and delete everything in front of it, the behaviour is OK.

It seems to be that any route that is 100% calculated by the XT is OK.
Any route that has been pre-calculated, and then modified by the XT may get stuck in repeated U turns.

The use of Waypoints created in MyRouteApp or in Basecamp for the route creation may be a cause. I don't know about Waypoints created in Explore.

One thought occurs to me about this - and what the difference is

When a route point is transferred to the XT it is usually recalculated. (But not Waypoints). It is also slightly relocated. By this, I mean that the XT seems to look up the name of a known place near to the plotted point, and uses the coordinates for that plotted point instead of the ones that it has been given.

When a route point is changed within the XT from a Via to a shaping point, that point gets relocated.

When I create a route I like it to keep the name that I give it - because I like to have numbers to represent a sequence or a mileage. Because the XT will rename them unless they are created as Waypoints, all of my tests have been created with some Waypoints.

I am wondering if my behaviour is actually causing the problem !

I wonder if I just make a route using drag and drop points that the XT renames and slightly repositions, will result in a faultless route.

I haven't had the time to get back to Garmin about any of this since Christmas. Various other priorities have prevented it, and weather has prevented my going out on the bike.

Your test is one that I haven't tried - almost, but not exactly the same. So it will be interesting to find out what you discover.