Zumo User Forums

proofresistant wrote: Fri Jul 25, 2025 8:22 pm ...

This previous post was a little bit Updatet yet

Hi @proofresistant

I've downloaded your routes Rat AlteSchule and had a good look at them. Found 6 misplaced route points and corrected them.
I didn't repeat any of your tests - I did some of my own for the things that I want to look for. The results were Ok, and not a great deal different from what I expected. It's probably not quite the tests that you would make, but it's all information.

I'm curious. You said the route was created in Basecamp. Yet the route points have flags and you cannot set the icon for ad hoc route points. Only for Waypoints. Also the route points have all kept their names - which is unusual for a Basecamp produced route.

Yet I corrected done placement errors and recalculated both routes before doing any tests. I wondered what you had done - if anything.

My recalculated routes, like yours, have the subclass 00000FFFFF string, which surprised me. Perhaps I should have changed the maps, recalculated and then changed the maps again. Hang on- I'll have a go at that -...... No. Every single subclass definition that follows a via or shaping point definition has the 'ooooffff' string. WHich I'm sure wasn't always the case - 'cos I had noticed that when that was there, the route point names were never changed.


Anyway, I digress. pdf attached.

@proofresistant

jfheath wrote: Sun Jul 27, 2025 10:38 am I'm curious. You said the route was created in Basecamp. Yet the route points have flags and you cannot set the icon for ad hoc route points.

jfheath wrote: Sun Jul 27, 2025 10:38 am My recalculated routes, like yours, have the subclass 00000FFFFF string, which surprised me.

Maybe, just maybe this is the result of the Postprocess function of TrIpManager?
I have never noticed weird results with CEP when a GPX was postprocessed by TripManager, but if it would, I would like to know.

Frank

I did wonder, but he never mentioned it, (I might have skimmed over that important piece of info) so I just imported them direct to Basecamp, created new gpx files and placed them on the SD card.

But if so, it is odd that even after changing maps and recalculating twice the 'null' subclass field remains intact.

Doubtless he will let me know.

jfheath wrote: Sun Jul 27, 2025 2:22 pm But if so, it is odd that even after changing maps and recalculating twice the 'null' subclass field remains intact.

Once a subclass is reset to '00ff' it remains that way. Only when creating a new Shaping/Via you can be sure the subclass is initialised.

Hi @jfheath,
first of all, it is a pleasure to understand zumo XT2 behaviors together

jfheath wrote: Sun Jul 27, 2025 10:38 am I did some of my own for the things that I want to look for. The results were Ok, and not a great deal different from what I expected.

ProofResistant CEP Test Results.pdf
Shaping Point CEP Test - Location : Test 01
Significant difference. The route takes the motorway to meet the route between shaping points 04 and 05 (red circle)
With Via Points, It took the motorway until it crossed the original route (green circle)

I honestly cannot understand why anyone would expect this result,for the first sections of the route.
This is a significant CEP issue, at least in my opinion. Too many shaping intermediate points are simply being skipped here.
However, I had already recognized that U-turns can be a problem for CEP, as in this case, and described this in my previous summary.

jfheath wrote: Sun Jul 27, 2025 10:38 am
I'm curious. You said the route was created in Basecamp. Yet the route points have flags and you cannot set the icon for ad hoc route points. Only for Waypoints. Also the route points have all kept their names - which is unusual for a Basecamp produced route.

"you cannot set the icon"
That's only half the truth, or rather, it's only partially correct.

• Yes, you can't change them!

• But alternatively:
  • 1st you can copy them from other routes.
  • 2nd you can set flags before in the Map and then use the rubber band to place the route point exactly on such a flag.
    You can see then, the shaping point will also inherit this flag (some times ).

Whether points retain their name or not has not been of interest to me to date, so I have no opinion on the matter.

jfheath wrote: Sun Jul 27, 2025 10:38 am Yet I corrected done placement errors and recalculated both routes before doing any tests. I wondered what you had done - if anything.

When I say that I create in Basecamp, that doesn't mean that I always start from scratch, but rather that I some times copy from my other collections. May thats it.

jfheath wrote: Sun Jul 27, 2025 10:38 am

My recalculated routes, like yours, have the subclass 00000FFFFF string, which surprised me. Perhaps I should have changed the maps, recalculated and then changed the maps again. Hang on- I'll have a go at that -...... No. Every single subclass definition that follows a via or shaping point definition has the 'ooooffff' string. WHich I'm sure wasn't always the case - 'cos I had noticed that when that was there, the route point names were never changed.

I have never really analyzed or understood the subclasses.
I just noticed, at least in the last tests, that there are default values for waypoints, which are set differently depending on the program, such as Tyre and/or Basecamp, etc.
In the example I mentioned here, the data was saved with Basecamp, and I don't think I processed it with TripManager (otherwise the 4-5 poorly placed waypoints would not be as they were found).

jfheath wrote: Sun Jul 27, 2025 10:38 am Anyway, I digress. pdf attached.

Thanks for this


PS
It may not have been useful to rebuild the route and/or place the points “correctly” in order to test it in such better variant.
More important is to try out the specific causes for CEP issues.
The bad waypoints were definitely not the reason for my CEP issue in this test case, it's the U-turn.

FrankB wrote: Sun Jul 27, 2025 2:10 pm I have never noticed weird results with CEP when a GPX was postprocessed by TripManager, but if it would, I would like to know.

May i can find one example.

I mentioned that I am creating another route for simulating CEP issues.
I will then have this optimized by TripManager, but I believe the problem will remain. We'll see, it remains exciting.

I've been trying to sort out what is happening with my recent strange CEP behavior. When I get far enough to share I will update here.
However, while working on that, I realized that CEP was ignoring my avoidance settings - specifically CEP was taking a highway to start the route even though I had the XT2 set to avoid highways.
I tested that independent of CEP and found it to be an issue on its own. I reported that separately viewtopic.php?t=3442

I had not observed the avoidance settings in the CEP, so I have no opinion on this.
What I did observe is that the preferred shortest or fastest route is taken into account when calculating the CEP.

I am currently in the process of creating a test GPX and a guide for it, the merging is actually doing most of the work
Apart from that, my findings remain as recently updated. Is comming soon

So, it's done for now.

I have now also worked out the appropriate demo for the compilation of CEP findings I posted a few days ago. Don't worry, it's only a “small” route with 2 issues included.


Additionally the CEP issue also is documented step by step. You can see how two CEP Issues issues appears and how I was able to fix them myself during the trip with the zumo XT2 as an example. Please take a little time here


Attached is the matching PDF and the matching GPX template.
Content al little bit more like this
Hopefully my experiences will also help you.
May also for the old zumo XT (1) but I can't test that myself.

I would be pleased to receive constructive feedback