Good morning Community,
sometimes we've got problems with triangulation in InRoads in our company, it throws the message "Triangulation halted. ..." and I know that it's a known issue descriped in http://communities.bentley.com/products/road___site_design/w/road_and_site_design__wiki/triangulation-halted.aspx. But unfortunately we're not able to solve the issue as it's descriped in the article. The problem occurs irregular and there seems to be no visible problem with our geometries in the DesignFile.
Some testdata you'll find attached. I uploaded a DesignFile which is the source of the DTM I attached as well. Furthermore I added some screenshots of my DTM import options and the error message itself (for both configurations listed below).
What I tried so far is altering tolerances in InRoads Project Options, copying a part of the DTM that's corrupted into a new surface, importing the geometries in different ways (selection, by level or fence), using the extended data check option for triangulation and using the Resolve Crossing Segments feature. But all in all it was not successful, the error still occurs.
In my current version of InRoads the error message points to the location where the issue occurs, but in later versions it just says "No point found" and that's it - without any hints - that's really unsatisfying.
I tried to test in the following two setups:
Configuration 1: Intel Xeon CPU 5160 Microsoft Windows XP SP3 Microstation SS2 08.11.07.443, InRoads 08.08.00.46, Regional setting: Germany (but with . as decimal separator) all files are stored locally
Configuration 2: Intel Core i7 3770 Microsoft Windows 7 SP1 Microstation SS3 08.11.09.357 InRoads SS2 08.11.07.566 Regional setting: Germany (but with . as decimal separator) all files are stored locallyI hope someone may provide a solution for that issue, since we need the DTM in our business process.
Thank you in advance,
Maik Schulze
Something is wrong with your dgn data on Level 61! What is the source of this data? Are this contour lines?
Regards
Frank
since 1985: GIS, CAD, Engineering (Civil) Senior Consultant : [Autodesk Civil 3D , Esri ArcGIS, VertiGIS: in previous days : Bentley MS V4 - V8i, GeoGraphics, Bentley Map V8i, InRoads, HHK Geograf, IBr DAVID] : Dev: [C, C++, .NET, Java, SQL, FORTRAN, UML][direct quote by: http://en.wikipedia.org/wiki/Helmut_Schmidt]: "Wer Kritik übel nimmt, hat etwas zu verbergen"Wer Grammatik- und/oder Rechtschreibfehler findet, der darf sie behalten :-)
Yes, there's something wrong with geometries in the dgn. We are using triangulation and the Crossing Breaklines feature for quality improvement of our dgn data, means to detect invalid geometry.
But in the past we didn't have such a problem and geometry quality sometimes was much lower than in this case here.
Have you been able to triangulate the surface?
Maik
edit: Right now I realized that when I import level by level with Load From: Level option, triangulation succeeds. When I import with Fence oder Single Element option it still fails.
Unknown said:Have you been able to triangulate the surface?
No Maik, even with IR 8.11. If I import the breaklines (lines and Linestrings) from Level 61 the Problem "Point not found" accur!
I even did a DGN REPAIR and checked some Linestring-Errors (vertices with even values) but I did not find somthing strange.
I suggest you call Bentley InRoads Support Group here in Germany or try the new SS3 open Roads DTM functions.
Maik,
I'm running MicroStation 8.11.09.357 with InRoads 8.11.07.615 on a Windows 7 box. I had issues with your files initally but I then:
I made a copy of your file, deleted ALL graphics from it and then compressed it with ALL options turned on. I used the file as a seed file to create a new DGN file. I referenced in your DGN file, copied the graphics into my DGN file and then detached the reference file. I imported the graphics into the TEST.DTM surface by level. I assigned the default symbology and breakline feature type to the line work and the random symbology and random feature type to the points on level 63. I had no issues triangulating the TEST surface. I used Bentley's out of the box XIN file as I'm in an english environment. I hope I attached the files correctly.
Paul
One other issue is that some of the line work appears to be contour lines. These are the light and dark blue lines to the left side. There are a few other light blue lines throughout the file that do not appear to be contour lines. Contour lines should be imported as feature type contour to improve triangulation. I noticed this when I went back to look at the generated triangles and I notice that around the blue line work on the left, I had triangle verticies that were NOT coincident with feature points. There error is very small but it is there.