• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

HELP!! Crash on SPLITGRID

The man deserves more than a rest!! I was thinking perhaps some kind of Nobel Prize for Flight Simulation?
 
Back from my vacation now. Is the file attached to post 9 still the best polygon to try to debug this with?

Concave or convex shouldn't matter for scenproc btw. I'll check what is going on here.
 
Back from my vacation now. Is the file attached to post 9 still the best polygon to try to debug this with?
Concave or convex shouldn't matter for scenproc btw. I'll check what is going on here.

Thanks! That polygon in the attachment in post 9 is one of many. I can provide smaller ones or bigger ones. I'm really giving scenProc and the batch processing a workout with lots of large data, some large files, many features in each file, and it's holding up and running beautifully with the exception of these crashes. When I get them I drill down slowly into the data until I reach a point where only one or two specific polygons which cause the problem are left, all others processing perfectly.

Let me know if I can help in any way!

ian
 
Arno
Here's an (bigger) example of a particularly interesting batch run where a pretty complex polygon is processed without a problem (see log file) but a second isn't.
ian
 

Attachments

Hi,

Ok, I'll check. In general the smaller the data, the quicker I can debug issues.
 
Hi,

I can reproduce the issue here with your data. But I need to have a deeper look to understand why it goes wrong. Hopefully I can fix it later this weekend.
 
Hi,

I think I have fixed the bug now. Can you try the new development release tomorrow? I have tested wit the files you provided and they all seem to work now.
 
Hi Arno,

:( The new release seemed to fix some of the "bad" polygon issues perfectly but not others (or at least I couldn't be certain). I probably have too many types of rogue polygon to fix. It's funny though that I can process some amazingly complex polygons (large, multi-part, convex, concave, with holes) and other apparently simpler ones fail.

I'm still working through to see if there's an obvious difference between specific polygons. I haven't had a chance to test enough yet though - it's been taking me all day. While I've been doing this it seems that importing files is taking longer. For a couple of areas it failed to IMPORT (or hung with no progress display) where this didn't happen previously. I'm probably just imagining.

Also, I had a couple of runs where scenProc processed the IMPORTOGR, SPLITGRID, and then IMPORTAGN for about 5,000 features in total but EXPORTED 90,000 agn files! It's possible there were around 90,000 files in the original folder in total but I never imagined it would IMPORT all of them. On deleting the target folder files and running again it created what seems to be the right number (a few thousand). Perhaps I'm working with simply too much data.

Rather than throw all these crazy problems at you, I think it best to leave it and I'll process as much as I can and then "fix" individual polygons manually in Annotator if I can.

Thanks for your help though and especially thanks for scenProc in the first place!

ian
 
While I've been doing this it seems that importing files is taking longer. For a couple of areas it failed to IMPORT (or hung with no progress display) where this didn't happen previously. I'm probably just imagining.
Arno - do you have a link for the immediately prior development release? I've run it multiple times on data that loaded without any problem previously. This one is definitely not working for me - either loading exceedingly slowly or not at all (hangs with no progress or only part progress).
Thanks
ian
 
Hi,

You can find all old release here:

http://scenerydesign.org/old-releases/scenproc/

Do you still get the same kind of crashes? I have also seen that on some polygons the code seems to enter some endless loop and just hang. It is related to the code the does boolean operations to get rid of holes.

I'm trying to figure out what goes wrong, but that is not always easy. The problem usually happens when you load a lot of data and it is hard then to isolate the polygon that has the problem.
 
I agree about isolating problem polygons!! I divide up my data into ever-diminishing squares covering a smaller and smaller area polygons (successful or failed - the latter being further divided and processed) until I get the specific one or two that appear to break the code and not run. Sometimes it is hard to see why because they appear to be simpler and even smaller than some that have previously been processed.
For me the attached file illustrates the point. I can explain if necessary. Basically I've taken the geometry/geopgraphy down as low as possible to identify polygons that appear to "break" the code.
ian
 

Attachments

Arno - downloaded the release but I'm getting

upload_2014-4-21_8-6-8.png


What am I doing wrong? I copied all the files across from the zip to a new folder in Program Files.

Sorry to be such a trouble-maker... :oops:
 
Hi,

Did you download the complete version or only the update version? The update version is supposed to be installed over an existing installation (so that you have all GDAL and OGR DLL files as well).
 
Arno
Just checked the sub-folders and all the dll files seem to be in place. I'll have another go at it.\
ian
 
Hi,

Did you download the complete version or only the update version? The update version is supposed to be installed over an existing installation (so that you have all GDAL and OGR DLL files as well).
Definitely got the full release version from the ftp... :confused:
 
So you have the subfolders gdal-data and x64 in your installation folder as well? The message indicates that some required files are missing. I added this check to prevent issues later on.
 
So you have the subfolders gdal-data and x64 in your installation folder as well? The message indicates that some required files are missing. I added this check to prevent issues later on.
Yes - got them both Arno.

upload_2014-4-21_8-42-2.png


Each with containing files...

upload_2014-4-21_8-43-1.png


and

upload_2014-4-21_8-43-58.png

upload_2014-4-21_8-44-38.png


Any missing? There's also the Tools folder.

ian
 
Back
Top