Sign in to follow this  
Guest Stormshadow

Mesh problems

Recommended Posts

Hello ppl,I've taken up mesh again after sometime and started off by converting an experimental 1024x1024 pixel home-made test image containing 256 grey-levels (0-255) as horzintal blinds and a 140-grey level circular area towards the top of the image.I converted to DEM using grises50.Everything ok - problems start arising as soon as I thread on the .inf file!Here is an example: LOD = auto DestDir = "..." DestBaseFileName = "Test" UseSourceDimensions = 1 ; These are the bounds of my mesh ;NorthLat = 035.903320312 ;SouthLat = 035.859375000 ;EastLon = 014.531250000 ;WestLon = 014.472656250 [source]Type = ElevS16LSBSourceDir = "..."SourceFile = "HBlinds.dem"CellType = S16LSBLat = 35.9033203125Lon = 14.472656250NumOfCellsPerLine = 1024NumOfLines = 1024CellXdimensionDeg = 0.000057220458984375CellYdimensionDeg = 0.0000429153447265625ScaleinMeters = 1.0So here's the troubleshooting:==============================- With these settings I get a bgl, and when viewed in TMFViewer, I get only part of the mesh. That is, the Northern and Western bounds are OK, but the mesh does not span through to the Southern and Eastern bounds! Circular region is displayed perfectly. BGL doesn't show up in FS.- If I change the 'NumOfCellPerLine' and 'NumOfLines' to 1025 (as should be the case - ImageWidth + 1, ImageHeight + 1, I get the circular region distorted to an elliptical area and position very close to the Western bound. However, now, the area is bounded perfectly.- I am caluclating the 'CellXdimensionDeg' like this: ImageWidth/(WestLon - EastLon) and 'CellXdimensionDeg' as ImageHeight/(NorthLat - SouthLat).Can someone kindly point me to where I'm messing up!? Cause it's all plain and clear and I'm getting what I'm expecting.Thanks a great deal,Storm

Share this post


Link to post
Share on other sites
Help AVSIM continue to serve you!
Please donate today!

Hi Storm.Your total span is so small that "auto" tries to make a mesh at LOD13. You could change "auto" to "14", and the mesh should compile including more data... of course it won't display in the sim at all, as LOD10 or 11 is about our limit without changing CFG files. But TMFviewer shows it just fine.If Steve Greenwood sees the post, he might be able to explain the FS9's limits for mesh LOD.The source data span needs to be slightly larger than the destination. And it must be large enough to use a span in degrees that will satisfy the LOD size wanted, as explained in the SDKs.Unless you have very accurate data, an LOD of 8 is probably more than enough for the sim ( and that includes the 3 arc second SRTM data ). It makes a real nice bumpy mesh.Anybody can make an LOD11 bumpy mesh. Few people can make an accurate LOD11 bumpy mesh, because we are not privileged to accurate elevational data. If someone makes mesh from poor data at LOD9 or 10 or 11, then that will cause problems when good mesh is made available at those LODs.If we restrict our attempts to LOD7 or LOD8, there should be few problems for future mesh-makers when they make better mesh.I'll get off the soapbox, now. ;)Dick

Share this post


Link to post
Share on other sites

...and the soap resumes!Thanks for the help - I managed to get some trials at LOD11 and everything seems to be ok.Just one thing which I noticed. Mesh close to the shore in most cases fails to appear. I had previously modified the coast for this region and obviously set water parts to 0m in LWM. Thinking that this might be the root cause, I changed the water height to mesh clinging, and voila, mesh close to the shore (note, where there is land!!), appears. Its as if the water LWM takes up part of the land LWM aswell...this is getting mind boggling!!CheersStorm

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this