Hello All
Some wisdom is required please - I am creating large scale terrain model for export to LumenRT to do an animation. I have created an STM in Descartes, exported it to a DTM and then draped a raster file over it as a dcdrape in Microstation Connect. I find this gives me the best quality surface, but the drape is giving me trouble. Firstly, the image does not always export. randomly, the file will export with the image apparently not there. I am not sure what is causing this - it is as though it is not finding the raster file.
I have managed to get around this by creating a custom material in a file where it does work and applying that in a file where it doesn't work. My concern is that the location of the raster image may not be correct when I attach it this way.
Secondly, I am struggling with the resolution of the draped image itself. I have read everything I can find on the forum regarding this, and tried all the lighting tricks and material settings, my result in Lumen is just not up to scratch. The material I am creating is pretty much default, type Special and DCDRAPE and I have turned the specular value right down.
The is the STM looks great in Descartes but fuzzy as soon as I get it into Microstation CE and fuzzier still in LumenRT. Any suggestions as to how I might improve this would be much appreciated.
-W
I came across this post on the WIki, could this be a similar issue?
http://communities.bentley.com/products/3d_imaging_and_point_cloud_software/w/wiki/28678.stm-file-incorrectly-exported-into-lumenrt
Does anyone know if I can apply this patch to LumenRT 201502424?
I've ran into very similar issues trying to get Descartes STM quality draping to LumenRT also.
What is the Bentley suggested / best way to get high resolution draping for visualization? It's my understanding that DCDrape has resolution limitations (1600 x 1600 pixels in SELECTseries, not sure if that is increased in CONNECT). Would we be better off using the MicroStation visualization tools with a Descartes STM & high res drape vs. trying to take it to LumenRT?
Are ContextCapture reality meshes now the expected vehicle for having a high resolution existing condition in LumenRT visualizations?
Wendy,
You have my sympathy, getting a decent draped image into LumenRT is tough. I ended up using the vizualization tools in MicroStation to drape the image and got something that looked OK. I didn't start with a hi-res aerial so I did not notice a drop in resolution on export to Lumen.
Good luck
Tom F
Mike
Subsequent to this post I have added the drape to the STM in Descartes and the resulting LumenRT export is significantly better than from Microstation, although I do believe there is some work to be done here.
On the advice of the clever people at Bentley locally, I exported the terrain as an obj file (where you have some control over the output settings to increase the tolerance) and placed it in LumenRT manually. Results were much improved.
Your point about ContextCapture is well made - am finding increasingly that clients' expectation for a realistic environment is hard to achieve in LumenRT. It is geared towards creating an environment, but I am finding it falls down on importing an existing environment and combining it with the model. There is also a need to incorporate drone footage with the model, I am hoping to see some more case studies using this on the forum. There is one that uses camera matching etc, very clever and well done, but it is hard to achieve this level of skill when visualization is not your bread and butter, but something you may be called upon to produce once or twice a year.