Lidar topography

[Ian Madin]

I am trying to import lidar topography to go with borehole and geologic map data, and am having a problem with very slow visualization. The data was a 50 MB ASC grid (which was dramatically down sampled from the native resolution), and when I view it in a 3D window, simple navigation is so slow as to be unuseable.  Is there some way to speed up the 3D redraw? What is the practical limit on grid size?

[Katy D]

Hello Ian,

Very large grid models may produce out-of-memory errors within the RockPlot2D and RockPlot3D programs. Unfortunately, we cannot define the maximum model dimensions because it's based on a variety of variables (e.g. other programs that are currently loaded, RAM size, virtual memory size, etc.). The solution is to decrease the size of the grid or the node density. You may also wish to try using a Subset of your project to limit data being loaded to the map.

More information about creating grids from Lidar can be found in our online help documentation: https://help.rockware.com/rockworks/WebHelp/grid_lidar.htm

[Mitch]

There are 2 Lidar programs. one creates a grid under Grid|Grid Create and one that creates a Triangle Mesh under Utilities|3D. It sounds like you're doing the first. The question is how many cells there are in your grid. Its a common mistake to set grid dimensions that are too large. While your Lidar data has a very fine resolution, using that resolution will end up with a grid surface that is unwieldy. While there is no fixed limit to the size of a grid it would be good idea to limit the grid size to 300 x 300. You can adjust from there.

 

Let us know what dimensions you're using.

[Alison]

Just to give a little bit more input - I would say that 3D display can handle much larger grids than 2D display without running into memory errors.  I just did some testing and found on my computer that the program got pretty sluggish when displaying a grid with more than 1000x1000 grid cells.  This is more cells that you would want to use when creating a stratigraphy model or a solid model (lithology, I/P/T-data), but would probably be OK for simple 3D display.

 

That being said, I was even able to display a grid with 3000x3000 cells in 3D, but the 3D viewer did not respond quickly when I tried to zoom, rotate, pan, etc.

 

My experience has been that much larger grids can be imported into the program and then used to accomplish tasks that do not involve visualization.  For example, you could use one as input to the Grid|Math|Resample program to create a coarser grid to use during modeling.  You could also use it to determine borehole elevations through the Edit|Adjust Elevations from Grid tool in the Borehole Manager.

 

As both Mitch and Katy mentioned, this is all very dependent on your computer specs, so you may just have to do a bit of experimenting to see how many cells your specific system can handle.

 

 

[Ian Madin]

Thanks for all the input.  It sound like I really misunderstood the grid handling capabilities of the program. I am working with 100 km2 of 1m lidar DEM, so I will clearly have to downsample dramatically and chop things into blocks.  What I really want to do is visualize several hundred boreholes spread out over this area with enough topography behind them to be able to work out relationships. Sounds like I need to use a 30m dem at most.