Dynamic Terrain from GIS data (like S.I.E.V.E.)
by ENJohnson · in Artist Corner · 04/09/2008 (12:38 pm) · 3 replies
I don't want to reinvent the wheel here, so I was wondering if anyone could point me in the direction of converting TIN, raster, or DEM data to a format that is compatible with Torque. Precise elevation is more of a concern than efficiency, so 8-bit greyscale PNG limited to 256 height deviations is not quite sufficient enough especially since 256x256 provides too small an area. I need something that will work properly with ATLAS. I would also like to preserve data for the geographic coordinates for cross-communication.
If anyone can point me in the right direction, or let me know what middleware components might suffice in aiding me, your help would be greatly appreciated.
Thanks,
Eric
If anyone can point me in the right direction, or let me know what middleware components might suffice in aiding me, your help would be greatly appreciated.
Thanks,
Eric
#2
So, as long as you have some means of converting the elevation data to RAW format (and that should be trivial, too), you should be fine.
You're probably right about Atlas being a better bet than legacy terrain.
As for the geographic coordinates thing, I take it you want to be able to correlate vertex data with their original geographic location. Simply create an extra map with that data and use the floating-point texcoords to index into that data. For elevation-based datasets, that should work fine.
04/23/2008 (3:51 pm)
Atlas' current heightfield-based generators use 16bit raw files which should be fine for your purpose. If you need even more precision, it's trivial to extend the generators.So, as long as you have some means of converting the elevation data to RAW format (and that should be trivial, too), you should be fine.
You're probably right about Atlas being a better bet than legacy terrain.
As for the geographic coordinates thing, I take it you want to be able to correlate vertex data with their original geographic location. Simply create an extra map with that data and use the floating-point texcoords to index into that data. For elevation-based datasets, that should work fine.
Ryan Jaeger