David Megginson writes:
> The new scenery code is still rough, and some tiles fail to build at
> all, but I am extremely impressed with Curt's recent work on TerraGear
> combined with the better Canadian elevation data available through the
> SRTM.
David,
Thanks for the kind words. Yes there is a few rough edges (so to
speak.) I see at least one bug in tile edge matching has crept in
along the way somehow. I need to look into that, but overall I'm
becoming really happy with the SRTM data and the new terrain fitting
approach which Norman Vine pointed me towards.
Here are some more pictures taken in and around the Bay area:
http://www.flightgear.org/Gallery/Source/terrain1.jpg
http://www.flightgear.org/Gallery/Source/terrain2.jpg
http://www.flightgear.org/Gallery/Source/terrain3.jpg
http://www.flightgear.org/Gallery/Source/terrain4.jpg
http://www.flightgear.org/Gallery/Source/terrain5.jpg
http://www.flightgear.org/Gallery/Source/terrain6.jpg
Here are a few things I've noticed with the new terrain fitting
algorithm:
- The concentration of elevation points in the final result seems a
lot better balanced.
- Major features such as ridges and valleys are much better
represented.
- It seems like we are achieving much better detail with the same or
similar number of polygons.
I'm also pretty happy with the quality of the SRTM data. If/when 3 or
1 arcsec terrain data is released for the entire word, I'll need a 1
gazillion terrabyte HD to do all the processing and a 256 node super
computer cluster also wouldn't hurt. :-)
Regards,
Curt.
--
Curtis Olson IVLab / HumanFIRST Program FlightGear Project
Twin Cities [EMAIL PROTECTED] [EMAIL PROTECTED]
Minnesota http://www.menet.umn.edu/~curt http://www.flightgear.org
_______________________________________________
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
