Notes on the height-map array and error arrays

More here.On the preprocess structure:

I use simple arrays for a 1024x1024 example--
        int lmax;                  // finest level of resolution
        float **h;                 // vertex height
        float **e0,**e1;           // max error in triangle 0/1
Note that there are half as many vertices as triangles.

I just sub-sample the h's, which is by far the simplest thing when you get
to the dynamic triangulations (else what would the hieght be when eight
*different* size triangles meet at a vertex?).  The wedgie thicknesses
can actually be tighter than mentioned in the paper, but then you can
get splits causing *more* error on occasion and also the view-frustum
cull can be flakey because the bounds are not nested.  I use the tightest
possible wedgie bounds computed by exhaustive sampling and just deal
with the flakeyness issues (hacks and kluges, mostly).
 
 

Contact Information

Updated Dec 22, 1999