Kim,
it does sound difficult to implement as a general case. Actually, I think that the only way to meet all current techniques with as common a scenegraph structure as possible is to render the heightfield to texture (generated somehow, e.g. FFT on CPU, sum of sines, FFT on GPU), then use vertex texture fetch to draw the vertices (e.g. projective, tiled, etc.). The main drawback (IMHO) is current performance : vertex texture fetch is slow on many cards; even fewer seem to support GL_LINEAR in hardware (which you need). Undoubtedly this will get better, but when I looked at this a little while ago, it was sluggish in comparison to techniques not involving VTF. To be honest I've don't think I've seen a real time > ocean simulation with geometric wake formations, they usually just use > texture overlays. I agree; vertex deformation by wakes is... uncommon (we've done it in the past, but for very very application specific reasons that probably wouldn't be appropriate). However applying a deformation to your _normal_ field works easily, and works especially well at high altitudes in calm seas when you can't / shouldn't be able to see the vertices anyway. It's equivalent in construction to the way the surface wake foam textures work. One other point : for semi-infinite tiling oceans, have a look at the new instancing stuff that I recently saw contributed into OSG. I haven't played with yet it myself, but just repeating the base (high res) tile as an instance is probably higher performance than having to bother with LOD, skirts, and all the rest. Best regards, David
_______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
