Hi Roman, > I played with 3D clouds patch in my sandbox and forgive me if I'm not > familiar with whole flightgear/simgear system. > Now I try modify existing code of 3Dclouds to pointsprite usage. > Roman
I am not sure point sprites are a good solution here... Altough they're put to good use in small-scale particle systems with high vertex counts (e.g. something like Mark Harris' cloud rendering), they're probably not so well suited for this implementation (based on Niniane Wangs' Paper, AFAICT), where clouds are modeled at a somewhat coarser scale. For example, point sprites are clipped like points (at least according to spec), so billboards will pop into/out of view as the center vertex crosses the viewport. Perspective-correct point size scaling is possible by fiddling with the point size attenuation, but still the billboards are constrained to be square and to be aligned with the imaging plane, so we would loose some flexibility in modelling the clouds. I think the current system would be better optimized by using impostors, because performance is probably impacted more by pixel throughput than by geometry (which btw might perform better by using arrays instead of single quad draw calls in CloudShaderGeometry.cxx). Would be interested in your benchmarks anyway - what I said being mostly theoretical :-) Manuel ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
