Hello. I refered earlier to two European papers. Now I found this at their site: Patents on sound rendering. We are in the process of filing for two US patents, together with N. Tsingos and E. Gallo on "Using graphics cards for audio signal processing" and "Perceptual techniques for sound spatialisation" I'm sure the first patent is intended to cover all audio processing in GPUs. This is sad. Even if somebody has given a prior thought on that GPUs could be used for audio processing, it does not help if all these patents cover the only way to transfer and process audio in GPU. The second patent deals with sound source clustering. I have given a thought for that kind of clustering earlier than them when I wrote GVerb. The number of early echoes are reduced by grouping the virtual source points in 3D space and by representing one group with one spatialized source point. They specially mention the early echo reverberator as an application of their method. When a person is killed, it means 8-14 years of prison: a life is cheap compared to the intellectual property, which is worth of 20 years. So, I suggest I keep up with my intellectual property and we implement our version of the invention. Your help is needed in coding. Juhana -- http://music.columbia.edu/mailman/listinfo/linux-graphics-dev for developers of open source graphics software