4 ms·
Looks really interesting. Is this feasible to use for acoustic simulation inside buildings?
by ImHereToVote 2y ago
Looks really interesting. Is this feasible to use for acoustic simulation inside buildings?
- magicalhippo 2y agoGoods question. I've been thinking about implementing this[1] method for real-time sound propagation as a hobby project, which relies on this[2] paper. Here's an excerpt from the second one: In this work, we present a technique which relies on an adaptive rectangular decomposition of a 3D scene to enable efficient and accurate simulation of sound propagation in complex virtual environments. It exploits the known analytical solution of the Wave Equation in rectangular domains and is thus able to achieve at least an order of magnitude performance gain compared to a standard Finite Difference implementation, while also being memory-efficient. Consequently, we are able to perform accurate numerical acoustic simulation on large, complex scenes in the kilohertz range which, to the best of our knowledge, has not been previously done on a desktop computer. Note on the performance claims, it's about a decade old. Anyway, was looking for a reference implementation so also curious if OpenCFS could be it. [1]: https://gamma.cs.unc.edu/PrecompWaveSim/ https://gamma.cs.unc.edu/PrecompWaveSim/ [2]: https://gamma.cs.unc.edu/propagation/ https://gamma.cs.unc.edu/propagation/