作者: Stefan Bilbao , Craig Webb
DOI:
关键词:
摘要: Physical modeling sound synthesis for systems in 3D is a computationally intensive undertaking; the number of degrees freedom very large, even and spaces modest physical dimensions. The recent emergence into mainstream highly parallel multicore hardware, such as general purpose graphical processing units (GPGPUs) has opened an avenue approach to reasonable amount time, without severe model simplification. In this context, new progra mming algorithm design considerations appear, especially ease with which given may be parallelized. To end finite difference time domain methods operating over reg ular grids are explored, regard interesting non-tr ivial test problem, that timpani drum. chosen here because its sounding mechanism relies on coupling between 2D resonator acoustic space (an internal cavity). It also large dimensions, thus simulation high computational cost. A presented, followed by brief presentation methods, foll owed discussion parallelization GPGPU, results.