GPU Based Sound Simulation and Visualization

As the era of Moore´s Law and increasing CPU clock rates nears its stopping point the focus of chip and hardware design has shifted to increasing the number of computation cores present on the chip. This increase can be most clearly seen in the rise of Graphic Processing Units (GPU) where hundreds or thousands of slower cores work in parallel to accomplish tasks. Programming for these chips represents a new set of challenges and concerns. A visualization of sound waves in the room was desired so that phenomena like standing waves could be quickly identified. In order to produce the visualization quickly, some form of acceleration would be required. The GPU was chosen as the accelerator using CUDA to produce the data for the simulation. The prototype was tested on a computer with an Intel Core2 Quad core CPU Q9450 and an NVidia GeForce GTX 480 GPU which contains 15 groupings of 32 compute cores for a total of 480 compute cores and has 1.5 GB of on board memory.