I aspire to find both realistic and efficient ways to achieve spatial audio in complex, dynamic environments. Rather than ray-tracing toward a listener, I simulate sound propagation on the GPU with a 2D wave equation (compute shaders, 512×512), recording reflections into a texture. From that data I drive a full DSP chain — interaural timing difference, inter-channel level panning, pinnae shadowing, distance attenuation, per-band air absorption via FFT, and Schroeder reverb — to localize sound and simulate occlusion, diffraction, and diffusion around obstacles.