Audio in games has always been a bit of a cheat. In the earliest days of games, a simple pre-recorded sound was replayed when a gun was shot or step taken. Bit rates went up and the quality increased. Audio could be played in stereo to help isolate location of the sound. Games then added full 5.1 and 7.1 surround sound and binaural sound. They now even take into account the material of object collisions such as metal on concrete vs metal on carpet – but it’s not fully dynamic.

Moving forward, what if we could fully simulate sound in a dynamic virtual location? We could take into account how sound reacts with different materials, walls, and shapes in the environment automatically. The idea is not new – but the compute required was far too high for realtime games.

Vercidium has tried to re-create the idea using modern hardware. What is unique is the idea of using this system for visually impaired people. This isn’t an idea that is limited to virtual worlds, but what if alarms in real life had such visual projection (lasers, lights, etc). Interesting.

There’s also some interesting comments in the video discussion that can lead you down different implementations and ideas.