Acoustic sensations in virtual worlds

Virtual Reality applications are characterized by two main features: realtime processing and close integration of humans by novel output and interaction devices. Processing tasks are simulation of world behaviour, computation of world the visual and acoustic representation and interpretation of user actions. Specialized computing nodes distributed in a net offer full processing power: compute servers a la cray for simulation purposes, high performance graphic workstations a la SGI reality engine for fast visual rendering and multimedia workstations a la Next with audio capability for sound rendering. Moreover ther are add-on peripherals as synthesizers, effect generators, speech processing and 3D-audio-cards. To manage and utilize the full processing power of a heterogenous net a distributed system is required. In the virtual reality toolkit of IGD a sound server addresses multimedia workstations as well as MIDI-devices via a common programming interface. The server stores a table of d igitally sampled sounds, which can be manipulated.