In recent years it has become pracical to predict coverage patterns of loudspeaker arrays using computer simulations. Comparisons of low-frequency array performance with computer analysis reveal three interesting truths: (1) Some favorite array geometries perform better in practice than they should. (2) Conventional calculations of polar patterns are inadequate to predict the performance of large arrays. (3) Computer optimization of array geometry can yield substantially better performance than what we have come to accept.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.