Do we still need diffuse field theory?

Abstract

More than twenty years after Murray Hodgson’s “When is diffuse field theory applicable?” paper, we gathered more and more evidences that diffuse field is mostly a chimera. If we consider the two most important implications of the diffuse field model, i.e., sound pressure level uniform distribution and reverberation time invariance, it is quite easy to say that, based on actual measurements in a number of different spaces, such conditions are hardly found. Ideal sound diffusion requires ergodic and mixing conditions, which are not obvious to happen, particularly when sound absorption is unevenly distributed or rooms are not proportionate. So, apparently, diffuse field theory could even be dismissed in favour of more accurate approaches capable of taking into account the specific nature of each space. Nowadays, we have several instruments spanning from the many variations of the ray-tracing algorithm to the “brute force” numerical solution of the wave equation. However, such methods rely on the measurement or estimation of other coefficients that, if not properly made, may introduce even bigger inaccuracies. A critical analysis is carried out showing that diffuse field theory still represents an important way to understand sound propagation in enclosed spaces.