Uses and limitations of statistical and geometrical modeling of nonexponential decay curves in performance spaces

Abstract

The conditions under which the high-frequency, diffuse-field model of coupled rooms is a valid approximation to geometrical acoustics have been examined by comparison with ray-tracing predictions of decay curves in two and three room systems. Results particular to coupled rooms were noted. Because nonexponential decay shape is sensitive to both decay rates and relative energy densities, corrections to these parameters meaningfully improved the diffuse-field model. Variations in the free-path distribution of each subroom, introduced by coupling, complicated the use of improved decay models. Also, the expected decrease in energy density with distance from the source was found to result in spatial dependence of decay shape for certain coupling geometries. Insights from this study were used in the construction of diffuse-field and geometrical computer models of Bass Hall, a 2056 seat multipurpose auditorium with an acoustically coupled stage house. Preliminary results indicate that high-frequency decay curves in each of the subrooms predicted by geometrical acoustics are well matched to the predictions of diffuse-field models. The use of both models as prediction and design tools is assessed by comparison at high frequencies with measurements made during occupied and unoccupied conditions. [Research supported by the Bass Foundation.]