CONCERT HALL ACOUSTICS - 2008: LEO BERANEK IDENTIFIES EDT AS METRIC BEST CORRELATED WITH SUBJECTIVE PREFERENCE

摘要

In the early 1980's the author was hired to review an existing recital hall identified as "too reverberant." Reverberation time was measured to be 1.7 seconds, unoccupied. However, while listening to talkers, musicians, and impulses within the recital hall, the reverberance sounded like a 2.5 second space. This experience began a long period of inquisition by the author into the perception of reverberance and its relationship to the acoustic design of music spaces. Later in the decade, the author's consulting firm while working with G. Siebein and S. Berkow, created a 1/10th scale physical model for the 1,600-seat multi-use hall at the California Center for the Arts, then in the design phases. Using a spark as a sound source and other proper physical modeling methods, the resulting impulse response was convolved with symphonic music recordings created in an anechoic environment (the Jupiter Symphony by Mozart.) Due to computer processing limitations of the day, the convolution was limited to approximately one-quarter of a second room decay combined with the anechoic music. Hence, technically, the reverberation time associated with the auralization was 0.25 seconds. However, when listening to the auralization, the reverberance of the hall was clearly audible as the music was ongoing. This listening experience perpetuated the author's belief that reverberation time is a secondary consideration regarding the enjoyment of reverberance and the audibility of reverberance can be created by sound reflections arriving much more quickly in time than typical reverberation time criteria would suggest.