Page 15 - Winter 2011

High frequency brilliance versus harshness
The high end of the frequency spec- trum deserves greater attention as well. Reverberation times tend to fall off at high frequencies in any room because the air absorbs high frequency sound more efficiently than low frequency sound. This may be a reason why room acousti- cians have tended to neglect high fre- quency information above 4 kHz or so. Despite this, high frequency sound plays a vital role in some of the more subtle qualities that distinguish good concert halls from great concert halls. While a “brilliant” sound quality is desirable, an over-abundance of high frequency sound can result in harshness.
Figure 4 illustrates portions of the spectrum of a violin that relate to subjec-
13
Fig. 3. Reverberation Times (T30) measured and calculated by Kirkegaard Associates in Royal Festival Hall, London. Error bars indicate the standard deviation of the measurements, which were made with an orches- tra seated on the stage and full audience.
It is interest- ing to note here that the ‘F’ region, which extends above 4 kHz, and therefore beyond the reach of standard room acoustics parameters, is related to a sharp or harsh tone qual- ity. The regions ‘D’ and ‘E’ are associated with brilliance. Perhaps a new high frequency room acoustics parameter that somehow relates the relative levels of these portions of the spectrum could be developed to distinguish between bril- liance and harshness in a concert hall? We suggest that the degree of attenuation between the ‘E’ and ‘F’ portions of the
tive qualities of violin tone.
spectrum may be significant in addressing this question. Another aspect of harshness we have observed in our consulting practice is the tendency of a room to sound harsh as the dynamic level is increased. An orchestra playing at a pianissimo or even mezzo-piano dynamic may sound lush and beautiful, but increasingly harsh and strident at forte and for- tissimo dynamics. While this effect begins with the instru- ments themselves, it would be useful to explore the role that
Fig. 4. Plot of a violin sound spectrum13 (With permission of A. Buen).
sound build-up curves, could be considered as alternative methods of analyzing the level and time behavior of low fre- quency sound.
In fact, the bass ratio, a traditional measure of bass response, has fallen out of favor as a predictive parameter for bass response in concert halls. Most researchers appear to be more focused now on the overall strength of low frequency energy, rather than its duration. This leads to parameters like G80(125)11 instead of a ratio of reverberation times, which probably is a move in the right direction, but still does not take into account energy below 100 Hz. A recent study of venues for amplified music in Denmark has emphasized the importance of reverberation times in the 63 Hz octave band,
in that case as a measure of suitability for rock music per-
12
formance.
In Search of a New Paradigm 11