Before proceeding, the concept of an octave should be explained. For those unfamiliar with music, an octave is considered the most perfect interval, both because of its musical harmony and the purity of its sound waves. The range from C to C is an octave, from F-sharp to F-sharp is an octave, and so on. From a scientific standpoint, as will be explained shortly, octaves blend easily because the wave patterns of the two notes fit nicely into each other, creating a pleasing sound to the ear. Keeping this in mind may help to understand the reasoning behind some of my decisions.

The first task was to decide how exactly I would define the "best seats." One could listen by ear, but this lacks scientific objectivity and, even then, what would one listen for? I decided that volume must be measured, and after some thinking and discussion with Dr. MacAdam, I concluded that volume consistency across different frequencies, defined here as octaves, would be the deciding factor. Using octaves to determine the frequencies was logical because of the perfect nature of octaves. A note that is an octave higher than another has exactly twice the frequency of the lower note, which would mean the wave patterns of the sound are identical except that the higher note has half the wavelength of the lower. The frequencies used in this study ranged from 65 Hz to 2093 Hz, with each successive note having twice the frequency of the previous one. Similar wave patterns are important in acoustical testing so that there is minimal difference in the reaction of the hall to the sound. Thus, the best seat could be defined as the seat with the most consistent volume across the range of six octaves.

The testing itself required a keyboard, something to measure the sound level, and an assistant . The use of a non-touch-sensitive keyboard was important because it eliminated the human variable of how loud one played at a particular moment. I obtained a seating chart of the Hall and selected twelve evenly spaced seats for measurement (Figure 1). I planned to have someone play the first octave on the keyboard while I sat in the seat and measured the sound level using an Extech Instruments Digital Sound Level Meter (Model 407735). After recording the measurement, I would give a signal to stop playing and move on to the next octave after the previous note had stopped resonating. This would be repeated until all six octaves had been measured, and then I would move to the next seat and start the process over again.

Having decided what to do, I had to obtain permission to use the Concert Hall. I spoke with Ms. Holly Salisbury, Director of UK's

Singletary Center for the Arts, who was enthusiastic and supportive of my endeavor. She personally gave me a short tour and an explanation of some of the specific acoustical tendencies of the Concert Hall, and she provided me a wealth of information about the architect and his philosophies. After I spoke with her, she referred me to Chris Musinski, the Singletary Center Assistant Production Director, to work out the details of the actual experiment. He told me we would be able to lower speakers from the proscenium arch to stage level in order to represent better the projection of a stage performer. He also explained to me some of the tendencies of the hall, including the dead zone of the corner seats in the very back of the Hall. We then agreed on a date for the actual testing.

Once the testing began I was surprised at how fast it went. In an attempt to ensure consistency and accuracy, I held the sound-level meter at ear height each time. However, due to time constraints, I was only able to do two complete measurements. The combined results are shown in Table 1. While the data may seem varied and inconsistent, one should understand that a difference of 3 dB (decibels) is simply a doubling of sound energy, which is perceived as only a small change by the average person. A difference of 10 dB, however, is a ten-fold increase in sound energy, equivalent to more than three successive doublings, and is a quite significant increase in volume. It is also important to realize when looking at the data that the individual values, as recorded in Table 1, are unimportant. The ultimate goal of this experiment was not to record volumes, but to compare the consistency or variations of volume.

The largest difficulty of the project thus far came at this point: what to do with these numbers? After much thought and discussion with Dr. MacAdam, the first and most obvious step was to average the two trials together to come up with a single volume for each seat in each octave. To account for the possibility that my keyboard or the sound system was unusually loud or soft in any octave, I then averaged all the volumes throughout the Hall in a particular octave together, resulting in six different averages. I compared the volume level at each seat against the volume level averaged over all seats for that octave, and this gave me the amount each seat differed from the average for each octave (Table 2).

The final step was to compute the range of the numbers for each seat, with the smallest range signifying the seat with the volumes most consistently equal across all given octaves (Figure 1 and Table 2). Based on the data collected, it would appear there is an arc running diagonally from the back left to the front right of the auditorium, while looking out from the stage, which would best be avoided. At nearly all the seats around this arc, the highest frequency proves to be a problem, appearing as either the upper or lower extreme. Of the twelve seats I chose to test, the results showed that seat E-40 proved to have the most consistent volume across all octaves. To take it one step further, I also computed the range after eliminating the octave with the most extreme difference, which still showed seat E-40 to be the best.