Normally, when we test a sound system, it’s in an unoccupied room, mall, stadium, etc. That, however, is not how it normally would be listened to. Many sound-system operators say that the presence of an audience can make a huge difference to the sound, whereas others think the changes are fairly small and hardly worth adjusting for. I can safely say, “It depends.” Primarily, it depends on the difference in the acoustic conditions with and without the audience present. So, what effects can audiences and building occupants have on sound-system performance?
Essentially, there are three main effects:
- Acoustic Changes brought about by the additional sound absorption introduced by the audience.
- Transmission Changes to the direct and reflected sound patterns caused by the audience absorption and associated temperature and humidity changes and gradients.
- Ambient Noise Increases due to normal audience/occupant activity, chatter, and spectator cheering and volatility.
Although all of those traits are well known, little has been written about them, and certainly not from a sound-system perspective.
People, on the whole, are pretty absorbing—from a sound perspective, certainly. Oddly enough, just how much sound people absorb is open to speculation, as it very much depends on the activity/posture of the person. Also, are the people all crowded up together, or are they spread out throughout the space?
Figure 1 shows a collection of absorption data that I have collected over the years. In many cases, it’s impossible to separate out the effect of the people from the seating in an auditorium, because the entire seating area acts in unison with the occupants. In reality, it’s highly dependent on the actual space being measured. In other words, a block of seats in auditorium A will not behave in exactly the same way if it’s measured in auditorium B, and it’ll certainly be completely different when measured in a reverberation test chamber.
However, as can be seen in Figure 1, people absorb more sound at mid and high frequencies than at low frequencies. Typically, people have absorption coefficients of around 0.40 to 0.60, but they absorb more when in a concentrated area in upholstered seats. (It should also be noted that people do absorb sound at 8kHz, but not many references have data for that frequency.)
So, now, we can see, in a reverberant space, why the sudden introduction of a large absorbing area (people), with an absorption coefficient of 0.4 to 0.6, occupying the floor or seats will have a big impact. To put it in perspective, consider one of the stands (side buildings) of a typical European stadium (Figure 2). In this example, when unoccupied, the measured reverberation time (RT) is around 2.7 seconds at mid frequencies (500Hz). But, it is predicted to diminish to 1.8 seconds when fully occupied.
That is a significant change, and it can be expected to have quite an impact on the intelligibility of the PA system, for example—in this case, improving the STI performance from an average of 0.46 to 0.72 STIPA. That is a huge improvement! But, as we will eventually see, it’s not quite the full story; this theoretical increase is rarely, if ever, achievable in practice. One point worth noting: The STI improvement is far greater than might be expected from the change in RT alone. In other words, introducing people into a space is very much more complex than simple reverberation theory might have you believe.
Figure 3 shows the measured RT characteristics for a medium-sized multipurpose hall (1,000 seats), with and without the audience present. It is rare to be able to make such measurements. During this particular test, the room was about 50-percent occupied. For comparison purposes, I have also shown the predicted RT for full occupancy of 1,000 people.
As noted, other acoustic changes often accompany the introduction of people into a space, so reverberation time alone doesn’t tell the whole story. Early reflections, in particular, might be affected. It’s for these reasons that, in critical spaces, such as theaters and concert halls, the absorption of the seating is selected to minimize the change in RT with and without the audience. This enables the sound during rehearsals, for example, to match the actual performance more accurately. Although some differences are inevitable, a stable acoustic is a useful base from which to work. Temperature gradient and humidity effects might well still occur, and they can cause the sound to diffract and be transmitted over audience members quite differently when they are present—particularly when the sound from the loudspeakers has low grazing incidence, such as from relatively low-height stage sidefills.
Next month, we will see how audience and crowd noise can affect the resultant sound-system performance, and how noise and sound absorption have a crucial balancing act.