Acoustic technology has developed apace over the past decade and the science is being studied with more precision than ever before due to the advancement of the measurement tools available.

Unfortunately, the application of the latest acoustic technology is generally well behind the level of electronic technology applied in studio control rooms.

Most of the recording industry seems satisfied to spend large amounts of money on the latest electronic developments only to install them into rooms which have either outdated acoustic designs or have very little acoustic consideration whatsoever.

What constitutes an acoustic monitoring problem in a studio control room? Monitoring problems can manifest themselves in two ways. Either the monitoring simply does not sound right — this can be poor overall balance, poor stereo imaging, the room sounding too dead or too live, ('booming', 'tinny') frequency suck outs, or ringing, or the monitoring sounds very good. However, the mix/recording, sounds very different when listened to in another studio or as the end product in a domestic situation.

This is a particularly dangerous situation to be in as it is so misleading. It is not the job of the monitoring system to flatter. The engineer/producer should be able to evaluate exactly what is going down on to tape.

The monitoring system consists of every element between the console and engineer/producer's ears which does not exist in the console/tape machine chain.

Because of the level of technology applied to electronic elements in this monitoring chain, (and even developments in monitor loudspeaker design) by far the weakest link in this chain is now usually the monitoring environment itself (the acoustics of the control room). Numerous problems can occur further back in the chain (particularly cross covers and monitor loudspeakers), however, they are usually over ridden by acoustic problems present in the room.

What happens to the sound after leaving the monitor loudspeaker? Very basically the wavefronts from the loudspeaker travel into the room until they encounter surfaces in the room. The wavefronts will then reflect from these surfaces.

The amount of energy in each reflected wavefront depends upon the absorption qualities of each surface — the absorption coefficient of virtually any surface is frequency dependent, therefore the reflected wavefront will have modified spectral forms, as well as having less overall energy than the incident wavefront. The wavefronts will carry on reflecting from surface to surface until they have negligible energy left.

At any point in the room space the direct sound from the loudspeaker will combine with the reflected waves. There are obviously path differences (= time delay) between these combining waves and thus phase differences between the corresponding spectral components. This causes the characteristic comb filtering and boosting of certain areas of the frequency spectrum, the degree of which depends on relative amplitude of the reflected components and their relative delay behind the direct sound.

The comb filtering and additive effects can cause the frequency response of a loudspeaker that anechoically measures ±2dB to become ±12dB at only eight feet away from it in a bad control room environment.

The development and decay responses of the soundfield in a control room are dependent on amount, type and position of absorptive and diffractive material in the room.

The decay of a room has been traditionally evaluated by the reverberant time (RT60). This is the time for a signal, once switched off, to fall to 60dB below the initial level. This figure gives no information about the shape of the curve and the individual reflections present. Therefore the RT60 figure is a fairly useless value with which to evaluate the room response. This can be seen for example, by comparing a small bathroom and a large auditorium, the mid band RT60 could easily be the same for both cases, however, the two rooms will sound completely different due to the shape of their respective decay curves.

What does the traditional 'real time' spectral analyser/pink noise measurement actually measure?

'Real time' analysis in this sense is a fallacy because these analysers in fact average all the signals arriving at the measuring microphone over a period of time. This period of time (integration time) has to be of the order of the reverberant time, (in order to read the display and average the random noise input) therefore the analyser is measuring the combined sum of the direct sound plus the reflected components.

The real time analyser display obviously needs very experienced interpretation to evaluate whether any anomalies in the display are due to the actual loudspeaker response or due to the room response and if they are due to the room response, what is actually causing them.

Unfortunately, it is this experienced interpretation that is all to often missing when engineers rush to their graphic equalisers in order to 'flatten' the room response.

Graphic equalisation should only be used to correct the actual frequency response of the monitor loudspeaker, it will not affect the actual room response at all.

It is possible, and unfortunately common, to use graphic equalisation to superficially cover-up problems in the room response, and that it is the consequence of the misinterpretation of the real time analyser data. It is no surprise, for this reason, that many studio owners, engineers and producers, distrust acoustic measurements. A 'flat' frequency curve measured on the real time analyser, using pink noise, does not give the whole story about the room response. The room can still sound awful.