Chapter 17

The Live-End, Dead-End Approach*

An earlier studio control room design philosophy than that of the Non-Environment room is the so-called Live-End, Dead-End (LEDE). As with the Non-Environment approach, this divides the control room into areas of contrasting acoustic function in an attempt to provide a listening environment that makes the room’s own acoustics subservient to those of the monitored programme.

The quote attributed to Jack Wrightson¹ in the account of the Non-Environment room (Chapter 16) provides an equally valid reference point for the LEDE philosophy:

The problem in the context of studio monitoring is that, regardless of the conditions, the room/monitor-loudspeaker combination places its imprint on all that transpires. For this reason, a control room should be neutral, it should add as few sonic colourations as possible to the sound generated by the monitor loudspeakers. In this context, poorly designed loudspeakers should exhibit their flaws; well-designed loudspeakers should demonstrate their assets. The aural purpose of a control room is to provide the best possible free-air representation of the signals carried by the studio’s audio system.

The aspiration for an LEDE room is thus to create a neutral monitoring environment. Where this approach differs from the Non-Environment approach is in the definition of neutral. The aim for the room is to create an environment in which the ‘context’ information is coming from the monitored programme material, not the room in which you are listening, while retaining certain aspects of the room’s natural acoustic. The things that make an environment  non-neutral are, principally, the early reflexions (which give a feeling of warmth or timbre) and the build-up of unconnected sound that combines with direct sound to create the impression that there is a tonal imbalance.

17.1 First Impressions

Any room lends some form of context to the listening experience. This remains true even if the principal objective is to avoid giving it any colouration – this is true by definition since it is a different experience to listening outdoors in a field. If you do nothing other than provide the listener’s eyes with a room to look at, this translates into an expectation of how the room should sound that is either fulfilled, or not, by the listening experience.

The way in which the ear is able to distinguish between, for example, a wooden drawing room and a stone basement is through the nature of the room’s early reflexions – their frequency balance and how closely spaced they are in time – and the way the reflexions are changed by the surfaces they encounter. A good impression of the size of a room is obtained as you walk into it, through the first sounds that you hear. From experience you have a model in you mind of how large the room is and how it ‘sounds’.

17.2 A Window of Objectivity

If a reflexion-free zone is produced so there are no reflexions in the desired 6–30 ms period (in a practical sized room, this is around 20 ms) it is possible to eliminate the usual colouration of a room acoustic and listen instead to that coming from the loudspeakers, whether it has been created in a real acoustic environment or with a digital reverberation unit.

Consequently, the design objective is to create a reflexion-free zone whereby the sound travels directly to the listener’s ears, and then there is as long a gap as feasible (within the constraints imposed by normal room sizes) before any reflected sound arrives at the listening position. When those reflexions do arrive, they should be as diffuse as possible – that is, there should be as many as possible and they should be unrelated to each other. Depending on the size of the room and the available budget, this is achieved by mounting the loudspeakers in the monitor wall (again, in common with the Non-Environment room). The geometry and broadband absorption at the monitoring end of the room are used to eliminate early reflexions at the listening position; and by returning energy to the room in a non-destructive manner. The use of absorption in the front halves of the rooms gives rise to the ‘dead’ element of the LEDE description.

We are trying to model an idealised energy-time response within the room (see Figure 17.1). The intensity of the direct sound and the early reflexions can be calculated, and their relative levels managed through acoustic treatment of the listening room. Through this, the overriding reverberation characteristic will be that of the monitored system, as the colouration offered by early reflexions will only be those created in a studio’s performing room or those coming from sound processors in the recording system. Achieving this involves consideration of where the sound energy enters the room space; how  it travels to the monitoring person’s ears; and how the room reacts with the energy introduced into it to give the listening context.

sss

Figure 17.1 Idealised energy/time response for a ‘Live-End, Dead-End’ control room. (a) Maximising the gap between the direct sound and the onset of ‘reverberation’ (or clustered late reflexions) by suppressing the early reflexions. (b) The minimum amount of reflexion suppression required for a reflexion-free zone

The first order improvement in the operation of a loudspeaker is to mount it in a wall such that it works into half space; there is no fringe or edge effect produced by the loudspeaker, and there are no early reflexions from the monitor wall or anything in close proximity.

Early reflexions are destructive (or colouring) to the sound in that they are coherent with the direct sound, and that due to differing path lengths they may arrive out of phase with the direct sound. Because they are coherent and delayed, they interfere with a range of frequencies at multiples of their fundamental frequency, giving comb-filtering effects. This causes aberrations in the perceived frequency response of the loudspeaker. By suppressing early reflexions within the room, the nature of the sound heard at the listening position is the nature of the sound coming from the loudspeaker, not any characteristic of the room.

It is necessary to retain subsequent reflexions, however, because a control room is finite in size (and generally small in acoustic terms) and complete absorption of all sound energy across broad frequency bands is not possible. It is, therefore, important that the returned energy from any reflexions that do occur is even with frequency. Thus mid and high frequency energy must be returned into the room to match the inevitable bass energy build up.

Careful design of the room will avoid modal problems as far as is possible, but the isolation requirements of real studio buildings necessarily imply bass level energy retention within the room (using massive walls to prevent egress of sound into unwanted areas – the performing room(s) even if nowhere else). Thus diffusion is employed in the rear half of the room to return energy in as even and non-coherent a form as possible.

If energy is returned to the room unmanaged, it is likely to cause similar effects as with early reflexions. Managing this energy is achieved by most designers by using proprietary diffusion products. These work by setting up interference between the arriving and reflected sound by the use of differing lengths of ‘pockets’ or ‘wells’ (see Figure 4.32) giving different delay times. This is similar to the way a film of oil on water produces different coloured reflexions, as internal reflexions in the oil film interfere with the light. A diffuser produces multiple, unrelated (termed non-specular) reflexions of sound that are returned to the room in different directions to the arriving sound, but retain their energy and frequency content. The energy is therefore retained in the room at the original frequency but without the ability to recombine destructively.

It can be shown that the intensity of direct sound decreases as the square of the distance from the source. The intensity of early reflexions can be similarly calculated, taking into account the additional path lengths and the characteristics of the reflecting surfaces, and can be shown to be inversely proportional to the square of the distance (and hence also time) from the source.² Absorption effects are highly frequency dependent but may be calculated by the use of the absorption coefficient for a given surface (see Tables 4.1 and 4.2, for example).

The Non-Environment approach to design would have it that the room contributes little or nothing to the monitoring mix. The aim is that all effects of the room are removed by very efficient, very broadband absorption throughout. Adherents to the LEDE philosophy have felt that it is very difficult to achieve highly uniform absorption of a wide range of frequencies. The bass end in particular is very difficult. If it were possible to allow all the bass energy to leak away, it would be more practical. However, a control room is frequently sited next to a performance space, and therefore it is necessary to contain the bass in order to make a normal studio operable; thus, a residual amount of bass remains in the room and builds to create perceived tonal anomalies.

Sound can be diffused by a number of means, and a number of different factors can determine effective diffusion. Given the space, diffusion can be maintained down to the order of 100 Hz. An additional benefit here is that there is a reasonable amount of bass absorption offered by the nature of the diffusers used at higher frequencies, in that the trapped air acts as elastic absorption.

The secondary effect of the diffuse high-end and mid sound energy that remains in the room is that it creates an even decay, producing a more neutral room. Hence the ‘live’ element in the LEDE design.

Taken as a whole, the frequency response of the reverberant field is shaped by broadband absorption, some bass absorption (of a broadband nature to maintain a smooth frequency response), and the maintenance in the room of some diffuse high and mid frequencies. The reverberation time of the space can therefore be made even with frequency.

Room modes can, and should, be minimised, though they cannot be completely eradicated. Modal effects can also be made more tolerable by arranging the principal listening areas to be free of modes. Any asymmetry in the room can make it difficult to predict modal behaviour, although there is a school of thought that suggests asymmetry in the front end of a room is a good thing because an irregular modal response is less perceptible than a regular one.

The net result is an early reflexion-free period in which the timbre and acoustic qualities of the recording environment can be monitored, and there is a context against which to evaluate this in that the monitoring space is even in reverberation time. This has additional benefits of broadening the area over which the stereo field can be maintained. It is optimised at the main listening position for numerous reasons – principally that the listener is stereo-symmetrical with the loudspeakers, which is ideal, and stereo-symmetrical with the room, so that any artefacts are even on both sides. Any phase information is also symmetrical. Figure 17.2 shows a practical realisation of the LEDE concept.

Figure 17.2 Rear end of a ‘Live End-Dead End’ room showing the diffusers. BBC, Maida Vale

17.3 Working and Listening Environments

The modern control room, however, is no longer a place where one technician in a brown coat makes hits out of the efforts of distant musicians working in a different environment. In many of the world’s music studios, 95% of the creativity now takes place in the control room, and is rarely the work of a single person – necessitating the creation of a suitably sized listening area with a common listening characteristic. Everyone involved needs a similar, common experience of what is being recorded.

Modern control rooms are built to impose huge amounts of acoustic control, and their acoustics have to be robust to the way in which they are used. This is important as significant amounts of additional equipment are often brought into the acoustic space, and this can lead to significant livening of the rear of the control room. In the LEDE philosophy, a number of semi-randomly distributed ‘workstations’ of equipment distributed around the rear third of the room introduce further diffuse sound to a diffuse field, rather than detracting from a designed-dead acoustic. It is less destructive to perturb further a diffuse energy field than to introduce reflexions into an area that is intended to be dead – and it has been found in certain studios that there is a limit to the amount of additional equipment that can be brought in before the monitoring environment is compromised. (This will be discussed in much more detail in Chapter 18.)

The practicality is that it is necessary to accommodate this because a lot of recordings are made this way. The operational considerations have to include the control room offering a working environment as well as purely a monitoring environment. Typically, this may involve creating drum patterns, creating MIDI sequences, manipulating samples of audio and layering keyboard tracks in modern music.

There is a further, less well-defined, criterion: the idea of a ‘comfortable’ working environment. This is not analogous to working in a domestic room, but rather a room in which to assess realistically the elements going into the creative process with respect to the variety of listening environments that exist outside of the studio. These range from living rooms and headphones, to cars and nightclubs, to optimised listening rooms and single-speaker radios. In common with other design philosophies, the objective is to allow you to evaluate how your work is likely to sound when it leaves the studio environment and enters the ears of the ‘real world’.

It is folly to try to create a ‘representative’ monitoring room because there is such a broad spectrum of listening environments and situations to address. Instead, what is required is to produce a non-colouring monitoring environment to assist in assessing the relationship of the work to these other environments.

It is important to note the difference between the listening and monitoring processes here. It is quite possible to listen to a sound, whether it is a familiar voice or a symphony, in a coloured environment and recognise the elements that characterise it. This is possible as the brain de-convolves the sound, in order to distinguish between the direct sound and the artefacts added by the environment. As such, it is possible to ‘listen’ in a wide variety of acoustic environments without being aware of either the effects of those environments or the systems through which one is listening. Through this mechanism, sounds become readily recognisable in a wide variety of different listening conditions. ‘Monitoring’, however, is not consistently possible in many such environments as these artefacts may conceal aspects of the programme or modify them in ways that make objectivity impossible.

This again is why there should be no early reflexions – even of a controlled and diffuse nature – in a monitoring room. In contrast, this is quite acceptable in a listening room if it makes the listening experience more enjoyable.

17.4 Summary

The object of the design approach is to provide a neutral monitoring environment.

The principal difference to the Non-Environment approach is in the definition of how neutral ‘neutral’ should be.

The first design objective is to create a reflexion-free zone (RFZ) around the monitors and the listener(s).

The rear half of the room is designed to be diffusively reflective in such a way that the overall room response has an even decay rate at all frequencies of interest.

This is usually achieved by the installation of proprietary diffusers based on specific mathematical sequences. See also Section 4.7.

Room modes should be minimised as far as practicably possible.

The technique is also said to be beneficial in providing a better musical ambience for musicians to work in. This is considered to be important now that so much of the playing of the instruments takes place in control rooms.

The installation of considerable amounts of equipment in the rear section of a room is considered less noticeable in its effect on the diffusely reverberant acoustics of a Live-End, Dead-End control room than on the acoustics of a room with a maximally absorbent rear wall.

Modern exponents of the Live-End, Dead-End control rooms aim for a comfortable working ambience, and not a domestically representative decay time.

* This chapter was written by Tim Goodyer, and was based on a series of conversations with the studio designer David Bell, an experienced exponent of Live-End, Dead-End control room techniques. Many hundreds of control rooms have been built to the specifications of this concept over a period of around 25 years, so this book would have been seriously lacking if an authoritative description of LEDE principles had not been included.

References

1  Wrightson, Jack, ‘Psychoacoustic Considerations in the Design of Studio Control Rooms’, Journal of the Audio Engineering Society, Vol. 34, No. 10, pp. 789–95 (1986)

2  Angus, James A. S., ‘The Effects of Specular Versus Diffuse Reflections on the Frequency Response at the Listener’, Journal of the Audio Engineering Society, Vol. 49, pp. 125–33 (2001)

Bibliography

D’Antonio, Peter and Konnert, John H., ‘The RFZ/RPG Approach to Control Room Monitoring’, presented at the 76th Convention of the Audio Engineering Society, New York, USA, Preprint No. 2157 (October 1984)

Davis, Don and Davis, Carolyn, Sound System Engineering, 2nd Edn, Chapter 9, Focal Press, Boston, USA, Oxford, UK (1997) Davis, Don and Davis, Chips, ‘The LEDE Concept for the Control of Acoustic and Psychoacoustic Parameters in Recording Control Rooms’, Journal of the Audio Engineering Society, Vol. 28, No. 9, pp. 585–95 (September 1980)

Davis, Chips and Meeks, Glenn E., ‘History and Development of the LEDE Control Room Concept’, presented at the 72nd Convention of the Audio Engineering Society, Anaheim, USA, Preprint No. 1954 (October 1982)

Wrightson, Jack and Berger, Russ, ‘Influence of Rear-Wall Reflection Patterns in Live-End, Dead-End Recording Studio Control Rooms’, Journal of the Audio Engineering Society, Vol. 34, No. 10, pp. 796–803 (October 1986)

See also Section 13.1.3.