In 2008, Paul Malpas reported on a Meyer Sound Constellation system installed in the auditorium at the Universidad Laboral on the outskirts of Gijón in northern Spain. This article takes the broader view and looks over the relatively few products brought to market over the last 40 years, which electronically extend the acoustic usability of a space.
As mentioned above, it has been over 40 years since the first electro-acoustic reverberation system was installed, by Prof. Peter Parkin in the Royal Festival Hall in London. This pioneering approach evolved into a system licensed by AIRO called Assisted Resonance (AR). The principle used an array of microphones, each paired to a specific loudspeaker located remotely from the microphone. Each microphone/loudspeaker pair was tuned to respond to a narrow band of frequencies by use of physical resonators attached to each microphone. Each was tuned to a different frequency band and, between the ensemble, a wide range of low and lower-mid frequency modes of the room were exaggerated, thereby extending the acoustic reverberation in those bands.
A development of this principle at the time was called the Multi-Channel Reverberation (MCR) system, marketed by Philips. The important difference was that the microphone/loudspeaker pairs were full-range instead of band-limited. This brought home the issue of feedback and colouration and a limitation on the channel gain of each microphone/loudspeaker pairing. Therefore, in order to achieve a useful and realistic lift in reverberation levels, around 50 to 100 channels are needed in a typical auditorium. MCR was originally developed in the 1960s and then again in the 1980s but news of MCR was mute for many years until XLNT Advanced Technologies / Team Projects of the Netherlands picked up the technology from Philips and have applied it in at least three medium-sized theatres since 2006.
Such systems as AR and MCR are classified as regenerative, because of how they extend the reverberation time by repeatedly relaying sound picked up in one part of the room to another.
In the mid 1990s, a system called CARMEN emerged from France's Centre Scientifique et Technique du Bâtiment (CSTB) that seemed to make an ideology out of this approach! Regenerative systems are also known as non-in line systems, to distinguish them from in-line systems. Examples of in-line systems are described below. Whereas the principled approach of regenerative systems is to pick up acoustic signals from one part of the room and to relay them via loudspeakers in a way that mimics physical acoustic reflections, the modus operandi of the in-line system is to add the reverberation electronically to a relatively ‘dry’ signal picked up from close and/or directional microphones.
Examples using in-line methods included the Early Reflected Energy System (ERES) and the Reverberation on Demand System (RODS), the latter being developed between Jaffe Acoustics in the US and the late Peter Barnett in the UK. RODS used gating on the microphone inputs to concentrate the efforts of the reverberators on the direct sound and not on adding reverb to reverb. More recently, in-line systems commercially available in Europe are dominated by SIAP, ACS, LARES and Meyer’s Constellation.
The System for Improved Acoustic Performance (SIAP) takes the approach of picking up the sound through Supercardioid microphones, to get as ‘clean’ a direct signal as possible, but relaying this with only subtle reverberation added. This approach seems to rely on accentuating the natural acoustic response of the space by amplification of the source through loudspeakers remote from the audience. SIAP also includes some time-variance to apply to the signal to help control regenerative colouration.
The Acoustic Control System (ACS) applies the principle of using sparse microphone arrays to ‘capture’ acoustic wavefronts, which it would process, redistribute and reconstruct into virtual reflected wavefronts through sparse loudspeaker arrays. Latterly, the system has concentrated more on the application of simulated reflection sequences.
The Lexicon Acoustic Reinforcement and Enhancement System (LARES) uses a small number of directional microphones close to the sound sources over the stage area and a large number of loudspeakers distributed around the space. Between the two is a matrix of reverberators that also employ time variance as a method of limiting regenerative colouration.
Meyer’s Constellation system (previously VRAS) is based on a hybrid method proposed by Mark Poletti. Essentially a non-in line system that also adds reverberation electronically and can control early reflection patterns, all microphones feed into one or more reverberation and/or early reflections processors, the multiple outputs from which are distributed to loudspeakers throughout the auditorium. Constellation controls regenerative colouration by using a proprietary method to ensure linear transfer functions between all microphones and loudspeakers. These functions will include precise equalisation to pull out peaks in the spectral level response. Similarly, the function includes what are referred to as ‘unitary’ digital reverberators – essentially a mathematically pure form of reverberation tail – intended to add frequency independent reverberation or, if required, a smoothly controlled reverberation frequency response. This all conspires to maximise the ‘gain before feedback’, and thus helpfully reduces the number of discrete channels required. Furthermore, banks of independently-seeded reverberators are used to decouple the channels from each other acoustically, reducing the risk of the sort of ‘spread localisation’ you get when a monophonic signal is played through a number of loudspeaker simultaneously.
Sustaining the Argument
In the regenerative vs in-line systems argument, proponents of the former, such as CSTB, argue that any attempt to add electronic reverberation to a signal that inevitably would already contain an acoustic response of the actual space results in the unnatural aural impression of two acoustic responses interwoven. Of course, if the microphones are somewhat directional and close to the source, and if the natural acoustic of the space is relatively dry, then perhaps this issue is outweighed by the sheer power and flexibility of an system based on electronic reverberation and (signal) time delay patterns. The idea that a bank of Digital Signal Processors can impose complex acoustic reflection sequences and precision controlled reverberation tails is one that becomes increasingly seductive as the affordability and sheer power of available DSP continues to grow. Taken to its logical conclusion, the in-line system approach allows us to create a highly tailored acoustic response, perhaps one gathered electronically from a completely different venue and applied automatically, computationally accounting for the install venue, the installed system and the desired virtual acoustic.
Then again, classical music audiences are well known for being highly discerning and quite sceptical about technological interventions, often with good cause. It is very easy to anticipate venues and applications where the puritanical approach of the regenerative camp might not just be esoterically ‘nice to have’ but actually quite necessary to the subjective success and critical acceptability of the result. For my money, the more plausible ‘augmented’ acoustic has been where the new reverberation time has been no greater than double the natural RT (though in recent years I have started to consider relaxing this rule). In other words, to get up to, say, 2.2s we ideally should start with a space with at least 1.1s to start with, in which case the ‘double reverberation’ criticism levelled at the in-line systems applies.
Tailing Off
So the balance of benefits to be weighed between these two broad approaches is by no means a done deal. As processing muscle-power and software know-how continues to evolve, you might think that the ‘fix it in the mix’ possibilities of in-line would nail it every time. But then consider the hard-to-convince market, as well as a maturing of expectations for aural quality in the broader range of concert-goers. Or perhaps one of the hybrid solutions will have the scope to find the right balance for your project, which then brings the success very much down to the care and attention in the commissioning. In any case, these are not systems that can be easily AB tested in your hall and neither are they yet used so commonly in projects that it is likely you will have heard half a dozen this month. As is nearly always the way in good audio, the choices will come down to just how much you trust the people behind (and in front of) the particular solution route you decide to head down, and how much your goals and needs are aligned with their aspirations for your project.