Sabine was founded in 1979 by three engineers who had between them worked on two opera houses and a nuclear power station, and who had noticed that the acoustic advice given to architects was mostly adjectives. The practice has been named after Wallace Clement Sabine from the first day, which sets an obligation: he arrived at his equation by measuring a room several hundred times with an organ pipe and a stopwatch until the numbers stopped disagreeing with him, and a practice trading on his name is not entitled to be vaguer than that.
There are nineteen of us. Eleven are acousticians, four are engineers who came from structures and vibration, two run the measurement equipment full time, and two keep the practice from falling over. We work across concert halls, opera houses, theatres, recording studios, teaching rooms, places of worship, workplaces and transport buildings, and we take on around fourteen commissions a year — a deliberately small number, because the useful part of this work is the arguing, and arguing takes time that cannot be subcontracted.
How we work
We would rather move a wall on a drawing than hang absorption on it afterwards, and we say so at the stage when moving it is still free. Most of what determines how a room will sound is settled in the first eight weeks of a project, in decisions about volume, proportion and section that are made by architects and clients who are usually not thinking about acoustics at that moment. Being in the room then is worth more than any amount of later cleverness, and a practice that is only brought in at Stage 4 is being asked to repair something rather than to design it.
We compute every room twice, by Sabine’s equation and by Eyring’s, and publish the gap between them. In a lightly treated room the two agree and the gap is uninteresting. In a heavily treated one they diverge, and the size of the divergence is an honest measure of how far the statistical assumption has been stretched to produce the answer. Quietly picking whichever figure suits the report is a habit we have seen often enough to have made a rule about.
We measure what we predicted, and we publish the comparison whether or not it flatters us. Every project page on this site carries both the computed curve and the measured one. Where they part company we say why. A discipline that only reports its successes is not measuring anything; it is marketing.
What we will tell you
We will tell you when a room cannot do what is being asked of it, as early as we can work it out. We will tell you when the answer is a loudspeaker rather than a wall, and when it is neither and the real problem is the plant on the roof. We will tell you which of our recommendations we expect to survive value engineering and which we do not, so you can spend the argument where it matters. And when we are overruled — which happens on most projects, because a building is a negotiation — we record what was proposed, what was built, and what the consequence turned out to be, in a document the client keeps. Several of those consequences are described on the project pages of this site.
Equipment
Two class 1 measurement systems, a calibrated dodecahedron source and a subwoofer for low-frequency work, an eight-channel array for spatial measurements, accelerometers and a vibration analyser for structure-borne work, and a small reverberation chamber at the Rope Walk unit for testing sample assemblies when the published coefficient for something does not look right. That last item earns its floor space about four times a year and has twice changed a specification.