In 1895, Harvard asked a young physics instructor named Wallace Clement Sabine to fix the acoustics of a recently completed lecture room in the Fogg Art Museum. The room sounded terrible: sound from a single spoken word lingered audibly for over five seconds, making lectures nearly unintelligible. Sabine's investigation lasted three years. It involved borrowing seat cushions from Sanders Theatre across campus, transporting them to the Fogg hall each night, and running measurements from 2am to 6am to avoid the noise of the Harvard Square streetcars. From that obsessive empiricism emerged the equation that defines architectural acoustics to this day.
Reverberation Time and Why It Matters
Reverberation time (RT60) is the number of seconds it takes for sound in a room to decay by 60 dB after the source stops. Sabine chose 60 dB because that roughly corresponds to the difference between a loud sound and the threshold of inaudibility in a quiet room.
The decay rate depends on two things: how large the room is, and how much of the sound energy is absorbed by the surfaces each time a wave reflection occurs. A large room with reflective walls decays slowly. A small, heavily treated room decays quickly. Sabine quantified this relationship with a formula that remains fundamentally accurate 125 years later:
T60 = 0.161 * V / AWhere V is the room volume in cubic meters and A is the total acoustic absorption in sabins. One sabin equals one square meter of a perfectly absorptive surface (an open window, for instance). A surface with an absorption coefficient of 0.5 contributes 0.5 sabins per square meter.
Sabine discovered the formula not by deriving it from physics, but by measuring it. His experiments with different quantities of seat cushions revealed that reverberation time decreased in a predictable way as total absorption increased. The constant 0.161 (or 0.049 in imperial units) fell out of the data.
The Boston Symphony Hall, completed in 1900, was designed using this formula. Sabine calculated the geometry and surface treatment needed to achieve a reverberation time of just over 2 seconds, which he judged ideal for orchestral music. The hall still ranks among the finest listening venues in the world.
What Sabine Got Right, and Where It Breaks Down
The Sabine formula makes one significant assumption: absorption is distributed evenly, and losses per reflection are small. That is a reasonable approximation for lightly treated rooms where the average absorption coefficient across all surfaces is below about 0.2. In that regime, the formula provides a useful approximation of RT60.
In heavily treated rooms, the assumption breaks down. If half the surfaces are covered with dense broadband absorbers, some waves encounter absorptive panels on nearly every reflection. The Sabine formula overestimates the reverberation time by 15 to 30% or more when the average absorption coefficient exceeds 0.4.
Carl Eyring derived a correction in 1930 that handles this case:
T60 = -0.161 * V / (S * ln(1 - α̅))Where S is the total surface area and α̅ is the average absorption coefficient. For small α̅ values, ln(1 - α̅) approaches -α̅ and the Eyring formula collapses to Sabine's. The two formulas agree to within measurement error when α̅ is below 0.2. Above 0.3, Eyring is the correct choice.
Practically, this matters for studio control rooms and treated recording booths. A room lined with thick rockwool panels will have a much shorter measured RT60 than the Sabine formula predicts, and designing treatment based on Sabine alone will lead to under-specifying absorption.
Ideal RT60 Values by Room Type
Reverberation time is not universally desirable or undesirable, it has a target range that depends on the room's function:
Control rooms: 0.2 to 0.4 seconds. A mixing environment needs minimal reverberation so engineers hear the reverb in the recording, not the room. Longer RT60 causes reverberant buildup that obscures transient detail and low-frequency definition. The target is roughly 0.2 seconds across the full band.
Live recording rooms: 0.3 to 0.5 seconds. Some natural ambience contributes life to acoustic recordings, but enough control is needed to capture instruments cleanly.
Concert halls: 1.8 to 2.2 seconds for symphony orchestras. The reverberation is load-bearing for orchestral music. A violin section of 16 players sounds larger and more blended in a room with 2-second decay. A direct-dry recording of the same section played through a speaker sounds thin by comparison.
Opera houses: 1.3 to 1.8 seconds. Opera requires more speech intelligibility for text delivery than pure orchestral music.
Cathedrals: 4 to 8 seconds or longer. Gregorian chant was composed for these acoustics. Modern popular music in cathedrals sounds unintelligible for the same reason.
The trade-off is between spaciousness and intelligibility. Short RT60 clarifies, long RT60 envelops. The optimal value depends entirely on what the room is for.
Beyond RT60: The C80 Clarity Metric
RT60 describes overall decay but says nothing about the balance between early reflections and late reverberation. Clarity (C80) fills that gap:
C80 = 10 * log10(E_early / E_late)Where E_early is the acoustic energy arriving in the first 80 milliseconds after the direct sound, and E_late is everything after 80 ms. C80 is measured in dB.
The 80 ms boundary is not arbitrary. Psychoacoustic research has established that the auditory system integrates early reflections with the direct sound, treating them as part of the same acoustic event. Reflections arriving within 80 ms contribute to perceived loudness, fullness, and ensemble blend. Reflections arriving after 80 ms are heard as reverberation, and at high levels they reduce intelligibility.
Positive C80 values indicate more early energy than late energy, which tends to sound clear and detailed. Negative C80 sounds diffuse and enveloping. For orchestral music, C80 values between -2 and +2 dB are generally considered optimal. A room with very long RT60 will have strongly negative C80; a dry studio will have strongly positive values.
Note that C80 requires actual measurement with an impulse and impulse response analysis. You cannot calculate it from room dimensions alone.
Critical Distance
One more concept the Sabine formula implies: critical distance. This is the distance from a source at which the direct sound and the reverberant sound contribute equally to the total sound pressure level. Inside critical distance, direct sound dominates and the position of the listener relative to the source matters. Beyond critical distance, the reverberant field dominates and listener position barely matters.
Critical distance scales with room absorption. A heavily treated control room (T60 = 0.25 s) has a critical distance of roughly 2 to 3 meters in a mid-size room. A concert hall (T60 = 2 s) has a critical distance close to the stage. This is why studio engineers work in the near field: the monitors are placed within the critical distance to minimize the influence of the room.
Practical Implications for Studio Treatment
Three things follow directly from this physics:
Thin absorption is irrelevant for low frequencies. Porous absorbers (foam, mineral wool) only work when their thickness is comparable to a quarter-wavelength of the target frequency. A 50 mm panel has a quarter-wavelength cutoff around 1.7 kHz. Below that frequency, thin panels are essentially acoustically transparent. Bass treatment requires thick panels, membrane resonators, or corner-mounted broadband traps of 200 mm or more. This is also why the Sabine formula gives misleading results at low frequencies when absorption coefficients are measured only at mid frequencies.
Adding more absorption has diminishing returns. Once average absorption across the room is above 0.4 to 0.5, further treatment has little effect on RT60 because the existing treatment is already capturing most of the energy. The Eyring formula quantifies this. Chasing a true anechoic result in a live room by piling on more foam gets expensive and ineffective; beyond a certain point the room mode problem (which is a different phenomenon) dominates and additional absorption does nothing.
Sabine never solved the problem he was assigned. The Fogg lecture hall acoustics problem was only partially addressed. Sabine reduced the RT60 but the room geometry still created problematic reflections. He died in January 1919, at 50, without having fully resolved the problem that started the work. During World War I he had been assigned to advise on aircraft instruments and served on the Western Front on aerial reconnaissance techniques. His acoustics research stalled during those years and was never completed.
What he gave the world instead was more durable: a formula that made acoustic design quantitative. Before Sabine, a building's acoustics were a matter of intuition and luck. After him, they were engineering.