Saturday, November 21, 2009

The acoustics of concert halls and my living room

Music and Brain Blog #1
Brian Graiser

Jourdain, Robert, Music, the Brain, and Ecstasy, HarperCollins Publishers, NY, 1997. Chapter 2: …to tone…, subsection “Concert Halls,” pgs. 46-51.

Synopsis: Often overlooked in importance by audiences, the concert hall acts as an acoustic resonator for the instruments played within, and could therefore be considered an extension of those instruments. Soloists and ensemble conductors must thoughtfully adapt to a particular hall’s resonance, temperature, humidity, and other factors in order to produce the best possible sound. The nature of unrestricted sound is to spread out in all directions and dissipate quickly; by controlling the path of the sound through surface materials and positions, concert halls make it possible for sound to travel all the way to the back of the audience without losing too much intensity. There are three kinds of sound that concert hall acousticians are concerned with: direct sound, which travels directly from the performers to the audience; early sound, which is comprised of the initial sound reflections from the walls and ceiling reaching the audience; and reverberation, which is composed of the subsequent echoes and sound reflections heard after the early sound. A venue is considered “intimate” when the first reflections of the early sound reach the audience no later than 0.02 seconds after the direct sound; this quality is impossible to produce in a larger concert hall due to the wide space and vast walls and ceilings, so it is typically reserved for solo or chamber concert venues that are smaller in nature. Conversely, while reverberation is a rare occurrence in nature or smaller venues, its prominence in large concert halls was a key factor in the development and evolution of large ensembles such as symphony orchestras; the longer resonance led to a new concept of harmonic blending and progression that was inconceivable in a drier, “intimate” room. Acousticians consider a sound’s reverberation to have stopped once it has dwindled to one millionth of its original strength, and as such can be used as an ideal measuring tool: it is desirable to achieve one second of reverberation for chamber music, 1.5 seconds for early Classical or minimalist large ensemble music, and as much as 2.25 seconds for lush Romantic symphonies. However, the majority of music listening today occurs at home, not in a concert hall. The smaller room delivers a high degree of intimacy, but the reverberation is warped; the numerous objects and surfaces in the room produce a stronger-than-usual intensity in the reverberation, but also cause it to disappear within 0.5 seconds. To solve this problem, most recording studios add in synthetic reverberation to their products; however, these false reverberations are compounded when introduced to an irregular venue (such as a living room), further distorting the sound from its original form and quality.

Reflection: The physics of acoustics and resonance has always fascinated me; as a percussionist, I deal with the acoustic properties of marimba resonators and timpani bowls on a daily basis, so I really don’t have much of a choice in the matter. However, one thing that has really gained my interest is the science behind why certain halls make certain percussion instruments sound much better than others. While Walter Hall is a great venue, it can also lend a dryness to some ensembles, and a lushness to others. Perhaps this is a product of sound directionality and performer positioning; many concert halls have a spatial threshold, where performers beyond a certain point of the stage do not get “picked up” by all the reflective surfaces of the hall. Another thing to note is that unlike most wind instruments or vocalists, most percussion instruments project their sound straight up (excluding instruments such as bass drums, which are purposefully angled towards the audience); the sound from a marimba or vibraphone bar travels down into the resonator tube, hits the cap at the end, and is reflected back up towards the ceiling (inexperienced personnel do not know this, and as result many microphones are erroneously placed underneath the bars of the instrument, rather than above them).

Another thing that piqued my curiosity was Jourdain’s condemnation of recording techniques aimed at the home market. I wouldn’t go so far as to say that living rooms “ruin” music recordings, but it’s obvious that the sound quality experienced in a concert hall seat is much different than that experienced on my couch. In terms of mass marketability, adding in false reverb in the editing phase of a recording makes sense; it gives a vague sensation of being in a larger room, and adds a degree of resonance that would not naturally occur in the average living room. However, for the true audiophiles out there, isn’t there a better option?

I had a thought (a potentially expensive one, but worthwhile in my opinion): what if recording engineers placed more microphones in various locations of the concert hall during a recording session, and focused less on capturing the immediate, intimate sound that occurs on the stage itself? Beyond the standard “ambient sound” microphones, I would be interested in seeing what would happen if an array of two dozen or so microphones were spread throughout the hall (particularly in the “sweet spot” of the audience, which is typically the middle of the mezzanine level), and only a handful of microphones were actually placed on the stage with the musicians. From there, studio engineers could mix the levels to create the desired balance, but they would have many more sources of early sound and reverberation in relation to the usual abundance of direct sound. Transfer those properties to a 5.1 surround sound speaker system, with the rear speakers assigned to only producing the appropriate reflections of early sound and reverberation, and I believe one can create a much richer and more realistic listening experience.


Linnea said...

While reading this chapter of Jourdain’s book, I was stimulated to reconsider the role of acoustics of a performance space in the performance. While I had thought of them to some extent and considered superficially the implications of a particularly “dry” or “resonant” hall, I had never thought of them in great detail.
Last year, I took a new music course in which we listened to Alvin Lucier’s “I am sitting in a room.” As far as I understand and remember, this work explores the sound of the room by playing a recording of a voice speaking “I am sitting in a room…” The recording is then played back and rerecorded such that more of the sound of the room is picked up. This repetition is continued until the words are far from clear. Ultimately what is being generated is the sound produced by a combination of the speakers and the acoustic properties of the room in which it is recorded. I found this to be a brilliant composition that really shifts the attention to the acoustic properties of a performance space.

joe schacher said...

One thing that is missing from this discussion is that whenever you record a sound SOME aspect of the room acoustics is also recorded. So if you record a saxophone, the sound that is recorded includes direct sound, early reflection and reverb. Then, when simulated reverb is applied in the studio, all three sounds (direct, early and reverb) are subject to the applied effect. Certainly, this is not ideal and makes the simulation less effective.

How can a recording engineer alleviate this problem? The answer is simple: record in a venue where the performers sound best and record both direct sound and ambience. If the acoustic space works well enough with the ensemble, simulated reverb is unnecessary. When I record drums, I never record in a music studio. I always rent a church or another venue which has excellent acoustics for drum recording.

As for Jourdain's condemnation of recordists who record music for living room listening: I don't understand his logic. Yes, people should go to the concert and listen if the want the real experience. But if you are making a recording for sale, wouldn't you want your recording to sound best on the most common system on the market? It may not be 'pure', but if most people listen in their living room on $200 systems, a recording engineer should try to make the best album possible for THAT system.


Your idea of using many microphones to capture the full ambience of a hall is a good one. The only problem is that for each microphone you add to the equation, you get more problems with phase relationships. 24 microphones would each accurately record the sound from their perspective, but when you mix them together in a studio, you get either missing or overly accentuated frequencies.

To achieve your goal, you would need a hall which was specifically built for such a purpose. I know that the Glen Gould Studio at the CBC has quite a few mics to capture ambience and that they built it specifically for recording. I think I remember about 12 mics spaced around the hall. Check it out if you can - it is the most amazing recording studio I have seen.