Back to REGONAUDIO

" The most beautiful sound in the world"--Wolfgang Sawallisch describing the orchestral sound in the Bayreuth Festspielhaus .(personal conversation)

The orchestra in Bayreuth is under an overhang--the audience hears no direct sound at all. Remember this when people start to talk about how musical reproduction is about imaging and transparency. These things are non-existent if the orchestra is under a roof--and yet that Bayreuth sound is "the most beautiful sound in the world".

Everyone who attends concerts of unamplified, acoustical music quickly notices that concert sound is quite different from recorded sound. Part of the reason is space. Concert halls are much larger than living rooms, and you hear the size of the room around you. But there is also a major difference between recorded and live in the character of the musical sound itself. Recorded sound is almost always brighter and more transient emphasized, more aggressive, as it were, than the real thing.

These differences arise primarily from a single, simple fact: The microphones are closer to the instruments than you would be in the audience. Why does this make such a big sonic difference? Why is the sound in audience locations so different from the sound right where the players are? In Issue 38, I treated the subject of why the close-up sound is brighter ("Records and Reality: How Music Sounds in Concert Halls"-also reprinted on this site). There I discussed tonal balance, primarily; my emphasis now is a little different. Let me start at the beginning.

How Sound Works in Concert Halls

Imagine a musical instrument playing a note on stage. The note begins with an "initial transient." Then, with a non-percussion instrument anyway, the note continues with sustained tone for a little while. Typical notes last about a quarter- to a half-second. Really short notes last around a twelfth of a second. (Consecutive notes at a rate of more than 12 per second tend to blur together, turning a scale into a glissando.)

Out in the audience, what do you hear? First, there is the direct arrival of the initial transient. This tells the ear/brain the direction in which the sound source lies, and later, related sounds, such as reflections, seem to come from that same direction, unless they are delayed so long they become an echo. Fairly soon after the initial transient's direct arrival, one hears the early reflections. In a typical concert hall, these begin to arrive in about 15 to 20 milliseconds. These are lumped in with the first arrival by the ear/brain and give an enhanced sense of the sound beginning. Unless they are very strong and asymmetric laterally, they don't change the perceived sense of the directional position of the source. As the note is sustained, one begins to hear tone, from the direct arrival of the sustained sound, from the early reflections, and from the build-up of the reverberation in the hall, the "reverberant field."

Now, the reverberation in a hall takes a long time to decay after the tone stops. In a good concert hall, it takes about 2 seconds for the sound level to drop 60 dB (the "reverberation time"). But the build-up of the reverberant field is much faster. This is a math thing (details omitted!), but the "rise time" of the reverberation, which is defined as the time for the reverb to get within 3 dB of its full value, is only 1/20 of the reverberation (decay) time. So 'if the decay time is the usual 2 seconds, the reverb is coming on strong within 1/10 of a second after the tone started.

The reverberant field is a big part of the sound, too. Anywhere beyond the edge of the stage, it is likely to be more than half the total sound. And out in the hall, where distance has diminished the direct sound, the reverberant part is much more than half of what you hear. Furthermore, the reverberant field has a different character from the direct sound. For one thing, it contains much less high frequency energy. The reverberant sound has bounced around the hall, so it has been traveling through the air a long way. And air absorbs high frequencies much more than it absorbs lower ones. So do most materials, especially the soft flesh of the people in the audience, and the seats, too, unless they are wood. (Even "hard bodies" are soft as far as sound absorbing is concerned). The steady-state frequency response of a hall is always strongly rolled off in the highs. The top octave is pretty much out of there, except in the direct arrival and the early reflections.

There is a second effect that is more subtle, but is perhaps even more important musically: The hall eats transients. Distance attenuates the direct sound, the first arrival of the initial transient. But the reverberation picks up and amplifies the steady, tonal part of the sound. The transient part is not amplified in the same way because it is over long before the 100 milliseconds (=1/10 sec) rise time of the reverberation.

You can think of this in another way: The hall acts on sounds as a kind of time-delayed amplifier. The reverberation, all those later reflections combined, amplify the sounds made on stage. But the process spreads out transients. They contribute their energy to the reverberant field, all right. But that energy is spread in time and therefore is not as "transient" as the transient sound directly from the instruments.

So what happens? The sound in the hall is much less transient-laden, far smoother, far more legato, far less staccato, less "bangy" and "crunchy," than close-up sound. It is also harder to understand the words at a distance. Theaters for speech (plays, etc.) have to have much less reverberation than a concert hall in order for speech to be comprehensible (speech comprehensibility depends on transient definition).

The more you think about this and the more you listen for it in concert environments, the more important and all-encompassing the difference becomes. And if you have a chance to listen to the close-up "main" mikes in a recording session, compared to the far way "ambience" mikes, the situation will be burned into your memory forever. The ambience mikes have the sonic character of a concert. The close-up mikes are far more aggressive, more emphasizing of transients. On piano music, the difference is startlingly - almost incredibly - large.

Acousticians are well aware of these effects. One of the main judgments involved in concert-hall design is the balancing between the transient-enhancing early reflections and the reverberation. Various measures have been introduced for this: R. Thiele's "Deutlicheit," which is the ratio between the energy of the first 50 milliseconds of a transient from on stage to the total energy received at an audience location; Reichardt and associates' "clarity," which is the dB difference between the first 80 milliseconds and afterwards (this difference is usually negative, properly so in a music hall: the later energy is greater).

What This Means for Music

For music, as opposed to speech, where articulation is the vital point, the smoothing out by the hall is intended and desirable, even indispensable. Modern musical instruments ate designed to be played in large halls. The instruments' sounds are tailored for that environment. As halls became larger during the Nineteenth and Twentieth Centuries, instruments were made more and more brilliant to compensate for the high-frequency roll-off in the larger halls. And old instruments were rebuilt for this purpose. (Almost no original-condition old Italian violins still exist. Almost all have been rebuilt to accommodate the higher string tensions of modern metal-wound strings that give increased power and brilliance. And the soft-sounding sweet-toned violins of Jacob Stainer, regarded in bygone centuries as ideal - J .S. Bach played one - have become collectors items, virtually unusable for modern concert performance.)

The changes are particularly easy to follow in the case of pianos, since the pianos of various periods are still around. (I once had the chance to play on a piano used by Chopin, a haunting experience even for a violinist with near zero pianistic ability.) The Erards of Chopin and Liszt are a different sonic world from the modern concert grand, far softer in tonal as well as dynamic terms, and having far less attack. Even an 1890 Pleyel is a far cry from a modern piano. Listening to the old instruments makes one understand Schumann's famous remark about Chopin's piano playing being like an aolian harp, and contemporary descriptions of how in Debussy's playing one could almost not tell when the notes began.

Modern instruments are musically well adapted to their intended environment, the large concert halls of our time. The smoothing and de-brightening effect of the hall turns their bright, aggressive, direct sound into the beautifully balanced, rather smooth sound that is musically appropriate.

How It Can All Go Wrong

The trouble is that if you record a modern piano or a modern orchestra at close range, even, as is often the case, with the microphone inside the piano, what emerges is utterly wrong in musical terms. The smoothing effect of the hall is eliminated. The resulting sound is far too bright and, worse perhaps, far too bangy and aggressive, more like a jackhammer than the Chopin/Schumann aolian harp. The Heifetz recordings by RCA in the early stereo era show how these effects operate on the violin: In real life, in the concert hall, Heifetz was ultra-smooth, almost too smooth for some tastes (Virgil Thomson called it "silk underwear"). On the records, his style seems crunchy and somewhat rough. I keep coming back to this example because having heard Heifetz live -- an unforgettable experience -- I find the misrepresentation on records is particularly disagreeable. But the phenomenon is general. Everyone is abused.

It is no wonder that musicians have found it hard to take audio seriously (Heifetz called high fidelity and stereo "high phooey and hystereo"). But of course this aggressive sound is not an intrinsic property of recording. It is simply a feature of too close microphone placement, egged on by generations of audio and record critics calling for "immediacy," "presence," and "clarity." (An empty hall is somewhat more reverberant than the same hall with an audience present. But this justifies only being slightly closer than in the concert situation, and slightly is the operative word here.) The intended sound of music is the sound in the hall. To the TAS motto, "the sound of real music in real space," we should add, "at a sensible audience location." In our own time, there has been a trend toward a more distant perspective. Let us hope this trend continues.

Back to REGONAUDIO