http://www.angelfire.com/music2/davidbundler/acoustics.htmlI saw an article by contributing editor Jerry Rosen entitled "Shaping the Sound
of Concert Halls" in the August 1986 issue of Mechanical Engineering, the
monthly magazine of the American Society of Mechanical Engineers. It was a good
primer on the subject of concert hall acoustics and fueled my curiosity about
some deeper technical issues that the author didn't cover. I decided to do an
acoustics article myself and called up Ron McKay, who at the time was with Bolt,
Beranek & Newman, one of the preeminent acoustical consulting firms in the world
(he eventually left BBN and started up his own consulting firm, McKay Conant
Brook). I visited him at his BBN office and laboratory in Canoga Park,
California. After an hour, I came away with a clearer education and fuel enough
for a story that ran in the November 16, 1986 issue of the Pasadena Star-News
and the following one, which appeared in the August 1996 issue of 20th-Century
Music.
Anyone who has read the mass of vague, confusing, sometimes contradictory
descriptions of concert hall acoustics knows just how much nonsense has been
written on the subject. Look at the hand-waving report cards and hand-wringing
eulogies of Davies Hall in San Francisco, Philadelphia's Academy of Music, or
Carnegie Hall ~ and you'll soon long for the security and crisp language of home
electronics with all its talk about flat response, oversampling and
sound-pressure level in dB.
And why not? To most of us, acoustics looks like a black art. We can抰 see it. We
can抰 touch it. Language often fails us when it comes to describing the vagaries
of sonic textures. Add to that the fact that some of the greatest sounding halls
date from the late 19th century ~ when builders had virtually no practical
scientific knowledge of acoustical principles ~ and it would seem that our
century has been backsliding from the dark ages.
Fortunately, there does exist a standard of excellence for concert hall
acoustics, and experts and lay listeners alike have no trouble coming to an
agreement on it. Ask anyone who knows, and the list of favored sites will
include Vienna抯 Grosser Musikvereinssall (built in 1870), Leipzig抯 Gewandhaus
(1885), the Concertgebouw in Amsterdam (1895) and Symphony Hall in Boston
(1900).
Each of these halls is steeped in a wealth of historical traditions. They each
can boast of a great resident orchestra. All are built from the standard
construction materials of their day such as wood and masonry and, perhaps more
to the point, their interiors share roughly the same aspect and proportions. All
were built at around the turn of the last century, a time when engineering and
architectural techniques were still in their infancy compared with the last half
of the twentieth century. Yet for their acoustics and the sheer quality of the
listening experience, each of these structures far surpasses concert halls of
more recent vintage. Why is it, then, that these halls sound so wonderful? What
secrets did their builders uncover only to be lost again ~ like the secrets of
the Stradivarius ~ to later generations of architects. Finding out the answers
to these questions has been a decades long journey that may finally have brought
us full circle to where we were a hundred years ago.
The search for perfect acoustics was on well before the beginning of this
century. One of the earliest pioneers in this field was Harvard University抯
Wallace Sabine whose studies of reverberation and attenuation continue to
influence acousticians to this day. For the first half of this century,
designers concerned themselves primarily with reverberance and the goal of
achieving an even distribution of sound throughout the hall in accordance with
principles first laid out by Sabine. Then in the early 1960抯, Leo Beranek
discovered the importance of time delays in the arrival of successive waves of
sound. But it was not until the 1970抯 that acousticians fully came to appreciate
why listeners preferred the old European halls best.
Manfred Schroeder, a researcher at AT&T Bell Labs in Murray Hill, New Jersey,
wanted to know what qualities set the best halls apart from lesser ones. To that
end, he studied 20 European concert halls, surveying listener reactions and
correlating them to the physical and acoustic properties of the halls. What he
found was that listeners preferred the sound of long, narrow halls ~ like the
classic halls in Vienna, Leipzig and Amsterdam ~ over the sound of newer, wider
halls built for larger audiences.
The reasons for the perceived differences between narrow and wide halls are not
hard to understand. In either locale, the first sound a listener hears is that
of the sound waves coming directly from the stage. The next wave to arrive is
the sound reflected from the nearest surface. In a wider hall, this is usually
the ceiling overhead which produces a similar signal at both ears. In a narrower
hall, though, the first reflections arrive from the left and right walls. These
reflected sounds produce slightly different signals at the left and right ears ~
both because of their differing content and because they arrive at the ears at
minutely spaced times. Though the differences are subtle, those signals are
apparently dissimilar enough for the brain to perceive an enhanced
搒patial?quality to the musical signal in a narrower hall, thus giving a listener
the impression of being more fully immersed in sound.
The superior acoustical quality of the old European auditoriums didn抰 happen by
design. It was an accident of history. In the early days of construction, before
the advent of reinforced concrete and cold-rolled steel girders, the length of
beam spans was limited by the strength and stiffness properties of timber. These
antique halls had to be built narrow in order to carry roof loads without
resorting to the use of obstructive columns inside the hall. As construction
technology improved ~ resulting in both more accurate engineering analysis and
better construction materials ~ roof spans got wider. This allowed auditoriums
to hold more people and permitted more of those people to sit closer to the
stage. On the face of it, better construction technology had struck a blow for
democracy, opening up the concert experience to an ever larger audience.
However, as the shape of the listening space slowly evolved, there was a cost in
terms of lost acoustic quality - and democratically enough, everyone paid the
price.
In the last 20 years, the recognized importance of lateral reflections has
fueled nostalgia for the proportions of the 揷lassical?European halls. During
that time, the virtues of the rectangular 搒hoebox?design have become lore among
architectural acousticians. Many recent designs have incorporated it as their
basic shape; e.g., Meyerson Hall in Dallas and the Cerritos (California) Center
for the Performing Arts. However, some designer think the case for lateral
refections has been overstated. Ron McKay, who designed the acoustics for the
Ambassador Auditorium (Pasadena, California) and was consultant on the
acoustical renovation of UCLA抯 Royce Hall a decade ago, doesn抰 discount the role
that lateral reflections play but says it doesn抰 matter where the first
reflections come from as long as they reach the listener抯 ears quickly. [At the
time this article first appeared, both of these halls were out of commission.
Ambassador was closed for inadequate funding by its owner, the World Wide Church
of God; the hall remains closed today. Royce Hall was damaged in the 1994
Northridge Earthquake and was not reopened until 1998.]
揂ctually, both quick reflections and side reflections are important,?says McKay,
揵ut a quick reflection from overhead also is important and valuable. The time
delay between the direct sound reaching you from the stage and the first
reflection ~ whether that reflection is from overhead or from the side ~ is very
important. If that time delay is short, then you say there抯 a great deal of
presence to the room, or there is an immediacy to the sound. And all that抯
important there is the time delay question, not whether the reflection is from
the side or from above.?
To illustrate this from a designer抯 standpoint, imagine an auditorium in which
there is a piano recitalist on stage and a solitary listener out in the
audience. When the pianist strikes a key, the sound of that note radiates out in
all directions. One of these sound paths ~ the shortest one ~ leads directly
from the piano to the listener抯 ear. Another wave ~ the first reflected sound
the listener hears ~ bounces from piano to wall to ear like the trajectory of a
billiard ball and arrives a fraction of a second later. What Leo Beranek
discovered in the 1960抯 was that the ideal time delay for that reflected wave is
about 0.02 to 0.03 sec. after the direct sound hits. That delay time, multiplied
by the speed of sound in air (1100 ft./sec.), says that this detour to the wall
should add an extra 22 to 33 feet to the wave抯 path. Clearly, this has
implications for the ideal sizing of a hall (not to mention the ideal place for
a concertgoer to sit). For a listener, the immediacy of sound is enhanced when
nearest reflecting surfaces are close by. Too long a delay between the first two
waves, say 0.05 to 0.07 sec., and the effect is akin to sitting in a cavernous
space like London抯 Royal Albert Hall (not a great place for a concert, McKay
says, 揵ut it抎 be a great place to play basketball!?. Any delay above a tenth of
a second would actually be perceived as an echo.
Thereafter, says McKay, it's important for the listener to continue receiving
many reflections at very short time intervals ~ and from all directions ~ in
order to give the sound more fullness. As an example, he cites his redesign of
Royce Hall. The sound that critics reported as having a tremendous presence or
immediacy was the result, he says, of successive reflections that get to you
quickly ~ "that don't take a long time to go off to distant surfaces and come
back."
As the waves continue bouncing around inside the auditorium "box," they lose
energy with each successive reflection. The sound dies down either rapidly or
slowly, depending on the absorptive properties of the wall and ceiling
materials. Therein lies the second important parameter of acoustic design ~ one
which can be measured in seconds rather than milliseconds. Reverberation was a
quantity first studied around the turn of the century by Wallace Sabine, the
Harvard acoustician who designed Boston Symphony Hall. Technically,
reverberation is defined as the time required for a sound to decay by 60
decibels after the source stops. In plainer language, it's simply the time it
takes a loud sound to decay away to inaudibility.
The ideal reverberation time depends upon the type of music being played.
Liturgical music involving either an organ or a chorus sounds best with a
reverberation period of 3 or 4 seconds. Romantic music ~ say, late Beethoven or
Brahms ~ sounds best in an acoustic of 2 seconds. Debussy and more contemporary
fare that involves complex harmonies is better at 1.6 seconds, an interval that
is also well-suited to chamber music. The difference may seem trivial, but
according to acousticians, anyone with a good pair of ears can hear the
difference between an auditorium with reverberation times of 1.8 and 1.9
seconds. In the theater, where the spoken word makes the highest demands on
clarity, reverberation should be 1 second or less. For this reason, it's not
uncommon to hear complaints about muddy acoustics and garbled dialogue when a
stage play is mounted in a theater whose primary function is musical rather than
dramatic.
All of this raises an important question about the flexibility of multiple use
facilities. One of the classic problems in architectural acoustics is church
design, where the extreme reverberation requirements of liturgical music and the
spoken word come into play; the standard solution here is to design the space
for music and to rely on electronic amplification for sermons. Opera, being part
speech and part music, demands a reverberation time somewhere between the
2-second ideal for orchestra and 1 second for speech (possibly 1.5 to 1.7
seconds, leaning a little more to the musical end of the spectrum). In general,
though, auditoriums have often been called upon to serve many different
functions with as many unique acoustical requirements.
In the 1960s, architects met that challenge by building the American classical,
community, multi-purpose auditorium that was expected to do everything. Drama,
touring Broadway musicals, local symphony concerts, even the opera ~ all were
supposed to be presented under a single roof on a different night of the week.
The unfortunate result was that this building was a compromise hall. A little
too reverberant for speech and not reverberant enough for music, it didn't serve
any function particularly well. For that reason, subsequent designs in the late
1970s and 1980s came to incorporate variable acoustics.
While there are numerous ways to vary the acoustics, one of the primary tricks
is to move very large quantities of draperies in and out of the auditorium.
Draperies, being sound absorbing, cut down on reverberation time and allow
greater clarity for theater and lectures. At Ambassador Auditorium, for example,
the interior has a false ceiling made of bronze bars. Above the bars ~ and
invisible to the audience ~ is an enormous open space. For speaking events, a
large quantity of black curtains may be pulled out into the space (in speech
mode, the reverberation time is about a second). For musical events, the
curtains are drawn back into large storage pockets to open up a big, hard,
reverberant space for music (which then resonates for a more ideal period of 1.6
to 1.7 seconds).
Rock music, jazz and other popular forms ~ which are typically amplified ~
require a relatively dead, non-reverberant space. It helps if the interior space
is large (though the choice of an arena's size for a rock concert has more to do
with gate than with acoustic quality). If the sound energy at an amplified pop
concert isn't dissipated quickly enough, the environment can get too mushy and
overloud. Ideally, the quality of the listening experience at pop concerts has
more to do with the audio system than with the building's natural acoustics.
Given the proportional and volumetric limitations inherent in the requirements
for a good hall, there are ideal sizes for halls depending on the type of music
being played. Acoustic designers find it more difficult to create good symphony
acoustics in auditoriums seating 3,000 or more than, say, halls with capacities
of 1,800 to 2,500. Accordingly, the next generation of orchestra halls are being
built leaner. The newly opened Bridgewater Hall in Manchester, England;
Seattle's Benaroya Hall (opening in 1998); and Los Angeles' proposed Disney Hall
all seat about 2,500 people.
The interior shape of an auditorium plays a key role as well, and designers have
experimented with almost everything imaginable, including fan-shaped, reverse
fan, shoebox, and circular rooms. A fan-shaped hall is undesirable. When a room
fans out from the stage, the sound energy dissipates as it travels to the back
rows; as the room gets wider and wider, the energy is spread thinner and
thinner. In a rectangular shoebox hall, by contrast, the energy doesn't have the
opportunity to spread from front to back; consequently, there is greater
consistency of listening condition and greater uniformity of loudness.
Reverse fans ~ in which the auditorium is widest at the front and narrows toward
the back ~ are poor for both acoustics and sight lines (where a listener sitting
in the front row corner has to watch a performance sideways). In the reverse
fan, the side walls are set far apart so that it takes the reflections much
longer to come back, thus violating the short time delay criterion for good
acoustics.
"The goal that everybody's after," says McKay, "is getting those early
reflections to everybody in the audience. You always have to balance those
things against the requirement for 3,000 seats for box office. And there is
always a tendency to go toward some kind of fan shape in order to get good sight
lines. What you have to do is be ingenious enough to find ways to do that and
still not let the walls get pushed so far out or let the ceiling get pushed up
so high that you lose the early reflections.
"The classical halls and the newer ones like in Salt Lake City start with a
rectangle, admittedly, but it's a very wide rectangle in order to get enough
seats," McKay says. "Now how to narrow that down? Well, I start cantilevering
out some balconies so that if I look at a section view, I have balcony-main
floor-balcony. Now the hall, rather than being 80 feet wide, is effectively 60
feet wide between balcony faces."
Some designers have played creatively with more complicated geometries in order
to experiment with acoustics. A good example is Segerstrom Hall in Costa Mesa,
California, whose interior surfaces vaguely suggest the Death Star in "Star
Wars." Here, the basic fan shape is broken up by outcroppings of raised seating,
creating vertical reflecting surfaces throughout the auditorium. The hall has
good presence, but one can sometimes hear the call of a phantom horn coming from
somewhere overhead.
Where should a listener sit in order to enjoy the best sound? In a well-designed
hall, it shouldn't make much difference whether the seat is on the main floor or
in the balcony. In some older halls with high ceilings, though, the balcony is
often a better place because of its proximity to the ceiling; here, a listener
begins to receive early reflections from overhead that would not reach him
otherwise. If a listener wants to hear an orchestra the way the musicians
themselves hear, it's best to sit down in front (or in some of the newer halls,
beside or behind the orchestra). At these distances, the sound of an orchestra
is powerful enough to overwhelm the sound of the hall. As one moves back eight
or more rows on the main floor, the hall's acoustics begin to come out. The
sound should be fine anywhere on the main floor except under a deep balcony,
where the sound is weaker and more remote than in the open hall. Some orchestra
halls like the one in downtown Phoenix have a rear balcony that isn't
cantilevered from the rear wall but stretches like a beam between the two side
walls and is open at the back. This "flying balcony" design allows the sound to
penetrate the rear of the main floor by going behind the balcony above; a valid
technique, it's one that takes much of the disadvantage out of sitting under a
balcony.
In a way, the 20th century could be called the acoustic century. In the last
hundred years, this most aesthetic of sciences has experienced a dynamic period
of fads, experimentation, intuitive trials and gross errors. Like the changing
winds, designers have plied their visions in divergent directions with a single
goal in sight ~ to find a sonic ideal for live music. Every concert hall built
was like a stepping stone toward that goal ~ a million-dollar experiment leading
to the next improvement. Along the way, those designs have been influenced by
equal parts physics, politics, art and caprice. There was even a time when some
concert hall designers were influenced by the dry, brilliant, unreal sound of
stereo recordings with 32-track mixing.
