Académique Documents
Professionnel Documents
Culture Documents
OM 70 - 80
ING CEILING
REVERBERATION
REVERBERANCE TIME(RT) ( 2 . 0 SEC)
LOUDNESS LOUDNESS
(GI OR
SURFACE
TEXTURE
AND SOUND LATE
SOUND ENERGY ARRIVING INTERAURAL CROSS
SPACIOUSNESS DIFFUSING MATERIALS FROM T H E SIDES (AFTER 1 00 MS) CORRELATION lACC
U R G E ROOM VOLUME (<0.50)
Table 1 . Summary of current hypotheses linking acoustical qualities with physical acoustic phenomena, architectural design features of
rooms and acoustical measures that have resulted from this hybrid research and design effort.
variety of locations in a scale model of a room being LABORATORY STUDIES OF SOUND QUALITY
designed. Changes can be easily made to the model allowing
Once one has developed a holistic understanding of the
several alternative design schemes to be explored. An
iterative series of impulse responses can be obtained that acoustical qualities of rooms derived from listening and
allows one to understand the acoustic effects of their design measurements in situ, hrther study can proceed by isolating
decisions while the building is still in the early stages of variables in laboratory tests. Sound reflections from archi-
design. tectural surfaces that contribute to the impulse response can
It is also possible to mix or convolve music or speech be isolated and studied effectively in physical or computer
recorded in an anechoic environment with impulse responses models of rooms. How individual sound reflections contrib-
obtained from scale models or computer models of rooms that ute to impressions of acoustical quality can also be studied
are being designed resulting in an aural simulation of the in laboratory situations.
acoustical qualities. One can then listen to the aural simula- Recordings of anechoic music that have been mixed or
tion of music or speech as it would sound in the room that is convolved with the impulse response of a specific seat in a
being designed. One can compare the aural qualities of hall can be altered and played back to listeners through
scheme 1 with scheme 2. For example, one could compare the headphones or loudspeakers in an acoustical listening room
aural effects of changing the angle of a ceiling panel or so they can qualitatively evaluate the simulated sounds.
widening the roomsomewhat (Beranek, 1996;Siebein, 1994). Similarly, using digital signal processing equipment, rever-
8 5 T H A C S A A N N U A L MEETING ANDTECHNOLOGY (ONFEKENC E 55
Fig. 2. Acoustical model of a new concert hall in the Singapore Arts Center constructed and tested at the University of Florida with Stirlingl
Wolford Architects and Artec Consultants, Inc. This hall is an example of the Concert Hall of the 21st Century described below.
beration and individual reflections can be added to music or NEW METHODS FOR ACOUSTICAL ANALYSIS
speech. The reflections can be programmed to come from AND DESIGN
specific directions, at specific loudness levels and at particu-
The concert hall of the 21st century has evolved as an
lar time intervals after the direct sound to simulate the effects
interesting hybrid that assimilates principles of shape, pro-
of reflections from a ceiling or wall in a room. The arrival
portion, material, configuration and program from the his-
time, frequency content, loudness and direction of the reflec-
toric precedents of the late nineteenth century. The acous-
tions can be changed to simulate a second location for a wall
tical and architectural qualities of the halls of the future have
or an alternative material selection. The second sound can
be played to listeners so they can give an appraisal of the been extended beyond those of the precedent halls through
relative qualities of each condition (Ando, 1985; Beranek, an extensive research effort that has been integrated and
1996; Blauert and Lindeman, 1986; Cremer, 1982; Gold, explored in practice in an exemplary manner. The remark-
1994; Haas, 1972; Soulodre and Bradley, 1994). able acoustical qualities of these halls have been obtained in
This ability to isolate specific sound reflections in a a rich and varied contemporary architectural vocabulary.
complex impulse response has been a major contribution to The acoustical attributes of the precedent halls have been
the understanding cfroomacoustics. One can identify which identified through impulse response tests conducted in the
reflections are responsible for specific sound qualities. One rooms. Aural simulations have been used in laboratory
can also identify which architectural features of the room studies to evaluate the subtle qualities of sounds heard in the
have produced the reflections. This allows one to design the precedent rooms. These techniques allow specific acoustic
aural environment of rooms and evaluate the aural conse- or architectural variables to be isolated and examined in
quences ofthe design before the room is built in a manner that detail. This hybrid method that combines architectural
is similar to the way in which architects have traditionally analysis of historic halls, qualitative evaluation perfor-
designed the visual environrncnt of rooms. mances in existing rooms and simulated sound fields in the
56 ,ARCHITECTURE: MATERIAL AND IMAGINED
A . SOUNDREFLECTIONS
ARRIVING AT LISTENERS FROM
--
THE SIDES
I
B. ROOM WIDTH KEEPS THE
-
AUDIENCE MEMBERS CLOSE TO
.2
THE WALL SURFACES.
D. PERFORMANCE
AREA
LOCATED IN SAME SPACE AS
AUDIENCE
E. SOUNDREFLECTIONS FROM
THE CEILING
,
L - - - F. WALL, UNDER-BALCONY,
1 SOUND REFLECTION PATH
-
-
- - - - G . UPPER ROOM VOLUME ALLOWS
MULTIPLE SOUND REFLECTIONS TO
- - INCREASE REVERBERATION
--+ 1
-
-- -
<-
.
- - - +- -
- -
H. WALL, UNDER-BALCONY. SOUND
- - - - ---
REFLECTION PATH
J
SIDE WALLS (UP TO 100 MS).
w
>
W
-I /- REVERBERANT SOUND ENERGY ( 1 00
n MS TO 2000 MS [ 2 SECONDS]).
z
J
0
cn
laboratory, impulse response measurements in buildings and seats in the rooms. The room surfaces are located in close
models and a series of actual buildings constructed over a proximity to listeners in rooms of this size so sound
thirty year period has produced the first series of successful reflections can arrive at their ears shortly after the arrival
halls since the end of the last century (Johnson, 1990; of the direct sound. These early sound reflections will
Siebein, 1994). heighten sensations of clarity and intimacy. The sound
will also be louder in a smaller room. Adequate loudness
is an important factor that contributes to the dynamics of
THE ARCHITECTURAL AND ACOUSTICAL
music.' It is also important in providing the balance
QUALITIES OF THE PRECEDENT HALLS
between loud crescendos and very quiet, subtle moments
The concert halls of the late nineteenth century were rela- when only a single string or flute may play (Beranek,
tively narrow roo~nswith a high ceiling and parallel walls. 1996; Johnson, 1990).
There were several narrow balconies wrapped around the 5 . Ceiling height and reverberation. The precedent concert
sides and rear of the rooms. The walls and ceilings of the halls typically had high ceilings and a relatively large
rooms were built of heavy, massive materials, yet they were room volume that contributed to the rich, full, reverber-
covered with ornate, diffusing surface textures. The halls ant sound for which they are known (Beranek, 1996;
provide a rich blend of articulate, clear sound that is simul- Johnson, 1990).
taneously enveloping and richly reverberant. A summary of 6. Ceiling height and early sound reflections. The high
the architectural features ofthe rooms and the corresponding ceiling height, however, did not provide an overhead
acoustical qualities is presented below. surface close to the main floor seating from which early
1. Narrow room width. The rooms were typically 70-80 ft sound reflections could amve at listeners ears. The early
wide. This narrow width allows sound reflections from sound reflections from overhead have been shown to
the side walls to arrive at the listeners' ears shortly after increase the sensations of loudness and clarity if the
the direct sound. These early lateral reflections contrib- reflections arrive within 80 milliseconds after the direct
ute to a sense of acoustic envelopment and a broadening sound (Haas, 1972).
of the apparent size of the acoustic source (Barron, Many musicians have trouble hearing each other or
1993).' maintaining a sense of ensemble in rooms where the
2. Shoebox shape. The long, parallel side walls that resulted ceiling is too high. Musicians rely on sound reflections
from the shoebox shape helped to increase the sense of from the stage enclosure to keep time with each other and
reverberance in the rooms. Multiple sound reflections to maintain a sense of ensemble (Beranek, 1996; Gade,
would move between the parallel surfaces from the front 1989; Siebein, 1994).'
of the room to the rear. This would create a longer 7. The location of the orchestra. The orchestral platform
reverberation time than would otherwise occur for a room was located within the main volume of the room. Many
of a given volume and absorption (Johnson, 1990).4 halls even had some seating behind the orchestra. This
3. Multiple tiers of narrow balconies. The rooms were resulted in an enhanced sense of acoustic intimacy in the
wrapped with two or three tiers of narrow balconies. rooms.7 It established a direct connection between the
These balconies extended along most of the length of the music played by the performers and the audience that far
side walls and along the rear wall. In some halls these exceeds that which is achieved when an orchestra plays
balconies even wrapped around the front wall ofthe room within the confines of a proscenium stage (Johnson,
behind the orchestra. There are several interesting acous- 1990).
tic effects that are attributed to these tiers of narrow 8. Floor slope and seating. The low floor slopes and
balconies (Johnson, 1990). unupholstered seats in many of the precedent halls con-
Sound from the orchestra could be reflected from the tributed to relatively high loudness levels. As the floor
bottom of the balcony to the side wall of the room and slope increases, more of the direct sound strikes the
back to the center of the main floor. This "cue ball" absorbent surface of the audience and is attenuated. This
reflection contributed to the sense of acoustic envelop- absorption is increased when heavily upholstered seats
ment in the rooms and also made the acoustic image of the are used (Johnson, 1990).
sound source appear wider than it actually was (Barron, Strong direct sound waves and early reflections also
1993; Cremer, 1982). allow one to localize or pinpoint the position of the sound
The faces of the balcony rails were usually curved or sources on stage. This is an interesting acoustical quality
ornately detailed so they would diffuse sound into the because one notices that composers deliberately move
room. This surface would help create the differences in the source of sound through the orchestra at various
the sounds that arrive at the left and right ears of the moments in a musical piece (Siebein, 1994).
listeners increasing their preference for the sound field 9. Materials. The materials in the rooms were generally
(Gold, 1994). heavy and massive. Thick plaster that was richly de-
4. Room size. The rooms were relatively small by today's tailed, masonry walls and other heavy construction ma-
standards. There were usually between 1,000 and 2,000 terials were typically used. These materials are sound
58 ARCHITECTURE: MATERIAL AND IMAGINED
A . REVERBERATION
CHAMBER
ABOVE
6 . SOUNDREFLECTIONS ARRIVING
AT LISTENERS FROM T H E SlDE WALLS
AND BALCONIES
C . TWOL E V E L MAIN FLOOR ALLOWS A
SLIGHTLY WIDER ROOM WITH PARTERRE
WALL TO PROVIDE EARLY SOUND
REFLECTIONS
E. MOVABLECANOPY AT CEILING
F. PERFORMANCE
AREA
LOCATED IN S A M E S P A C E A S
AUDIENCE
G . REVERBERATION CHAMBER TO
EXTEND REVERBERANCE.
H . WALL, UNDER-BALCONY,
SOUND REFLECTION PATH.
. K . WALL, UNDER-BALCONY,
SOUND REFLECTION PATH.
Fig. 4. Summary of the architectural design features of the Concert Hall of the 21st Century.
8STH ACSA A N N U A L M E E T I N G A N D T E C H N O L O G Y C O N F E R E N C E 59
reflective in all frequencies.Vhe relatively large surface be directed back to the musicians across stage increasing
areas insured that even low frequency sounds with long the sense of ensemble or support.
wavelengths would be reflected in the room. There were The canopies can also be raised and lowered to change the
few if any sound absorbent surfaces used in the rooms acoustical qualities of the room for different types of
with the exception of the audience and seats. This performance. For romantic symphonies with full orches-
resulted in maximizing the reverberation time and loud- tra, chorus and organ, the canopy will be located very
ness in the room for a given size and shape (Beranek, high in the room. This allows the full organ sound to
1996, Johnson, 1990). propagate into the room. It also allows the sound of the
10.Textured walls and ceilings for sound difhsion. The large source (orchestra, chorus and organ) to mix or blend
walls and ceilings were usually covered with surfaces that somewhat in the vicinity of the platfonn before it moves
would diffuse the sound. Sound difhsing surfaces in- into the room. The high canopy location also provides the
cluded the faces of several narrow balconies that lined the maximurn access of sound to the reverberation chambers
walls; the recesses and protrusions of pilasters and other allowing extremely long reverberation times to be
plaster relief work; and statues or large niches that lined achieved in the rooms.
the walls. The difhsing surfaces would break-up or When smaller groups such as chamber orchestras per-
scatter the sound that hit them. This would cause the form, the canopy can be lowered. This allows the
sound to actually reach the listeners' ears from many strongest sound reflections from overhead to move to-
different directions. A sense of acoustic spaciousness wards the audience and arrive at listeners' locations
was achieved by the distinctly different reverberant closer in time to the arrival of the direct sound. This will
sound fields that arrive at the left and right ears of people enhance the sensations of intimacy and clarity of string
listening in the rooms (Beranek, 1996; Chiang, 1994; passages for example and increase the relative loudness
Schroeder, 1974).9 level of the weaker source.
Reverberation chambers. Large, sound reflective vol-
umes are acoustically coupled to the main room volume
EXTENDING THE ACOUSTICAL QUALITIES OF
to increase the sense of reverberance, spaciousness and
THE PRECEDENT HALLS
warmth in the rooms. Large, massive doors can be
The Concert Hall of the 21st century returns to these opened or closed allowing more or less sound to move
precedents as its basic parti. It extends the acoustic into the reverberation chambers. The sound will reflect
potential of performance and the architectural envelope of off surfaces within the chambers and then move back into
space by adapting the form of the precedent halls to the the main room volume. Recent concert halls in Tampa
needs and desires of the future. Massive, moveable cano- Bay, Dallas and Birmingham incorporate reverberation
pies are placed over the stage to improve the sense of chambers with volumes that are equal to more than one
ensemble among the perfonners and to provide variability half or more of the volume of the main house. Designs
in the response of the room to accommodate various types for newer halls in Singapore and Kobe have even larger
of orchestral performances. The sensation of reverberance reverberant chamber volumes that are equal to or greater
can be increased rather dramatically by coupling large than the volume of the main house.
adjacent volumes to the hall itself through a series of The doors that couple the chamber volume to the main
operable openings. These reverberation chambers have house can be opened and closed in a variety of configu-
been extremely successful in the new halls at Dallas and rations to extend the rich reverberant sound up to six
Birmingham. Acoustical banners can be lowered to cover seconds in some cases. The sound from the chambers
portions of the side walls of the room to allow the room to enters the main room at a lower sound level than the
change from very reverberant to extremely dry. The design decaying sound field in the room. This creates a decay
features of these rooms are summarized below (Johnson, curve with two slopes. First is an early decay that falls off
1990; Siebein, 1994). relatively quickly in the main room volume. This pro-
1. Canopy. A large sound reflecting canopy is constructed vides clarity and loudness for the music. Second is a later
to float above the orchestral platform and the front rows decay at a more gentle slope as sound from the chambers
of seats to enhance several of the acoustical qualities of continues for quite some time. This adds a rich, but clear
the precedent halls. The height of the canopy can be reverberant quality to the sound in the room.
adjusted in many rooms to accommodate a variety of The extent to which the chambers have become an
performance types. Early sound reflections from the integral part of the design of many concert halls has
orchestra will strike the canopy and arrive at listeners' increased over time as well. In Tampa Bay, the chambers
seats shortly after the direct sound. These early reflec- covered only the upper portions of the room. The
tions will contribute to an enhanced sense of acoustic reverberant sound was perceived as coming from above
intimacy, increase the loudness level at main floor seats or hovering above the room. Beginning with the Dallas
and increase the perception of clarity in the room. Some hall and continuing through the halls currently under
of the early sound reflections from the canopy will also design in Singapore and Kobe, the chambers not only
60 AK(HITE(TL1KE. MATERIAL A N D IMAGINED
surround the upper portion of the room, but also extend materials are usually stored behind the walls where the
down the side walls. This allows the reverberant sound sound from the orchestra can not strike them.
to be heard as coming from all parts of the room,
immersing the audience in the fullness of the music. CONCLCSIONS
The reverberant chamber concept and the resulting double
The Concert Hall of the 21st Century is emerging as an
sloped decay have allowed the apparently paradoxical
exciting acoustical and architectural building type. The
conditions of clear and reverberant sounds to coexist in
acoustical qualities of these rooms have been developed
the same room. This series of architectural experiments
through careful studies of historic precedents, modeling,
has contributed a new acoustical condition to the palette
simulating, listening and designing ina hybriddesignmethod.
of designers.
The impulse response has been the catalyst that has fused
3. Acoustical banners. Sound absorbent panels or retract-
current knowledge in music, electrical engineering, com-
able banners are also used in the new concert halls. At
puter simulation. virtual reality, architecture, acoustics and
times when the room should be acoustically dry, such as
statistics in a highly successful approach to design rooted in
during a rehearsal or during a speaking function, the
knowledge that is emerging from all of these constituent
banners or panels can be extended to cover most of the
disciplines (Beranek, 1996; Johnson, 1990; Siebein, 1994).
wall surfaces in the room. This will reduce the loudness
and reverberance in the room. It will provide a suitable
environment for amplified speech for a lecture or debate. REFERENCES
Some smaller areas of banners may be used when cham-
Ando, Y. Concert Hall Acousrics, Springer-Verlag, Berlin:
ber groups or renaissance music is played in the hall 1985.
where clarity is more important than reverberance and Barron, Michael. Auditorium Acousrics and Architecrural Design.
fullness. When the hall is used for concerts, the absorbent E & FN Spon, London: 1993.
Fig. 5. The new symphony hall in Birmingham, england by Artec Consultants, Inc. represents one of the interim steps in the emergence
of the Concert Hall of the 21st Century.
85T"ACSA ANNUAL MEETING ANDTECHNOLOGY CONFERENCE
Beranek, L.L. Concert and Opera Halls: How They Sound. If a system under test is subjected to an impulse, the response
Acoustical Society of America, Woodbury. New York: 1996. obtained is a result of the properties of the system plus the
Blauert, J. and Lindemann, W. "Auditory Spaciousness: Some characteristics of the impulse.
Further Psychoacoustic Analyses." JASA, Vol. 80, 1986. pp. Impulse responses were originally recorded from a hand clap,
533-542. a bursting balloon, a gun shot or an electronic pulse played
Carvalho, A.P.O. "Objective Acoustical Analysisof Room Acoustic through a loudspeaker. They were recorded on oscilloscopes
Measurements in Portuguese Catholic Churches," Proceedings and photographed for analysis. Currently sophisticated pseudo
of 1994 National Conference on Noise Control Engineer~tig, random noise signals such as maximum length sequences and
Fort Lauderdale, FL. complex sine sweeps are programmed in signal processing
Chiang, W. "Effects of Architectural Parameters on Six Acousti- software in a portable computer, sent through a digital to analog
cal Measures in Auditoria," A Pl1.D Dissertation, University of converter, amplified and played through loudspeakers into the
Florida, Gainesville, FL: 1994. room. The pseudo random noise signals are more repeatable
Cremer, L. and Muller, H.A. Principles andApplications ofRoorn than the actual impulsive sources such as the hand clap or gun
Acoustics, Vol. 1 and 2, English Translation with Additions by shot.
T.J. Schultz, Applied Science Publishers, New York: 1982. ' The lateral fraction is an acoustic measure that relates the
Gade, A.C. "Acoustical Survey of Eleven European Concert amount of sound that is received at a seat from the sides to the
Halls," Report No. 44, The Acoustics Laboratory, Technical total sound energy that arrives at the seat. The lateral fraction
University of Denmark, Copenhagen, 1989. has been related to the acoustical quality of envelopment and
Gold, M.A. "Subjective Evaluation of Spatial Impression: The source width in laboratory studies where simulated sound fields
Importance of Lateralization," Proceedings of the Wollace were constructed to isolate the acoustic effects of early lateral
Clement Sabine Svmposium of the 127"' Meeting of the Acous- sound reflections.
tical Society of America, Cambridge, June 1994, "everberation time is the amount of time it takes for the sound
Hass. H.. "The Influence ofa Single Echo on the Audible Speech." to decay to a level that is 60 dB less than the original level. It
JournalofAudio Engineering Society. Vol. 20, No 2: 1972. pp. is derived from a least squares fit of the reverse integration of
146- 159. the impulse response of a room.
Hawkes, R.J. and Douglas, H. "Subjective Acoustic Experience in The relative strength or relative early loudness are acoustic
Concert Auditoria." Acustica. Vol. 24 No. 5: 1971. pp. 243-247. measures of loudness. Many of the precedent halls have
Johnson, R. "Reflections on the Acoustical Design of Halls for remarkably consistent loudness levels throughout the rooms.
Music Performance," V.O. Knudsen Lecture, A paper Pre- The hearing conditions on stage are currently evaluated by an
sentedon the 120 th Meeting of the Acoustical Society of acoustical measure called support. The energy of reflected
America, San Diego, November, 1990. sounds heard on stage are recorded over short time intervals.
Katsuaki, S., Kimura, S. and Hanyuu, T. "Analysisof Sound Field The time intervals were chosen based on interviews with
on Spatial Information Using a Four-Channel Microphone musicians regarding stages where adequate support for their
System Based on Regular Tetrahedron Peak Point Method," listening was provided by the room.
Applied Acoustics, Vol. 37, 1992, pp. 305-323. ' The sense of intimacy is related to the initial time delay gap.
Kuttruff, H. Room Acoustics, 3rd Edition, Elsevier Applied This is the difference in arrival time between the direct sound
Science, London: 199 1. and the first reflection from the room surfaces.
Schroeder, M.R. "New Method of Measuring Reverberation "he bass ratio is an indication of the relative warmth or the
Time," The Journal ofAcoustical S0ciet.v ofAmerica. Vol37, support of bass or low frequency sounds in the room. It is a ratio
No. 4, 1965, pp. 409-412. of the reverberation times in the lower frequencies to the
Schroeder, M.R., Gottlob, D., and Siebrasse, K.F. "Comparative reverberation times in the middle frequencies. The bass ratios
Study of European Concert Halls: Correlation of Subjective were usually high in the precedent halls.
Preference with Geometric and Acoustic Paramcters." The The treble ratio is an indication of the relative brilliance or
Journal ofAcoustica1 Society ofAmerica. Vol56, No. 4, 1974, support of higher frequency sounds in the rooms. It is a ratio
pp.1 195-1201. of the reverberation times in the higher frequencies relative
Siebein, G.W. "Acoustics in Buildings," A paper presented on the to the reverberation times in the middle frequencies. The
1271h Meeting of the Acoustical Society of America, Cam- treble ratios tended to be fairly high in the precedent rooms
bridge, May 1994. as well.
Souiodre, G.A. and Bradley, J. "A Subjective Evaluation of The interaural cross correlation or IACC is a physical measure
Acoustic Measures in Concert Halls Using Binaural Simula-
that is intended to indicate the relative dissimilarity of the
tion," Proceedings of the Wallace Clement Sabine $vnposium sounds that arrive at the left and right ears of people. In
of the 127'" Meeting of the Acoustical Society of America,
laboratory studies it has been found that seats in rooms that have
Cambridge, June 1994, pp.27 1-274. low lACC values are generally preferred by listeners. This
would mean that people prefer listening to music where the
ENDNOTES sound that arrives at the left ear is different than the sounds that
I The impulse response test method was borrowed from digital arrive at the right ear. Most of the precedent halls have
signal processing. It is a basic premise of linear systems theory. relatively low IACC values.