發表論文研究

Monograph ................................................................. 研究論文發表

樂團成員、觀眾與音樂廳音響性能的交互影響

Field Study of Stage Acoustics for Solo and Chamber Music

陳世堂*1 許晏坤*2 江維華*3

ABSTRACT

Stage acoustics were evaluated in 4 halls regarding solo and chamber music performances. The performers expressed their opinions about hearing oneself, hearing others, ease of ensemble, and overall impression at two locations with and without detached reflectors in each hall. Physical measurements were taken accordingly. As a result, most acoustical environments were acceptable to performers. Overall impression was significantly correlated with all of the subjective attributes. The optimum values of the measures associated early energy were close to the –12 dB suggested by Gade. The piano player preferred a high late-to-direct energy ratio. The strength of cello, horn, and trombone were essential for the balance in audience.

本研究於四個演藝廳現場進行獨奏與室內樂(二重奏、三重奏、五重奏)演出的舞臺環境評價實驗，曲目涵蓋巴洛克至浪漫派的原作與改編曲。樂手對自我聲音、他人聲音、合奏容易度與整體感受做即時評估，觀眾則對各聲部音量強度和整體感受作評估。在每個廳中並以演奏位置(中央與前緣)與兩側獨立反射板為控制變因。結果顯示大多數音響環境都被樂手所接受，接近舞臺邊緣並有反射板的組合在每個廳中普遍獲得樂手最高評價，也能提供觀眾席較佳的聲部音量平衡，樂手整體感受與其他三個主觀屬性間有強烈的線性關係。整體而言，音響性能參數ED80的最佳值約為-12 dB，接近Gade對交響樂團的建議值，而後期反射並未對主觀參數產生顯著影響，此外樂手間對環境的要求差異很大，如鋼琴樂手喜歡較充足的後期反射能量。就觀眾而言，大提琴、法國號、與長號的音量是決定音量平衡的主要因素。

1. INTRODUCTION

Not until Marshall’s study published in 1978, very little information about the acoustics of stage has appeared in the literature regarding room acoustics. His study indicated the importance of providing ensemble with early reflections in the frequency above 500 Hz and with delay time in the range of 17 to 35 ms [6].

In 1989 and 1990 Gade reported studies carried out in the laboratory as well as in the field. All studies revealed the importance of and support factor (ST1). The support factor ST1 (also denoted as STearly in other literatures) is given by the equation:

(1)

where E denotes energy measured 1 m from the source. Based on the field studies, Gade suggested an optimum ST1 of -12±1 dB for the orchestra. He also concluded that, however, chamber groups might feel more comfortable in smaller rooms [3][4][5].

In 1990 Bruck used a concert hall with an adjustable shell to conduct subjective tests by chamber music players. Overall, the percentage variance of the subjective preference explained by ST1 was lower than which reported by Gade. While brass players desired early reflections, string players favored sufficient reverberance [1].

2. METHODS

This research is conducted to investigate the subjective assessment of stage environment of solo and chamber group performances. Four halls were selected with varying seating capacities and reverberation times [2].

The stage of Hall 2 is coupled with reflectors only behind the performers while the stage of Hall 1 ,3 and 4 is fully enclosed. In each hall performers were seated in two locations with and without detached side reflectors. The 4-m² reflectors, 6 m apart from each other, were used to increase early reflections. After an approximately 50-sec long music segment was played, the performers expressed their opinions about hearing oneself, hearing others, ease of ensemble, and overall impression. Table 1 summarized the music works used for the evaluation.

Except early support (STe) and late support (STl), three measures were also calculated. Early to direct energy ratio (ED80) was a measure similar to STe but the interval of integration between 10 to 80 ms was used instead of 20 to 100 ms. Late to direct energy ratio (LD80) was a measure similar to STlate but the interval of integration between 80 ms to infinity was used instead of 100 ms to infinity. Echo criteria (EC) proposed by Dietsch and Kraak was used to evaluate echo. Temporal diffusion (TD) was used to evaluate coloration and fluttering echo. The Environmental Research Group, NTUST developed the measurement system. With a linear sweep sine wave played through a Norsonic 223 dodecahedron speaker, a B&K 4192 microphone was used to record monaural signals. Dummy microphones (Neumman Ku100) was used to record binaural responses. Fig. 1 Seating capacity, unoccupied reverberation time in the audience and profile of stage section comparing the 3 halls investigated.

Table 1. The music works evaluated.

	Segment 1	Segment 2	Segment 3
Violin Solo	Prokofiev, cof.115, Mov1, theme 2	Prokofiev, cof.115, Mov. 1, theme 2	Prokofiev, cof.115, Mov2, theme 1
Piano Solo	Bach, Bnv. 860	Chopin, Op.23	Debussy, Suite Bergamasque
Violin Sonata	Brahms, Op. 108, Mov. 1 (Allegro)	Brahms, Op. 108, Mov. 2 (Adagio)	Brahms, Op. 108, Mov.3 (Presto)
Piano Trio	Beethoven, Op.38, Mov. 1	Mendelssohn, Op.49, no.1, Mov.1	-
Brass Quintet	Handel , Water Music, (Arr. Bames), “Alla Hornpipe”	Handel, Water Music (Arr. Bames), Largo	Mozart. K.525, Mov. 1 (Arr. Robert. King), Allegro

Table 2. Correlation coefficients (strings and piano/ brass/solo) among parameters and acoustical measures. Coefficients with p£0.05 were shown in boldfaces. (SELF = hearing oneself; OTHR = hearing others; EASE = ease of ensemble; OVL = overall impression)

	SELF	OTHR	EASE	OVL
SELF	1
OTHR	0.70 /0.69	1
EASE	0.76 /0.57	0.77 /0.90	1
OVL	0.89 /0.66	0.90 /0.98	0.81/0.92	1
ED80	-0.78/-0.23	-0.81/-0.29	-0.64/-039	-0.85/-0.30/-0.62
STe	-0.75/-0.20	-0.78/-0.15	-0.68/-0.27	-0.87/-0.16/-0.68
LD80	-0.00/0.09	-0.16/0.41	-0.29/0.48	-0.00/0.49/-0.00
STl	0.08/0.03	-0.11/0.34	-0.25/0.46	0.06/0.43/0.07
EC	0.42/0.28	0.34/0.42	0.06/0.46	0.50/0.51/0.39
TD	-0.43/-0.28	-0.47/-0.30	-0.22/-0.33	-0.39/-0.27/-0.29

3. RESULTS

Subjective Evaluation. Except the brass quintet players scored over impression in one situation under 0, all of the acoustical environments were acceptable to performers.

Overall impression was significantly correlated with all three subjective attributes which were also correlated with each other.

Both early-to-direct energy ratio (ED80) and early support (STearly) was correlated with overall impression. With all observations included and an outliner excluded, a 2^nd order polynomial fit of overall impression can be derived. The regression shows an optimum ED80 of –12 dB. (FIG. 2) This is close to the suggested STearly by Gade for orchestra.

Table 3. Correlation coefficients among measures.

	ED80	STe	STs	LD80	EC	TD
ED80	1
STe	0.890	1
STl	-0.002	-0.062	1
LD80	0.047	-0.015	0.978	1
EC	-0.431	-0.391	0.617	0.659	1
TD	0.594	0.307	0.218	0.302	-0.122	1

Fig. 2 Overall impression as a function of ED80. A 2^nd polynomial fit can be derived if an outliner(x) was excluded

Acoustical measures associated with late energy (LD80 and Stlate) were found to be independent from the ones associated early energy. (TABLE 3) Subjective attributes were not correlated with late-to-direct energy ratio (LD80) with all of the data. (FIG. 3) However, significant differences existed among players. The data by piano player showed the preference of strong late energy in Hall 2 and Hall 4. (marked “o” in FIG. 4) More data are required to certify the effects of late energy or reverberation on stage.

The environments with side reflectors were favored by string and piano players while the front positions without the reflector were favored by brass players. (FIG. 5) The side reflectors were effective in enriching the strength of cello, horn, and trombone in the audience. (FIG. 6)

Fig. 4 Averaged data of overall impression (solid line) and hearing oneself (dashed line) in each hall comparing the violin player (□) to the piano player (o) and cello player (x).

Fig. 5 Attributes on stage comparing the front location (solid line) to the center (dotted line) and the presence of reflecting panel (x) to absence (o). (SELF= hearing oneself; OTHR= hearing others; EASE= ease of ensemble; OVL= overall impression; s= strings and piano; b = brass)

Fig. 6 Strength in audience comparing the front location to the center and the presence of reflecting panel to absence.

Much published information about the design techniques refers to orchestral performance although it appeared that early reflections were more desired by chamber groups than by orchestra players. In this study with ED80 in the range of –14 dB to –6 dB, almost all the acoustical environments were acceptable to the performers. The optimum values of the measures associated early energy were close to the –12 dB suggested by Gade. More data have to be collected to assess other physical characteristics on stage.

1. Bruck, D.C. “Musician’s Preference for Concert Stage Acoustics”, 109th Meeting, Acoust. Soc. Am., (1990)

2. Chiang, W. “Subjective Evaluation of Acoustical Environments for Solo Performance”, Building Acoustics, 21, 18-36, (1999)

3. Gade, A.C. “Acoustical Survey of Eleven European Concert Halls- a Basis for Discussion of Halls in Denmark”, The Acoustics Lab., Tech. Univ. of Denmark, Report No.44. (1989)^a

4. Gade, A.C. “Investigations of Musicians’ Room Acoustic Conditions in Concert Halls. I. Methods and Lab-oratory Experiments”, Acustica, 69, 193-203. (1989)^b

5. Gade, A.C. “Investigations of Musicians’ Room Acoustic Conditions in Concert Halls. II. Field Experiments and Synthesis of Results”, Acustica, 69, 249-261. (1989)

6. Marshall, H., Gottlob, D. and Alrutz, H. “Acoustical Conditions Preference for ensemble”, J. Acoust. Soc. Am., 64, 1437-1442. (1978)

ED80

STe

STs

LD80

EC

TD

ED80

1

STe

0.890

1

STl

-0.002

-0.062

1

LD80

0.047

-0.015

0.978

1

EC

-0.431

-0.391

0.617

0.659

1

TD

0.594

0.307

0.218

0.302

-0.122

1