Electronics, music and computers

Electronic and computer technology has had and will continue to have a marked effect in the field of music. Through the years scientists, engineers, and musicians have applied available technology to new musical instruments, innovative musical sound production, sound analysis, and musicology. At the University of Utah we have designed and are implementing a communication network involving and electronic organ and a small computer to provide a tool to be used in music performance, the learning of music theory, the investigation of music notation, the composition of music, the perception of music, and the printing of music.

ELECTRONICS, MUSIC AND COMPUTERS by A l a n Conway Asht o n D e c e m b e r 1971 U T E C - C S c - 7 1-117 ACKNOWLEDGEMENTS I w i s h to thank Professor Robert S. B a r t o n for his motivation, s t i m u l a t i o n a n d ideas in special-purpose systems design, information., r e p r e s e n t a t i o n and programming structures. I appreciate the patience and e n c o u r a g e m e n t of Professors David C. Evans and Wil l i a m J. Viavant du ring m y g r a d u a t e study. I am p a r t i c u l a r l y indebted to Robe r t B e n n i o n for many design suggestions, interesting discussions and the hardware implementation of the p r o t o t y p e equipment. INTRODUCTION E l e c t r o n i c and Computer Technology has h a d and will continue to have a m a r k e d effect in the field of music. Thr o u g h the years scien­tists, engineers, and musicians have applied available technology to new musical instruments, innovative musical sound production, sound analysis, and musicology. At the University of Utah we hav e designed and are implementing a communication network involving an electronic organ and a small computer to provide a tool to be used in music p e r ­formance, the learning of m u s i c theory, the investigation of music notation, the composition of music, the p e r c eption of music, and the print i n g of music. The c omputer-aided m u s i c tool is shown in figure 1. The computer serves as a communications device that aids the user by performing several functions such as the storage of information, p e rformance of pre - a s s i g n e d s e q u e n c e s , and retrieving of specific stored data. The computer receives input from a typewriter keyboard, a light pen attach­ed to the display scope, the organ manuals, pedals, and stop settings. In turn, the computer controls the audio tone production, lights which indicate the organ keys, the display scope and a color generator. There are many interesting things which can be done w i t h this h y b r i d configuration; for instance, computer assisted learning of music can be investigated, experiments of sensory p e r c eption may be carried out, k e y b o a r d p e rformance can be studied and new sounds, rhythms and melodies m a y be intricately interwoven. I w i l l discuss these and other possibilities further after I describe the computer-aided music tool fully (chapter 4). B u t first, it will be instructive to look at some acoustic pro p e r t i e s of sound, the development of electronic musical i nstruments a n d the use of computers in the field of music so that in the overall pic t u r e the value and position of our system can be assessed and evaluated. Finally, I will look at recent trends and speculate on the future of electronics and computers in the field of music. ■ „ 2 TABLE OF CONTENTS A C K N O W L E D G E M E N T S .............................................. ' ii A B S T R A C T ................................................. .. v INTRODUCTION ................................................... 1 CH A P T E R 1 MUSIC ENGINEERING ................................ 4 Wa v e f o r m G e n e ration ..................................... 5 W a v e f o r m M o d i f i c a t i o n ................................... 12 S o u n d Coordination and O r g a n i z a t i o n ......... .. 25 W a v e f o r m S t o r a g e .......................................... 30 Sound C o n t r o l ............................................ 32 Music N o t a t i o n .......................................... .. 40 C H A P T E R 2 HIS T O R I C A L D E VELOPMENT O F ELECTRONIC MUS I C A L I N S T R U M E N T S ....................................... 43 Novel Experiments b e f o r e the V a c u u m T u b e ............ 43 Early Developments befo r e World War I I .............. 45 Extentions of the Early D e v e l o p m e n t s ................... 63 Electronically Controlled S y n t h e s i z e r s ................ 76 C H A P T E R 3 COMPUTERS AND MU S I C .............................. 89 Generation of Sound Waveforms ......................... 89 Music C o m p o s i t i o n ....................................... 92 Music P r i n t i n g ............................................ 95 Music A n a l y s i s .......................................... 96 M u s i c o l o g y ................................................. 100 E d u c a t i o n ................................................ 101 Control of M u s i c - G e n e r a t i n g M e d i a ..................... 102 Page iii General D e s c r i p t i o n ......... .. ......................... .. 105 The O r g a n I n t e r f a c e ..................................... .. 108 Internal R e p r e s entation . .............................. .. 116 D ata File S t r u c t u r e ..................................... .. 129 E x a m p l e s ................... .................... .......... 144 E x t e n s i o n s ................................................. .. 147 C H A P T E R 5 USES, CA P A B I L I T Y AND EVALUATION OF THE M U S I C A T I O N A L T O O L ................................... 150 Music P e r f o r m a n c e ......................... .. ........... .. 152 E d ucation .................................................... 156 C o m p o s i t i o n ................................................. 160 Score W r i t i n g ............................................ .. 161 Study of M u l t i s e n s o r y Perception 163 R E F E R E N C E S ...................................................... .. 167 CHAPTER 4 THE COMPUTER AIDED MUSIC T O O L .....................105 iv ABSTRACT* The uses of electronic and computer technology in m u s i c are . s u r v e y e d and a c omputer-aided music tool is described w h i c h consists of a small computer, e l e c tronic organ, and a line-drawing display ' scope. The computer receives input from a teletypewriter keyboard, the o r g a n keys and stop tablets. The input information is processed a c c o r d i n g to stored p r o g r a m s , and outp u t is directed to the o r g a n tone g e n e r a t o r s , filter selection switches, lights wh i c h desi g n a t e the m a n u a l keys, volume control and the display scope. Thus, the organ is full-duplexed through the computer, since the keys are not directly .connected to the tone generators, b u t are linked v i a the computer. A b a s i c internal r e p r e s e n t a t i o n of stored m u s i c is deve l o p e d which c o n sists of an interpreted file containing a sequence of v a r iable l e n g t h commands and typed data. Information transfer rates of a seco n d a r y storage tape c a s sette r e c order is adequate for m o s t clas s i ­cal or g a n selections, w h i c h m a y be directly entered into the computer by p e r f o r m a n c e on the m a n u a l keyboards and pedals, or keyed-in, note by note, either from a nother type keyboard or from the organ keys t h e m s e l v e s . The interactive o r g a n - computer communication network allows real time visual, aural and w r i t t e n response. This makes the musi c a t i o n a l ★ i i . This r e port reproduces a thesis of the same title submitted to the D e p a r t m e n t of Electrical Engineering, Division of Computer Science, U n i v e r s i t y of Utah, in p a r t i a l fulfillment of the requirements for the degr e e of Doctor of Philosophy. v tool, as it is called, useful for p r o g rammed mu s i c courses, computer ai d e d i n struction of m u s i c theory, h armony and keyboard performance, ‘ and c o n d ucting h u m a n i n formation p r o c essing experiments in the inves­t i g ation of m u l t i s e n s o r y perception. _ The m e t h o d s of m u s i c production, p r o c essing power, and capabilities of the m u s i c a t i o n a l tool are compared to the techniques of other e lec­tronic. and c o m p u t e r - a i d e d mu s i c a l instruments and the influence of . el e c tronic and computer technology in m u s i c of the future is discussed. 3 Lights w h i c h i l l u m i n a t e the m u s i c keys Figure 1 The M u s i cational Tool CHA P T E R 1 MUSIC E N GINEERING The auditory sensation sound is p r o duced by the impact on the e a r d r u m of a pressure wave p r o p a g a t e d in an elastic m e d i u m . . Thus, s o u n d results w h e n air or some other m e d i u m is set into motion. The f r e q u e n c y of a sound wave is the n u m b e r of recurrent waves p a s s i n g a c e r t a i n observ a t i o n point p e r second. The physiological sensation d e p e n d i n g mostly on frequency is pitch. Auditory p e r c eption of the i n t e n s i t y or energy contained in a sound wave is the loudness of s o u n d In o t h e r terms, the volume of sound is the physiological counterpart to t h e amplitude of the wave. * Sounds may be orga n i z e d into five main categories [71]. 1) "Pure" tones, 2) ordinary musical tones, 3) clangorous sounds, 4) ordinary noises, and ~ 5) white noise. T h e first category consists of pure sine waves which can be re ­p r e s e n t e d as a function of time as s (t) = Asin(2irft + y) where A is the amplitude, f the frequency and y the phase shift of the sine wave. The so u n d of a sine wave to a listener is dull and becomes tiresome after a w h i l e due to the exact repetition of the waveform. The second gr o u p of sounds in their steady state condition may be represented as 00 a m i x t u r e of sine tones represented b y a Fourier Series: S(t) = Z ' n=l A sin(27rnft + y ). Two interesting waveforms in this group are square n n wa v e s and sawtooth waves. A square wave contains only odd harmonics oo A 1 and can be represented as S(t) = Z - sin [ (2n-l) 2irft]. The result , 2n-l n=l of a square w a v e is a holl o w sound, m uch like that of a clarinet. A sa w tooth w ave contains harmonics wh o s e amplitudes are inversely pro- " n Al po r t i o n a l to their frequencies and is represented as S(t) = Z - sin2mrft. n=ln n A s a w t o o t h w a v e produces a b u z z i ng-like sound not very dissimilar from tha t of an English Horn. The third category of sounds consists of those h a v i n g i n h a r m o n i c partials and include sounds produced from bells, gongs and cymbals. The next g r o u p .is.comprised of any kinds of sounds occur-ing contiguously, and the final category is composed of sizzling, h i s s i n g and h o w l i n g n o i s e s in which"a±i audible frequencies occur at r a n d o m times. M u s i c is the auditory pe r c e p t i o n of sounds which h ave b e e n o r g a n ­ize d in some meaningful way. E l e c tronic music is that music which is produced, m o d i f i e d or combined with the aid of electronic equipment. The organ i z a t i o n a l procedure of el e c t r o n i c music can be divided into six h i g h l y - i n t e r a c t i n g areas: W a v e f o r m generation, modification, coordination, storage, control and notation. W a v e f o r m Generation. G e n e r a t i o n of sounds can be done wit h any device capable of p r o ­duci n g un d u l a t i n g signals, w hether they be in a vibrating medium, an el e c trical circuit or in the air. The electronic musical instruments de s c r i b e d in later chapters use variations of the following five types of t one sources: 1) Purely electrical, 2) rotational scanning, 3) vibratory, 4) electrical pickup of concrete sounds, 5) digital. Purely el e c t r i c a l generators include o s cillating arcs, relaxation oscillators, v a c u u m tube circuits wit h feedback, thyratron oscillators, c o n d e n s e r-inductance discharge, radio-frequency b e a t systems and tr a n s i s t o r o s c i l l a t o r circuits. Electrical circuits e mploying an amplifier are m a d e to oscillate by feeding b a c k a portion of the out­p u t voltage in a certain phase relationship into the input bias to overcome the a t tenuation losses of the circuit and thus effect the oscillation. V e r y stable vacu u m tube oscillators were h a r d to design b e c a u s e of the fluctuating characteristics of the tubes w i t h time. Thermal effects in the tubes as well as variations in anode voltages and cathode currents h a d drastic effects in changing the constants of a tuned circuit. The first tube or valve oscillators to generate a sine w a v e f o r m w e r e especially unstable. One of the pop u l a r vacuum tube oscillators finally developed was the Har t l e y Oscillator which h a d fairly goo d stability. Figure 1-1 Hartley Oscillator M o s t of the p o p u l a r electronic organ m a n ufacturers including Allen, Baldwin, Conn, Electrovoice, Gulbransen, and Lowrey use d this o s c i llator or a m o d i f i e d version thereof in their tone generators. To m a i ntain the degree of stability needed, a coil inductance having a fairly high Q was used in conjunction with the Hartley. Versions of this o s c i l l a t o r p r o d u c e d sine, sawtooth, and pulse waveforms. A very useful w a v e f o r m is the sawtooth since it contains essentially all the harmonics of the f u n d a m e n t a l . Both vacu u m tubes and gas tubes were used in sawtooth g e n e rating circuits. One of the simplest relaxation c onfigurations used a neon tube. ...... ' The neon tube discharges whe n a certain potential is applied across its electrodes, and it remains conducting until the voltage is reduced b e l o w a certain cutoff point. The w a v e f o r m is p r o duced by the capacitor charging up e x p onentially until the tube ignition voltage is reached, w h i c h discharges the capacitor almost immediately. The frequency is determined by the values of E, C and R. One of the problems with neon tubes is that their firing characteristics change wit h time. Usually the tubes h a d to be "broken in" until the responses became somewhat stabilized. Neon tubes hav e also b een employed in frequency division circuits where every o t h e r incoming pulse triggers the next stage. These circuits are desig n e d so that the oscillation frequency of each R E Figure 1-2 N e o n Tube Oscillator stage is a little slower than half the frequency of the preceding stage. W h e n the first pulse comes in, the capacitor in parallel with the neon tube charges ex t r a but not enough to fire the tube. By the . time the nex t pulse hits, the capacitor has near l y reached the ignition voltage and the extra surge of voltage triggers the tube, and the p r o ­cess starts over. Multivibrators h ave also b e e n used extensively in frequency d i v i d i n g networks. ' Wi t h the a v a i l ability of-cheap transistors, m o s t present day oscillators and dividers cire built from transistors. W h e n transistors w e r e first d e v e l o p e d and attempts w ere made to use t hem to replace t u b e s , much t rouble was encountered b ecause of the hig h power dissipation and low tempe r a t u r e b r e a k d o w n of germanium. Silicon transistors, m a n i f e s t i n g a h i g h e r resistance to heat, solved this problem. The obvious advantages of transistors are long life, low voltage require­ments, little risk of h u m pickup, and compact size. Integrated circuits (ic's) are n o w b e i n g used in some organ designs, b u t they are still a little more expe n s i v e than discrete components for tone generators and divider circuits. The ic's certainly p rovide m o r e convenient and compact packaging, and they facilitate much easi e r debugging. W hen a failure occurs, the ic package can be replaced as o pposed to the testing of many capacitors, resistors and transistors to find the trouble. Se m i c o n d u c t o r manufa c t u r e r s are now p roducing specialized integrated circuits for the e l e c t r o n i c organ industry [117, 96, 113]. Flip flops, for instance, are used frequently in o s c i llator and frequency divider circuits. An interesting w a y of producing wedrd sound effects is wit h the beating, or h e t e r o d y n i n g of two radio frequency oscillators. The high frequency oscillations are above the audio range, b u t w h e n two differ­ing frequencies are mixed, a resultant beat occurs w h i c h m a y be in the audio range. The s o u n d effects are created b y rapidly changing the frequency of one of the two r.f. oscillators. „ _ „ ........ ....... resultant beats f \ m ' Figure 1 - 3 The H e t e r odyning of Two Signals Electrical g e n e r a t i o n of oscillating signals is just one of the man y ways w h i c h hav e b e e n used. Rotary devices w i t h s c a nning pickups were used e x tensively in m a n y electronic music instruments. The types of pickup used were magnetic, electrostatic and photoelectric. The acoustic siren and the phonograph, of course, operate upon this principle also. Of the man y r o t a t i n g devices, only the H a m m o n d electromagnetic p i c k u p succeeded commercially. In these systems an i rregular shaped rotating conductor alters the magnetic field in a perm a n e n t magnet and induces a current in a coil wound around the magnet. The varying current is then ampl i f i e d and fed to the rest of the circuitry. In designing electro m a g n e t i c generators the effect of hysteresis h a d to be considered. In some designs the lag was h elpful in obtaining complex waveforms. Rotating electrostatic p i c k u p was accomplished by h a v i n g t w o discs facing each other, one of w h i c h rotated. Voltage was i m p r e s s e d across the two discs and as one rotated the va r y i n g capaci­tance s e t up an o s cillating voltage in the circuit. One disc contained m e t a l e t c hings of des i r e d waveforms, while the other h a d the scanning pickup. In some schemes the disc containing the w aveforms rotated and in ot h e r instances the scanning plate revolved. As the area between the sca n n e r and w a v e f o r m varies, so does the capacitance. The third t ype o f r o t a t i n g generator was photoelectric. One of the simplest p h o t o e l e c t r i c generators consisted of a w h e e l w i t h a series of holes in it w h i c h a llowed a p u l s a t i n g b e a m of light to fall on a photoelectric tube. P h o t o e l e c t r i c cells can produce a current p r o p o r t i o n a l to the i ntensity of light falling on the cell, and the.v a r y i n g current becomes the e l e c t r i c a l signal. In more sophisticated designs, a w a v e f o r m is e t c h e d on the disc, and as it rotates variable amounts of light appear at the photocell. Sound tracks on m o v i e films cure an example of this technique. R o t ating drums and records have also b een used w ith the d i f f e r e n t p i c k u p methods. V i b r a t o r s mak e up a third class of tone generators. There are so m any combinations i n volving v ibrating e l e m e n t s , exitation methods and p i c k u p schemes that it is probably easiest to see three separate tables and imagine h o w each of the vibrators can be p u t into v ibration by each of the exciters, etc. [102]. 11 V i b r a t i n g Elements Exitation hitting, plucking, blowing, bowing, twisting, resonance magnetic forces Pickup W i r e s , r e e d s , s t r i n g s , bells , m e t a l p l a t e s , tubes, bars, tuning f o r k s , sounding b o a r d s , m e m b r a n e s , springs electrostatic, electro­magnetic, photoelectric microphonic, piez o e l e c ­tric, interrupted con­tact, m o d u l a t e d resist­ance, thermoelectric Concrete sounds, or those sounds made by ordinary p h y sical objects, are a source for man y electronic music enthusiasts w o r k i n g w i t h classical tape studio equipment. The sounds range from those of a howling coyote, an ambulance siren, and-downtown traffic to the m o o i n g of a cow, the creaking of a door and the dripping of a leaky faucet. These sounds are generally r e c o r d e d on tape by microphone and then modified by the various techniques of the sound modification category. ' The digital computer has opened up whole new areas of sound generation and w a v e f o r m processing. Digital sound generators consist of p r o g r a m m e d routines w h i c h produce lists of numbers w h i c h are trans­formed to analog w a v e f o r m s by a digital-to-analog (D/A) converter and a low pass filter. The numbers generated by the computer are samples of the w a v e f o r m at certain time intervals. The numb e r of samples per s e c o n d determine the h i g h e s t frequency possible in the so u n d generated. It turns out that the numb e r of samples must be 2n/sec wh e r e n is the h i g h e s t frequency. The digital samples are converted to voltage pulses b y D / A converters and these pulses are smoothed out by the filter to p ro d u c e the final a n alog signal. The generation of say 30,000 samples per second for a frequency response of 15,000 cycles p e r second takes some time on the computer, especially if the s y n t h e s i z e d w a v eform is complex. U sually the w a v e f o r m samples are computed and stored in an array to be later output to the D / A equipment. Anywhere from 2 seconds to several seconds may be re­q u i r e d to compute one secondfe worth of sound. The trend is n o w to store d i g i t a l samples of one p e r i o d of certain cyclic waveforms in tables -in the computer. A limited n u m b e r of samples are stored, and i-f-a sample value is need e d for a time n o t represented in the table, simple in t e r p o l a t i o n is done. The s t o r e d waveforms can be combined in v a r i o u s w a y s as I shall esqplain- later and the resultant sound can be p r o d u c e d in real time if a low frequency response is permitted. W a v e f o r m Modification. T h e seco n d category of electronic m u s i c production is the modifi-. cation of the g e nerated waveform. Th e r e are two ma j o r w ays to produce d i s t i n c t i v e wave f o r m s w h i c h may be further m o d ified by other means. The two met h o d s are additive and subtractive synthesis. Additive synt h e s i s consists of m i x i n g pure tones (sine waves) together to p r o ­d uce com p l e x w aveforms and subtractive synthesis consists of filtering o u t certain frequencies from a complex w a v e f o r m already containing a large s p e c t r u m of frequencies. In additive synthesis, the fundamental is c o m b i n e d w ith various h a r m o n i c s , p a r tials and overtones having d i f f e r e n t amplitudes to synthesize d e s i r e d sounds. An a l y s i s of steady state m usical tones have shown that distinctive tone colors m a n i f e s t a certain h a r m o n i c content. F o r example, a flute s o u n d consists of a d o m inant fundamental wit h a little b i t of 2nd and 3rd harmonics. V i olin sounds on the o t h e r h a n d have'large contributions fro m all of the first eight or nine harmonics, w i t h emphasis on the 12 fourth and fifth. Harmonics are multiples of the fundamental frequen­cy. In some e l e c tronic organs where additive synthesis is done (Hammond, for example) the tones used to form the result are e q u i temper­ed and thus the introduced harmonics are not exact multiples of the fundamental. The subsequent sound p r o duced is distinctive and sounds artificial. Additive synthesis is u sed in some e l e c tronic organs to p r o d u c e chiff, .the starting transients accompanying the speaking of an or g a n pipe. Here at the outs e t of the tone for a few milliseconds, certain h igh partials and usually a b i t of noise are added to the tone. Very r ealistic imitations of actual organ pipe sounds hav e been d e v e l o p e d by a n u mber of electronic organ manufacturers, p rimarily Saville, Al l e n -and Rodgers. . .T ~ The other major technique of tone coloring is subtractive synthesis. In this technique the filtering is done in a variety of ways, for i n ­stance, resonance and formant filtering, high-, low- and b a nd-pass f iltering and b a n d exclusion. An example of resonant filtering is the disti n c t i v e sounds p r o d u c e d b y differently shaped organ pipes. Here the natural resonance of the pipes impart an emphasis to certain fre­quencies and thus color the tone. Variously contoured speakers have be e n used in some e l e c tronic musical instruments to alter the tone qu a l i t y of the music produced. F ormant filtering is the electrical counterpart of resonance filtering. The formant theory was d eveloped by Hermann in Germany. This theory was that vocal timbres are produced by the effects of the throat, mouth and nasal cavities on the vocal pitch p r o d u c e d by the v i b r a t i n g vocal chords. The p u lsating air stream from the lungs through 13 . . the vo c a l chords set the several r e s onant cavities into damped oscillation, w h i c h may be inharmonic to the frequency of the vocal chords. H e r m a n n called these effects formants [31]. Electrical formant filters are resonant circuits w h i c h impart an emphasis region to cer t a i n frequencies. These filters o nly m o d i f y the e x i sting harmonics already in the waveform; they can't c o m p letely eliminate any or add any. The n a t u r e of m ost sounds depends not onl y on the harmonic con­tent, b u t a lso upon the spectrum or f requency response in time. F o r ­m a n t filters are generally applied to complex waveforms such as sawtooths w h i c h h a v e rich h a r m o n i c content. Formant Filtering Formants then, as shown in Figure 1-4, are frequency ranges in which ha r m o n i c components are predominant relative to neighboring harmonics. A p a r t i c u l a r tone quality may have several p e a k s , or formants along its freq u e n c y response curve. Formant filters usually consist of a p assive b a n d pass filter w i t h an amplifier w ith positive feedback. W h e n the gain exceeds the attenuation in the circuit,'it alternates at a resonant frequency. Waveforms sent through such a circuit receive a r e i n f o r c e m e n t of those frequencies at which the filter is oscillating. Passive filters are extensively vised to impart tone colors to complex tones. For example, a low pass filter produces an organ flute sound composed of m o stly low harmonics. The general classes of passive filters are d e p icted in Figure 1-5. . -------- WTT3--- ____ :___ I low pass 15 b a n d pass Figure 1-5 Passive Filters A pas s i v e filter is one w h i c h contains only r e s i s t o r s , capacitors and inductors, w hereas an active filter contains an amplifier. Some organ m a n ufacturers are now using active filters, which may be made to resonate along w i t h passive filters to imitate pipe organ sounds. The p r o b l e m w ith band-pass and resonant circuits is that they filter low notes and high notes differently. The trend now is to have many b a n d responsive filters for a p a r t i c u l a r tone color over the entire compass of the pitches. The rate of attack and length of decay of a sound are important pa r a m e t e r s wh i c h determine tone quality. A violin tone w ith a p e r ­cussive attack and gradual decay sounds like a piano, and a piano -tr yD - T T I b a n d elimination high pass tone given a time envelope of a v i olin tone sounds like a violin. Most i nstruments h a v e exponential attacks and decays as shown in Figure 1-6. A ( t ) = A ( l - e " k t ) Figure 1-6 Exponential Attack and Decay El e c t r o n i c m u s i c instruments simulate attack and decay by resistor-capacitor (R-C) circuits in which the charging and discharging time of a capacitor is d e p e n d e n t upon the values of the capacitor and the re ­sistor. P e r c u s s i v e sounds have been imitated very well with electronic circuits h a v i n g sharp attack times and long decays. There are many commercial e l e c t r o n i c b a n d boxes which c ontain pe r c u s s i o n sounds rang­ing from a b a s s d r u m to a cymbal crash. I n teresting sounds are obtained b y applying a p e r c u s s i v e formant to d ifferent waveforms, including noise. The gra d u a l attack and onset transients are very important in d eter m i n i n g certain Baroque pipe organ sounds. As mentioned earlier, these effects h a v e b een simulated very well. E l e c t r o n i c devices which impart time envelopes to any waveshapes put through t h e m also control the shape of the attack and decay curves as w e l l as the lengths. The computer generation of time envelopes is especially v e r s a t i l e since any curve can be d e f i n e d as the attack and decay. For instance, linear attacks and decays impart a science- fiction character to the sounds; whereas exponential curves give the sounds an air of familiarity. People w h o produce electronic music by m a g n e t i c tape man i p u l a t i o n often cut out the por t i o n of the sound ,. containing the attack. Sometimes very interesting and unusual results occur. A n o t h e r favorite technique is to splice the attack or decay of one sound onto another sound. Tape recorders are used in a variety of ways to m o d i f y sounds. F o r instance, a noise can b e recorded at one speed and p l a y e d b a c k at another, and this can be repeated ad infinitum. A p o p u l a r trick is to p l a y a recorded sound b eckwards either at the r e c orded speed, or faster or slower. The pic k up volume can be continually varied b y h a n d as recording takes place, or.the tape can be mad e to accelerate o r decelerate quickly b y yan k i n g it one way or another. The feeding b a c k of an output signal to the recording head p roduces a type of reverberation if the returned signal is attenuated. Sometimes, however, the feedback is amplified and the result is a shrieking how l like that p r o d u c e d in a microphone speaker feedback loop. If y o u h ave ever wonde r e d w h y some "electronic music" sounds the way it does you only n e e d imagine some of the above-mentioned shenanigans taking place. It mus t be conceded, however, that out of some of the bizarre, s o u n d - d istortion techniques much interesting mu s i c has emerged. There are other ways besides tape recorder playback loops in which artificial r e v erberation is created. Reverberation is the repeated e choing of a sound at subdued levels. It is the effect of sound waves arriving at a listener's ear at different times due to environment in w h i c h the sound is p r o d u c e d and heard. For example, in an auditorium 17 where live m u s i c is p e r f o r m e d the sound waves prop a g a t e in man y directions and are reflected and absorbed b y the surrounding objects - and finally reach the listener at d ifferent times and at various in­tensities. This ov e r a l l effect is called reverberation, and it gives s o u n d the q u a l i t y of "bigness." Thus it is i n t e r e s t i n g to simulate -• reverb e r a t i o n artificially. There are five b a s i c delay techniques u s e d to pro d u c e reverberation. They are a c o u s t i c d e l a y lines, m a gnetostrictive de l a y lines, electronic delay networks, m a g n e t i c tape loops a n d d igital delays. Acoustic de ­lays are p r o d u c e d b y the finite velocity of s o u n d energy in air or ot h e r e l a s t i c medium. The air chamber may range from a h o l l o w tube to a large r o o m or auditorium. The usual procedure' is to place a loud­speaker at one e n d of the chamber and several microphones at different p o ints to p i c k up various delayed s o u n d s i g n a l s , w h i c h are m i x e d to ­gether to p r o d u c e the reverberated signal. The propagations of vibrations in spring coils and m e t a l plates also constitute acoustic delay lines. The el e c t r i c a l s i gnal is converted by a t r a n s d u c e r to a mechanical vibration w h i c h is induced in the v ibrating coil or plate. Variously p l a c e d trans d u c e r s r econvert the vibrational ener g y to electrical signals wh i c h are c o m b i n e d to effect the reverberated signal. Acoustical delay lines have the d r a w b a c k that they distort the signal appreciably, especially in terms of frequency response. M a g n e t o s t r i c t i v e delay lines are formed from magnetostrictive materials w h i c h h ave the property that a m a g n e t i c field applied to one end of the m a t e r i a l induces a wave that p r o p agates to the other end [125]. The w ave p r o p a g a t i o n m a y be either longitudinal o r torsional. The 18 advantage of this type of signal propagation is that very little distortion of the w a v e results, even with a limited number of intermediate signal taps along the wire. To achieve these results wit h longitudinal propagation, however, the wire or b a r must be straight; otherwise, serious phase distortion results. Torsional w ave p r o p a g a t i o n provides longer delays per length of wire and the wire may also be coiled w i t h ­out affecting the p h a s e of the signal, so high frequencies m a y be p r o p a g a t e d w i t h o u t distortion. The third type of delay line is purely electronic and is produced b y cascading resistor, i n d u c t a n c e a n d capacitance delay circuits. The only p r o b l e m with this type of delay is that some 300 circuits are re­q u i r e d to give a delay of 50 m s . , thus the amount of hardw a r e becomes p r o h i b i t i v e . M a g netic tape loops are capable of producing any amount of delay b y the plac e m e n t of several playback heads with feedback at different intervals from the reco r d i n g head. Schober Organ Company advertises their Reverbatape, consisting of three playback heads, as b e i n g capable of prod u c i n g the s o u n d of a large auditorium to that of a small hall. A g ood tape reverberator can produce acoustical effects on some sounds indistinguishable from the real thing. The advantage of tape rever­b e r a t i o n is that the length of the delays and attenuation of the signals are controllable and it can have full audio frequency response with low distortion. One p r o b l e m w i t h the Reverbatape is tha t short, staccato notes are "bounced" back repeatedly instead of be i n g smoot h l y attenuated. The final m e t h o d of reverberation is digital. ' The processing of digital waveforms in a computer facilitates the simulation of reverberation • • 19 by programmable control. Mathematical models with iterative techniques (analogous to feedback loops) have been programmed to produce digital reverberation at the University of Illinois [66]. The trouble with digital reverberation is that it is not done in real time. The analog signal must be converted to digital samples, be processed and reconvert­ed to analog. A good system of reverberation at the University of Illinois. Experimental Music Studio uses a combination of both magnet-ostrictive delay lines and tape loops [125]. V i b r a t o is an o t h e r modi f i c a t i o n wh i c h can b e applied to give wa rmth to certain types o f music. Vibrato is a slow u n d u lation of pi t c h and ampl i t u d e of a s o u n d at about five to nine times a second. .' A trumpet p l a y e r p r o duces vibrato by slowly v a r y i n g the tension of his lips or by s l i ghtly a l t e r i n g the pressure of the m o u t h p i e c e on his lips. Both of these actions cause v ariation in air pressure and thus a slight w a v e r i n g of tone. In a similar m a n n e r puls a t i n g variations of air p r e s s u r e cause p i p e organ vibrato. In these cases a variation o f both p i t c h and a m plitude cause the vibrato. A viol i n p l a y e r utilizes mainly p i t c h v a r i a t i o n as he slides h i s finger b a c k and forth to p r o ­duce vibrato. Some e l e c t r o n i c organs vary the gain of an ampl i f i e r to produce a vibrato w h i c h is really an amplitude modulation. A true vibrato consists of both frequency and amplitude modulation, with the emphasis being on frequency variation. P e r i o d i c amplitude v a riation is correctly termed t remolo or t r e m u l a n t and is used instead of vi b r a t o oh some electronic musical instruments. T r e m o l o is p r o d u c e d very easily in electronic instruments by adding a s l owly oscillating signal to the bias of an 20 amplifier. In p h o t o e l e c t r i c instruments tremolo is caused by varying the amount of light falling on a photocell by moving a g r aded trans­p a r e n t slide back and forth in the light beam. The volume of capacitive p i c k u p generators can be v a r i e d by increasing and decreasing the d i s ­tance bet w e e n the plates. Tremolo is fairly easy to produce, but vibrato is a little more difficult. In many electronic generators, a slow oscillation is applied to the tone generators to pro d u c e vibrato. This works p r e t t y w e l l for i ndividual oscillators, b u t in a system w ith frequency dividers the vib r a t o does not p ropagate through. In the case of many individual tone generators several rates of vibrato should be used for different p i t c h ranges. A normal vibrato, w h e n applied to a low note of, say, less than 100 cycles p e r second, sounds bad because the percentage of vari a t i o n on low notes represents a significant change in pitch. Generally, vibrato is n o t imparted to the low pedal notes of an organ, b u t is saved for the m a n u a l tones. Higher p itched notes usually need a w i d e r and faster vib r a t o than the lower o n e s . V ibrato has b e e n p r o d u c e d in a number of ingenious ways on the v ariety of electronic musical instru m e n t s . In those cases w h e r e me c h a n i c a l rotating elements are used, a slight variation in speed p r o ­duces a good vibrato. An easy way to cause this v ariation is to use ecce n t r i c pulleys on the b e l t drive as depicted in Figure 1-7. The 21 ■ Figure 1-7 Vibrato A p p l i e d to Rotating Generators m o t i o n of the outside wheels is back and forth in opposite directions, aga i n s t and away from the b e l t drive. In rotating capacitance p i c k ­up g e n e rators wit h o p p osing disks, the stator can be slightly jiggled to and fro to produce vibrato. An i n t e r e s t i n g w a y of producing vibrato electrically was done by H a m m o n d w i t h a series of time delay networks. The signal is tapped o f f at d i f f e r e n t places along the delay line, and the phase differences of the c o m b i n e d signals p r o duces a vibrato. Schober Organ Company uses a p h a s e sh i f t network to electrically create vibrato. Phase shifting p r o d u c e s b oth amplitude and frequency modulation, similar to the doppler effect. U s u a l l y vibrato is applied onl y at one place for all the notes or else the same vibrato oscillator is used to create all the vibration. In b o t h these cases the tones all v ibrate together and are locked in phase. The r e s u l t is an unnatural so u n d w i t h o u t m uch warmth and depth. Some or g a n manufa c t u r e r s hav e overcome this p r o b l e m b y applying differ­ent rates of vibrato to various tones which make up the final sound. M a n y of the mo n o p h o n i c or melodic instruments were equipped with touch control vibrato. F o r example, the T r a u tonium had a liquid re­sist o r in the form of a long tube which was depressed by the keys to f orm gradations of volume. The lateral m o v e m e n t of the dummy rubber k e y de t e r m i n e d the pitch, so slight undulating movements of the fingers on the keys of the Trautonium produced a very controllable vibrato. Of course, the m o s t sophisticated and precisely controlled v ib r a t o can be pro g r a m m e d and p r o duced digitally, b u t additional time m u s t be s a c r i f i c e d since each wave form sample mus t b e modified. 22 A s t a n d a r d piece of equipment for many organ manufacturers is a r o t a t i n g s peaker which produces a vibrato effect. The large w o o f e r u s u a l l y is h e l d fixed while the smaller middle and high range speakers are rotated. The vibrato is a result of the doppler effect which is tha t the apparent frequency of a so u n d is i n c r e a s e d as the sound source approaches the listener, and the pitch drops as the sound source moves away. A variation of rotating speakers is rotating baffles into which the s o u n d is directed. Here again the virtual source of the sound is v a r i e d as the rotating b a f f l e spreads the sound in different directions. M o d i f y i n g electronic sounds w i t h vibrato is one way o f approaching the natur a l n e s s of some sounds, b u t it? is not sufficient in imitating pipe organ choruses. Sust a i n e d organ tones interact in a complex way, p r o d u c i n g a warm, indefinite m o v e m e n t called the choir effect. This r e sult can be arrived at e l e c t ronically b y the introduction o f noise to randomly control the tone oscillators. This is not the same thing as a d ding a little noise to the actual tone. The idea is rather to intro­duce r a n d o m variation in the oscillators. A similar effect can be o b t a i n e d by randomly affecting various vibrato oscillators, thus in t r o ­d u c i n g different o s cillation patterns among the many interacting tones. Gross modifications of pitch and intensity of sound are easily o b t a i n e d w ith electronic equipment. Wide variations of amplitude p r o ­d uce throbbing effects, and fluctuations in pitch create w a - w a and H a w i i a n guit a r results. A continuous gliding of pitch, or portamento, can be p r o d u c e d directly on some electronic musical instruments such as the Theremin and M o o g and portamento can also be p r o d u c e d by m o d ­i f ying discrete tone melod i e s by sending them through an integrator 23 followed b y a s moothing low pass filter. R i n g m o d u l a t o r s and multipliers are p o p u l a r devices for tone modification. The essential working of a ring m odulator is the m o d u l a t i o n of one signal b y another. The ring modu l a t o r accepts two signal inputs and produces their sum and difference as outputs, wh i c h . usually s o u n d e n t irely different from b oth inputs. Multiplicative m o d u l a t i o n of one signal b y another can also be accomplished. Ve r y s i g n i f i c a n t wor k in sound m o d i f i c a t i o n has been done with M a x M a t h e w ' s M U S I C IV b y J.C. Risset at the Bell Telephone Laboratories [123]. It is debatable whe t h e r this w o r k s h o u l d be classified under sound ge n e r a t i o n or sound modification because the computer generates the final samples directly. I have chos e n to include Risset's cata- ' logue of c o m p u t e r synthesized sounds in the p r e s e n t area of modification bec a u s e the c o m p u t e r programs mainly m a n i pulate arrays of numbers re­p r e s e n t i n g simple periodic waveforms and produce complex waveforms. All types of sound modifications can b e done very precisely b y digital w a v e f o r m processing. Attack and decay envelopes may be defined, r a n ­domly v a r y i n g or perio d i c vibrato and tre m o l o m a y be applied, additive synthesis or subtractive synthesis may be accomplished, all types of conce i v a b l e filters may be constructed, and thus unlimited timbres may be created. R i s s e t has published a catalogue of computer synthesized sounds, some of w h i c h are new and bizarre. He has produced timbres that hav e the tone qualities of f l u t e s , c l a r i n e t s , brass i n s t r u m e n t s , pianos, and pe r c u s s i v e instruments including bass drums, bongos, snare drums, and gongs. The simulation of a gong is especially interesting bec a u s e r a n d o m control has enabled the sound to h a v e natural waverings 24 ■ as are h e a r d from an actual gong due to the h i ghly interacting waves in the me t a l that determine the different vibration modes all over the gong. N e w and unusual sounds created by Risset include those made by e x p e r i m e n t i n g w i t h various shapes of attack and decay curves, different envelopes for h a r monic components of a sound and b i - d i r e c t i o n a l ■ glissandi. Two interesting features of this catalogue are that first it comes w i t h records which contain examples of the sounds, and second the descriptions of the waveforms a n d modification parameters are d e s c r i b e d completely for each sound. 'Hiis information is of course ver y valuable to anyone w i s h i n g to synthesize various s o u n d s , whether elect r o n i c a l l y or digitally. - Sound Coordination and Organization. The third ma j o r area of sound pr o c e s s i n g is the c ombining and m i x i n g of sounds. This area is concerned w i t h h o w various tones, timbres and melodies are organized and put together. There are three aspects of this area, which are rhythmic, synthetic, and spacial. The rhythmic aspect deals with the organization of the pitch structure of a sound. This topic includes the assignment of durations to the individual tones. Usually this is done directly by the p e r ­former on an electronic m usical instrument; as he plays melodies he a utoma t i c a l l y determines the rhythmic content. In the p r o d u c t i o n of el e c t r o n i c music wit h tape recorders, the durations of notes are alter­ed by diff e r e n t tape recorder speeds. There are recorders wh i c h will allow the speed-up of sound w i t h o u t the usual increase in frequency by rotating playback heads. The frequency of the playback is dependent 25 upon the r e l a t i v e speeds of the tape and the head. As long as the same ratio is m a i n t a i n e d for the playback hea d and tape speeds as was present b e t w e e n the r e c o r d h e a d and tape speeds at r e c o r d time, the frequency wil l remain t h e same. Conversely, if the tape speed is not increased b u t the r o t a t i n g h e a d is mad e to rotate, the r e s u l t will b e an increase of pitch b u t n o t of duration. There are v arious devices wh i c h control the durations of notes and these instruments w i l l be described under the control area. Suffice it to say, very complicated and sophisticated rhythmic p a t t e r n s can b e imparted to melodies. The s e c o n d aspect of the organization area is the synthesis of a po l y p h o n i c line of music from two or mor e melodies, which may be either mo n o p h o n i c or polyphonic. Again, this is d o n e-directly by a performer as he plays a p o l y p h o n i c instrument. The usual p rocedure in electronic mu s i c studios is to use tape recorders to combine melodies. Especially if the compo s e r has mo n o p h o n i c instruments to w o r k with, he wil l mix many m e l odies t o g e t h e r b y recording and r e - r e cording several times. An example of this technique was the prepa r a t i o n of Walt e r Carlos' selec­tions on the r e c o r d "Switched On Bach." The p i eces were prepared vising a M o o g Synthesizer, w h i c h is essentially a m o n o p h o n i c instrument, as follows: first, great care and time was taken to set up the filters and envelope shapers so that a particular tone q u a l i t y was produced (for instance, the sound of a harpsichord) then a two or four measure m o n o ­phon i c p assage was p l a y e d by Carlos on the keyboard. The notes were p l a y e d an o c t a v e b e l o w the final pitch range and recorded at half the playb a c k speed. N o t o nly did the monophonic passages have to be con­tiguously connected, b u t other monophonic melod i e s h a d to be mixed to pro d u c e the p o l y p h o n i c mucic. The effort took approximately a year w ith the crude manual tape splicing techniques. The m i x i n g of melodies is very easy if the music is stored in • digital form, or if the sound-production controls are stored digitally. In bot h cases definite and exact starting places may be determined for the synchronization and m i xing of melodies. For instance, melodies formed on the RCA Synthesizer are defined by digital control information p u n c h e d on a wid e p a p e r tape. Not o nly can an exact starting place be found, b u t the speed of the paper tape can be v a ried and thus the tempo. The first model of the RCA Synthesizer was driven \n synchronization w i t h a d i s k recorder w h i c h was able to record on several tracks simul­taneously. Melodies wer e m i x e d very simply b y 'recording'one on one ' track, another on a separate track, and playing the m back simultaneously. The later R C A Synthesizer, Mar k II, is now located at the Columbia Un i v e r s i t y Electronic Mu s i c Studio, and it is synchronized w ith a m u l t i p l e - c h a n n e l A m p e x tape recorder; in fact, the recorder was the first m u l tiple channel tape recorder that Ampex made. Synchronization b e t w e e n the synthesizer and the recorder is done by two signals which are p u t on the tape. The recorder is started up a few inches before the synchron i z a t i o n clicks so that the tape speed is m a x i m u m when the first signal is sensed. This signal turns on the synthesizer tape w h i c h has b e e n p r e p o s i t i o n e d to the desired starting place. The second click turns on the r e cording head on the specified channel to start the actual recording after the synthesizer tape has reached its normal speed. T w o tape recorders can be synchronized by the same method, and this is much m ore precise and easier than the manual techniques for synchronizing 27 tapes. W h i l e v i s i t i n g the Columbia Electronic M u s i c Studio I observed h o w Blilent A r e l m i x e d electronic music passages f rom two tapes onto a third. He first l i s t e n e d to the two tape selections and d ecided h o w m uc h of one he w a n t e d to m i x w i t h the other. T o determine where to cut the tapes h e m a n u a l l y s l i d the critical por t i o n b a c k and forth under a p l a yback h e a d a n d l i s t e n e d carefully to find a g o o d b r e a k i n g point. Af t e r h e h a d done this w i t h b o t h tapes he f o rmed two loops wit h the sections of tape h a v i n g the desired passages on them. T hen he ran each loop through a d i f f e r e n t tape recorder and l i s t e n e d to the results. He v a r i e d the vol u m e s of the individual recorders at the m i x i n g panel then he v a r i e d the speeds of one then the other loop, always listening, I supposed, for i n t e r e s t i n g combinations of sound.. He h a d m e a s u r e d the two tape lengths a n d m a d e sure they were n o t in a simple ratio of each other so that the sounds from each tape w o u l d i n t e r a c t differently as the loops c o n t i n u o u s l y played. Finally he d e c i d e d that a certain combination was w o r t h recording so he started up a third recorder from the mixing p a n e l a n d r e c o r d e d the resultant m i x i n g of the two tape loops for some time. I h a v e cited this as an example of some of the classical studio techniques tha t are carried on in p r o d u c i n g electronic music by mixi n g various r e c o r d e d sounds. The tape loop routine is very common­place and w a s p r o b a b l y introduced in Am e r i c a b y V l a d i m i r Ussachevsky, w h o was one of the p i o n e e r s wit h Otto Luening to b r i n g electronic music to the U n i t e d States [90]. Most electronic m u s i c has bee n produced by manual tape m a n i p u l a t i o n s that include the splic i n g of different sounds together and the r e c o r d i n g of one sound on top of another. The record­ing of one piece of r e c o r d e d sound on another has b e e n done in a wide 28 var i e t y of ways such as the variation of their relative s p e e d s , the chang i n g direction of one or both of the passages (playing the tapes b a c k w a r d s ) , the b l e n d i n g of one with the reverberated sound of the other, and the vari a t i o n of their individual volumes [147], ^ The electronic m u s i c composer usually works entirely with t a p e s , synthesizers and other electronic devices to produce the mu s i c to be performed. V l a d i m i r Ussac h e v s k y introduced the c o mbination of tape and live m u s ician per f o r m a n c e s w h i c h were w ell received. The synthesis of the final pr o d u c t i o n can be achieved by four means of supplying cues to the live performers: 1) each perf o r m e r can pla y at r a n d o m w h a t he wa n t s from the score and h o w fast he feels like p l a y i n g it; 2) cues are s u p plied at p e r i o d i c intervals; 3) s p ecial-synchronization cues come to the conductor or performers w h o are w e a r i n g earphones; and 4) each p l a y e r receives cues, b u t h e decides h o w he w i l l contribute to the mu s i c a l effect [149]. The third aspect of the organization and combining of sounds is spacial. The difference bet w e e n hig h fidelity and stereo illustrates an effect caused by the spacial m i x i n g of sounds. The p l a c e m e n t of sound sources in relation to the listener has a great deal to do w i t h the character of sound perceived. The size, shape, and m a k e u p of a room h a v e much to do w i t h the final sound effect from speaker outputs. The n u m b e r of sound sources (speakers, for example) their frequency response, and the directions they face is very important, as can be seen w ith B ose speakers, wh i c h r e produce recorded sounds so well. These speakers are composed of several small loud speakers wh i c h are d i r e c t e d at diff e r e n t angles and are usually p l a c e d facing a wall so that the 29 sounds f r o m the speakers are further m i x e d and r eflected against the h a r d surface. W i t h this arrangement the entire sound fills the room and the "stereo" e f fect is heard around the room. In the ordinary stereo s i t u a t i o n w i t h onl y two speakers, there is a limited range where, the ster e o e f f e c t is prominent. Various p r o d u c t i o n s of electronic m u s i c have u t i lized m a n y speakers p o s i t i o n e d all around the audience. Some speakers are p l a c e d in front, others on the two sides and behind, and still others above the listeners. Control pa n e l settings determine w h i c h sounds are d i r e c t e d to each speaker system. The a u d itory output can be mad e to s o u n d as though it w e r e m o v i n g from b e h i n d the audience to the front o r vice v e r s a ; it can sound as if it w e r e rotating around the r o o m one w a y and then the other, or it can appear to tie coming f rom all over. An impo r t a n t idea is t hat the directions, flow, and sources of the sounds can be electronically controlled. Interesting effects are c a u s e d b y sending one m e l o d y around the audience in one d ire c t i o n a n d a nother m e l o d i c line in the other direction, or in the same dire c t i o n e i t h e r lagging or leading, or trav e l i n g at a different speed. The spacial effects of sounds produce n e w m i x tures and com­bina t i o n s w h i c h add v a r i e t y and interest to electronic m u s i c 185]. W a v e f o r m Storage. The fourth area of electronic m u s i c pr o d u c t i o n is the storage of s ound materials. The obvious and most wide l y used m e d i a are magnetic tape and r e cord d i s k s . Storage is important becuase it facilitates the saving of certain melodies and sounds, w h i c h m a y*be later mixed w i t h o t h e r sounds; and it makes it possible to serially process some 30 sounds at different time periods. In a certain sense, all the tone generators are storage devices that yield up their sounds u p o n request. Pa p e r tape, long sheets of punched paper, thin magnetic wi r e s and . computer memory have all b e e n used to store music. C o m puter memory .. includes core memory, i n t e grated circuits, rotating m a g n e t i c disks and drums, magnetic tapes, paper tape, and punched cards. Mu s i c is r e p r e s e n t e d digitally on these mediums in four ways: anal o g w a v e f o r m samples, analog control information, digital control data, and algorithms. U su a l l y a combination of these ways is used. For example, in M a x Mathew's M U S I C V and GROOVE progr a m s w a v e f o r m samples are gene r a t e d b y algorithms w h i c h store the i n formation in tables, so a combination of the first and fourth storage methods is used. In a s y stem at the Un i v e r s i t y of T o r o nto,digital data representing analog control infor m a t i o n is con­v e r t e d to analog voltages w h i c h control v o l t a g e -controlled filters and oscillators. Our m u s i c s y s t e m at the University of Utah is an example o f stored information used as digital control. The digital data stored in the computer is sent directly to a matrix of latches w h i c h control the switching of analog tone signals. Core and i n t e grated circuit memories are now be i n g u s e d by the Rodgers and Saville organ manufacturers to store s t o p - a nd-coupler settings. By this means a combination button can be p r o g r a m m e d to ' determine a certain state of the organ stop settings and couplers. W h e n the b u t t o n is t h e r e a f t e r pressed, the appropriate state is invoked. If another organ i s t wants the same button to specify a d i f f e r e n t state, he simply reprograms the button, p u t t i n g the new state i n formation in 31 the memory. Subsequent depression of the combination butt o n causes the s t ored inform a t i o n to set the n e w state of the organ. S o u n d Control. The fifth area of electronic m u s i c pr o d u c t i o n is control. This area is m o s t i mportant b ecause it d i r ectly determines the interaction - o f the four previous areas; namely, the generation, modification, mixing, and sto r a g e o f sound. Control occurs in levels. For example, the h a n d can control capacitance which determines voltage l e v e l s , and the v o l tages can control oscillators, filters, amplifiers, and so forth. The d i f f e r e n t types o f controls interact wit h each other in diverse ways. Th e r e are four types of control: manual, electronic, stored d ata and algorithmic. ''1 •' ' - M a n u a l control consists of the m a n i p u l a t i o n of sounds b y persons b y means of keyboards, linear controllers, pa t c h panels, m i x e r panel knobs, touch sensitive devices, velocity sensors, stop tablets and cou p l e r switches. These devices are activated by movements of a p e r ­s o n 's hands, fingers, and feet. On the Theremin the positions of the two hands direc t l y control the pitch and volume of the resulting sound. M o d i f i c a t i o n s o f this idea allow the posit i o n o f the hand or finger on a w i r e or tube to control b oth pitch and volume. The pressure sensitive control tube on the later models of the T r a u t o n i u m is an example of p l a c e m e n t and force exerted by the h a n d in controlling the audio output. The M o o g equipment includes a linear controller which is a wire s t r e t c h e d o v e r a plate of resistance. The p o s i t i o n a.t which the wire touches the resistive plate determines a control voltage. The accom­p a n y i n g or g a n - l i k e k e y b o a r d works on the same principle; that is, the 32 d e p r e s s i o n of a key defines an o u tput voltage. The circuitry is a r r a n g ­e d in a serial fashion along the keyboard, so only one control voltage is d e t e r m i n e d even w hen a n u mber of keys are simultaneously depressed. E l e c t r o n i c organ keyboards, then, are used to control either m o n o p h o n i c or po l y p h o n i c instruments. There have been man y keyboards d e s i g n e d for different purposes. F o r example, the leather bands wit h t r a n s v e r s e welts on them wer e used as keyboards on various European s i n g l e - m e l o d y instruments (Chapter 2). The leather bands wer e p l a c e d o v e r s t r e t c h e d wires that could be p r e s s e d against resistance plates. The welts along the leather bands indicated the positions of a p r e ­d e t e r m i n e d scale (the we l t s could be s l i d to different p ositions along the bands). Keyboards hav e b e e n d e s i g n e d for .different Scales h a ving m o r e notes than twelve. Various scales have bee n proposed; for example, the P y t h a g o r e a n scale was b u i l t u p o n pe r f e c t fifths/ b u t it did not p r i m a r i l y contain octaves. 33 Name C D E F G A B C' C" J u s t Ra t i o to C 1 9/8 5/4 4/3 3/2 5/3 05/8 2 4 T e m e r e d Ratio to C 1 1.122 1.260 1335 1.498 1682 1.889 2 4 P y t h a g o r e a n Ratio to C 1 9/8 81/64 4/3 3/2 27/16 243028 2 2187/512 Figure 1-8 Scale Intervals The last number in the Pythag o r e a n row of Figure 1-8 is approximately 4.3, quite a b i t d i f f e r e n t from an octave. The just scale is interesting in t hat it has two diff e r e n t whole tone intervals. When the ratios b e ­tw e e n consecutive notes are noted, we see that E-F and B-C are in the ratio 16/15; D-E and G-A have 10/9; and C-D, F-G, and A-B all have 9/8. The 16/15 ratio represents a semitone, and the 10/9 and 9/8 are the two wh o l e tone intervals. Th e s e last two ratios differ by 81/80, w h i c h is called syntonic comma. If the difference of the comma is ignored, a just scale can be bu i l t up w i t h 22 notes (2 different . # V notes for C and D , etc.). An even b e t t e r approximation can be acc o m p l i s h e d w i t h 50 steps per octave, and the n ext improvement comes w i t h a r o u n d 300 steps [30]. It turns out that just intonation cannot b e maintained in all keys w i t h o u t unmanageable complexity. The introduction of sharps and flats makes things even more d i f f i c u l t be c a u s e theoretically there are m a n y kinds of F-sharps, etc.; and the F-sharps are not'equal to ' the G-flats. J u s t intonation can be approximated fairly w e l l with two equ a l l y t e m p e r e d scales, one w i t h A=440 and the other w i t h A=436.3. The i dea is to switch from one scale to the ot h e r depending on the key b e i n g played. The t e m p e r e d scale is constructed by d i v iding cin octave into twelve equal parts and by spacing the notes such that each one is - /2 times the frequency of the p r e c e d i n g note. This construction, as can be see n from Figure 1-8, approximates the fourth and fifth intervals of 4/3 and 3/2 fairly closely. For instance, the perfect fifth ratio of 1.5:1 is approximated by 1.498:1. Intonation-wise, these are i mportant intervals and they are approximated the same in all keys. H e l mholz argued against equally t e m pered scales in his "Sensations of Tone" by saying that it would h a v e a marked effect on our acuteness of appreciation for harmony. He remarked that the justly int o n e d chords "flow on w i t h a full stress, calm and smooth, w ithout beat." He continued, "Equally tempered chords sound b e s i d e them rough, dull, trembling, restless" [50]. Scales with 7, 12, 19, 2 2 , ' 31, and 53 number of steps p e r octave have b e e n suggested and some e x o t i c keyboards have b e e n b u i l t to control several of these scales of various lengths [30,126]. A novel design from the Netherlands was a th i r ty-one note organ k e y board which had terraced, interlaced keys col o r e d white, b l a c k and blue. .There were five divisions between each of the pa i r s C-D, D-E, F-G, G-A, and A-B, and three divisions be t w e e n E - F and B-C. The individual notes w e r e symbolized as follows: ii U _ Ji l _ _ _ C, C+, C , D , d ", D, D+, D , E , E~, E, E+, F _ , F, F+ . . . c " , C. The 31 notes w e r e chosen on the t h eoretical basis s u p p l i e d b y the famous Dutch s cientist and musician, Christian Huygens [51]. The claim was that on the 31-note pipe organ one could play the m u s i c of every civilization. The discussion of scales was m e n t i o n e d b ri e f l y b ecause m o d e m e l e c tronic control makes it possible to use scales of various lengths and to investigate their characteristics. Paul S. Rosberger has bee n c o ncerned for some time about the re-design of the common organ and p i a n o keyboard. In 1966 he s uggested a k e y board which required the d e p r ession of two adjacent keys simultaneously to produce a note [127]. Each key w i d t h was less than an average person's little finger, so the d e p r e s s i o n of a finger always p l a y e d a note. The next y e a r Rosberger p r e s e n t e d an improved version of his keyboard idea [126]. In this mo d e l there are no actual keys to press, but strips of capacitive s e n s o r elements wh i c h are spaced one centimeter apart. As Figure 1-9 36 . 16cm. 1cm. lcm. '■ Figure 1-9 New Keyboard D e s i g n - shows, each note, natural or sharp, is repre s e n t e d b y a similar strip. Those wh i c h c o r r e s p o n d to normal black keys are n otated b y the black bars w h i c h are for identification purposes only. There are several advantages o f this k e y b o a r d over the c o n v entional one; for instance, there are n o m o v i n g parts to break down, the keys are closer together so that a w i d e r span of notes is possible to the player, chords in d ifferent keys all h ave the same hand p o s ition since whole tones and semitones are always a fixed distance apart a n d the keys can be made touch responsive. A no t h e r n e w k e y b o a r d for controlling s y n t h e s i z e d electronic music is the t y p e w r i t e r - l i k e k e y board on the RCA S y n t h e s i z e r [60]. The keys resemble the old fashioned typewriter keys which were visibly attached to long levers. The function of these keys is to punch tape-control infor m a t i o n which directs the synth e s i z e r as is explained below. In the s i tuation wit h the p a p e r tape preparation, the composer is not i m m e d i a t e l y controlling the mu s i c output, b u t he is storing the control i n f o r m a t i o n for later use. In the s y s t e m at our university the teletype, or any ot h e r peripheral terminal connected to the computer, can be p r o g r a m m e d so that it directly controls aspects of m u s i c output. For example, the teletype keys could be used to pla y a melody, alter stop settings or dynamically control volume levels, etc. Other m a nual controls include patch and m i x i n g panels w ith their a ccompanying switches, knobs, tablets, and sliding bars. In prominent e l e c tronic mu s i c studios where there is a lot of equipment it is i mportant to hav e a m i x i n g panel wh e r e the various devices can be c onnected together. I ment i o n e d the mixi n g p a n e l several times as I d esc r i b e d Billent Arel's composing techniques to illustrate this point. Patch chords, s liding bars, and switches can also be set up so that certain regulation voltages result. These will be dealt w ith under e l e c tronic control. Variable resistive keying, capacitive force-s ensing switching, and v e locity-sensing keyi n g are all u s e d to provide precise and responsive touch control to the performer. E l e c t r o n i c control has become especially interesting and important since the d e v e l o p m e n t in 1964 by Robert A. Moog of voltage-controlled oscillators, filters, and amplifiers [110]. All of the Moo g voltage-c ontrolled modules are d e s igned w ith similar input and output impedances, v oltage and current levels and frequency response. Thus all the output signals from each m o dule can be input into any other module. Control signals consisting of voltages determine oscillator frequencies, filter characteristics, w a v e f o r m envelope shapes, rhythmic sequences and a m p lification levels. The control is linearized in most cases which means tha t an increase of one volt in the control signal produces a d i s crete o u tput signal change which is always the same amount over the 37 entire frequency range of the output signal [39]. Any electronic device w h i c h is able to produce an electrical s i gnal of any type, can be used as a control device to produce and m o d i f y audio waveforms. ' Store d - d a t a control has become a s i gnificant means of organizing sounds- and r e g u l a t i n g equipment. The RCA S y n t h e s i z e r can be controlled b y p u n c h e d p a p e r w h i c h passes over a drum. Holes p u n c h e d in columns along the p a p e r s p e c i f y the following sound p a r a meters which determine the output sound: frequency (from 20 cps, to 20,000 cps.), intensity (0 - 120 db.), attack, duration, decay, timbre, p o r t a m e n t o (free glide), frequency and a mplitude modulation (vibrato for i n s t a n c e ) , and random effects [116]. O r d i n a r y teletype paper-tape has b e e n used to supply data to v o l t a g e - c o n t r o l l e d equipment. The data-'is c onverted from digital form to vol t a g e levels by a D / A converter, and the resulting analog signal o p e r a t e s the electronic devices. A n y digital data, w h e r e v e r stored, can be u s e d in this w a y to control sound production. The natural conse q u e n c e of this is t hat the c o m puter w i l l p l a y an increasingly vital role in the synthesis and control of electronic m u s i c [66]. The p r o c e s s i n g power of computers w i l l facilitate the creation of very com p l e x mixtures of t i m b r e s , rhythms and melodic lines. Ano t h e r tape d r i v e n device is the Un i v e r s i t y of Toronto Hamograph [9] w h i c h is a programming device that has six control-information sensors. The control information takes the form of metallic foil strips w h i c h are attached to the surface of a tape film which can be driven at v a r iable speeds. The control bec o m e s d igital as it is p i c k e d up by the sense wires, and the resulting signals are used to turn on various dev i c e s or to control sound m o d i f i c a t i o n equipment. 38 Stored data, then, can be used to synchronize e l e c tronic equipment. As was p o i n t e d out in the area on sound mixing, the synchronization of recorders, s y n t h e s i z e r drives, and other me c h a n i c a l e q uipment is . important. An idea from the University of Illinois is to record . timing information alongside recorded audio, t hen synchronization caft be done p r e c i s e l y b y electronic controls [7]. The mixing of sounds from different tapes is done so m uc h that it w o u l d be interesting to be able to have se v e r a l tape c o n t r o l l e r s , all of w h i c h could be in­dependently addr e s s e d and each of w h i c h could b e s y n c hronized by timing information s t o r e d along w i t h the audio signal. - Digital control information is used in an i n teresting w a y b y Max Mathews in his G R O O V E computer pro g r a m [95]. "Manual control is sensed by the computer as digital control information w h i c h in turn modifies digital d ata that d i r e c t l y defines an output waveform. As the user enters control information, the computer p rocesses the control, modifies the s t o r e d w a v e f o r m samples, and sends out the resulting functions of time w h i c h may be used as analog control signals. The p e r s o n composing the sounds is able to have s e n s o r y feedback in real time. The feedback response may be audio or visu a l (CRT d i s p l a y ) , or both. The compo s e d functions of time may be u s e d to control any electronic equipment, as described above. Stor e d data in our system at Utah can r e p r e s e n t levels of control information. The b o t t o m level controls w h i c h o u t p u t switches are on (these affect pitch, volume, number of tones, and timbre) and other levels of s t ored d a t a control this ba s i c level. As m e n t i o n e d before, stor e d d a t a has b een used in core and i n t e g r a t e d circuit memories to control sto p and coupler 39 settings in a few e l e c t r o n i c organs. A l g o r i t h m i c control is closely associated w ith stored-data con­trol. In the case of algorithmic control, the dat a represents a p r o c e d u r e to be carried out. This p r ocedure m a y be a condensed r e p r e ­sent a t i o n o f s t o r e d dat a (a sine calculating routine as opp o s e d to the w a v e f o r m samples) or it may specify a complex transformation to be done a c c o r d i n g to certain conditions wh i c h exist. The p r o g rammer d e ­te r mines these conditions by inputting control information to direct the algorithms. In my system, as w e l l as in the GROOVE system, the input data, w h e t h e r from a keyboard, knobs, tablets, or light pen, may be i n t e r p r e t e d in any w a y b y stored computer programs. Since control is such a major factor in all areas of sound syn- •' thesis, the c o m puter assumes a vital role in electronic music production. M u s i c Notation. - The final area of electronic m u s i c organization is notation. E l e c t r o n i c m u s i c is a composer's w o r l d in w h i c h outsiders and n e w ­comers feel ill at ease and lonely b e c a u s e there is a lack of a d e scriptive s y m b o l i c language. Attempts have b een made and seminars h e l d to [23] d evelop notations for ex p r e s s i n g the procedures involved in creating a p i e c e of electronic music. The description of h o w to p e r f o r m certain ele c t r o n i c music is b a d enough, w i t h designated and r a n d o m cues to live p e r f o r m e r s , direction to turn knobs at sundry times and specifications for spacial effects. These actions as well as tape m a n i p u l a t i o n procedures, electronic device settings, etc. are difficult, if not useless to notate. Vladimir Ussachevsky has p a t ented 40 an e l e c tronic music score w r i t i n g n o t ation that has certain symbols to s pecify filtering effects, noise introduction, envelope shapes, re­verberation, etc. [147]. Junior h i g h and h i g h schools in Philadelphia encourage students to organize electronic music composition and then devise th e i r own notations. Many of the students r esponded by w r i t i n g p oet r y and c r e a t i n g accompanying art. W ith the a d v e n t of comput e r - c r e a t e d music came further attempts at languages to describe music. At P r i n c e t o n University, R e gner and R o b i s o n d e v e l o p e d S A M (System for Analysis o f Music) a n d IML (Inter­mediate M u s i c L a n g u a g e ) , respectively. Both systems are b e n t towards a linear r e p r e sentation of common m usical notation. Information such as durations, pitches, slurs, ties a n d articulation marks are stored in arrays a n d a c c e s s e d b y special routines. The Ford-Colunibia linear n o t a t i o n for ordin a r y m u s i c scores d e s i g n e d b y Stephan B a u e r - Mengelberg is p r o bably the foremost development a l o n g these lines. The effort is i n t e n d e d for use w i t h a computer-driven music printer, b u t the final form of the n o t a t i o n has not b een fixed y e t [65]. In this system as in S A M and' IML, i n formation describing various notational markings is e x ­t r a c t e d and p u t into tables which are indexed b y time [47]. A comprehensive a n d successful p r o g r a m for defining s o u n d structures is M a x Mathew's MU S I C IV in which the notation specifies w a v e f o r m p a r a ­meters. T h e b a s i c b u i l d i n g blocks are routines w h i c h function as os ­cillators, filters, amplifiers, a n d mixers. Here the music representation consists of numbers w h i c h define various paraineters< "for the routines that determine the w a v e f o r m s [142]. 41 A t the Un i v e r s i t y o f Illinois in 1963 a coirputer p r o g r a m MUSICOMP was d e v e l o p e d [5] that all o w e d the user to select various routines which y i e l d e d up r a n d o m numbers from various sets a n d distributions. These numbers w ere used to determine the so u n d parameters, pitch, octave, loudness, duration, timbre, a n d type of attack. On the R C A S y n t h e s i z e r the music notation is e xpressed in tabular columns o f b i n a r y e n c o d e d information [1]. Grap h i c a l p l o t t i n g o f s o u n d parameters enables b o t h the repre­sent a t i o n o f continuously v arying parameters, such as a glide in pitch, a n d of discrete events, such as notes a n d rests in a melody. Notation of musical o u t p u t has b e e n expressed in several graphical forms. For example, C h a n g a n d M a t h e w s wrote a graphical s c o r e - w r i t i n g p r o g r a m that p l o t t e d pi t c h a n d volume versus time of each separate voice sounding from the computer [21].. The o r d inate values wer e s c a l e d a n d no r m a l i z e d so t h a t the graph of a voice h a v i n g a w ide range of pitches was not larger tha n that of a voice w i t h a small p i t c h range. F o r example, in the former case the v e r tical distance of five inches on the graph p a p e r may cover five octaves; whe r e a s it could correspond to a single octave in the seco n d voice. O r d i n a r y music notation is m a inly p r e s c riptive in nature w i t h an emphasis on the structures, pitch, a n d rhythms. Graphical no t ation on the ot h e r h a n d can be very descriptive [133] w i t h precise information r e p r e s e n t i n g amplitudes, pitches, fluctuations in the basic pulse and attack envelopes. Various new and s p e cific notations have b e e n deve l o p e d and will continue to develop to r e p r e s e n t express and describe s o u n d production, modification, mixing, storage, and control. ' 42 ' HISTORICAL DEVEL O P M E N T OF ELECTRONIC . M U S I C A L INSTRUMENTS . ■The development of e l e c tronic musical instruments can b e b r o k e n • down into four periods. The first includes the attempts at electronic m u s i c pr o d u c t i o n b e f o r e the d iscovery of the vacu u m tube. The second p e r i o d extends from 1920 to W o r l d W a r II and comprises the development o f n o v e l single-melody instruments such as the Theremin, Hellertion, and Trautonium, and the development of polyphonic instruments such as the M i e s s n e r electronic p i a n o s and m a n y electronic organs. The third p e r i o d is an extension of the second period after W o r l d W a r II. This area covers the refinements and embellishments added to the i n stru­m e n t s developed before the war. The basic ideas of near l y all the instruments we have today w e r e developed in the peri o d b e t w e e n 1920 and 1940. Since the war, the v a cuum tube oscillators have b e e n re ­fined or replaced b y tr a n s i s t o r circuits; elaborate envelope shaping and tone forming circuits h a v e b een added, and novel p e r c u s s i o n sim­u l a t i n g circuits have b e e n developed. The fourth p e r i o d b e g i n n i n g in the 1950's includes the electronic music synthesizers w h i c h are c o n ­t r o l l e d by means other than ordinary keyboard manipulations. These i n struments include the RCA synthesizers and the Hamograph, the Moog, and the Buch l a equipment. N o v e l Experiments Before the V a c u u m Tube. * The first serious a t t e m p t at the widespread p r o d uction of music b y electrical means was m a d e b y Dr. Thaddeus Cahill at the b eginning CHAPTER 2 of the 20th century. In 1903 he p r o duced some mu s i c sounds wit h his giant Teleharmonium, or D y n a m o p h o n e , wh i c h consisted of about 30 car­loads of tone generators and controlling apparatus w e i g h i n g around 200 tons. The v a c u u m tube amplifier h a d not b e e n discovered yet and so large rotating m a g n e t o-electric generators w e r e u sed to produce the a l ternating audio signals. The large dynamos, that m e a s u r e d about two feet in d i a m e t e r and four feet high, produ c e d near-sine waves which were selec t i v e l y added by stop switches to synthesize the tones. Dr. Cahill h a d visions of transmitting his music to subscribers over telephone wires, b u t electrical engineering was not far enough advanced at that time to pre v e n t inductive interference bet w e e n the regular t elephone lines and the Teleharmonium, so after, m a n y complaints from telephone subscribers, the costly pro j e c t was abandoned [88,90]. There w ere early attempts before the v a c u u m tube was discovered to amplify p i a n o tones b y using telephone repeater amplifiers. In 1903 F a r r i n g t o n invented an instrument called the Choralcello w h i c h worked on the prin c i p l e of additive synthesis. To produce various timbres, a n u m b e r of h a rmonics were added to a fundamental tone. Even though some of the results were encouraging, the effort came to an end because of excessive cost and intricate and unstable e quipment [40]. Duddell's " singing arc" demonstrated in 1899 was one of the earliest devices w h i c h p r o d u c e d electrical music [35]. The p e r i odically discharging arc was par t of a tuned circuit which could be made to oscillate at d e f ­inite frequencies [34]. ; E a r l y Developments Before W o r l d W a r II. The development of the the r m i o n i c value, or v a cuum tube, by Dr. Lee de Forest in 1908 o p e n e d up the door for extensive p r o d uction of e l e c t r o n i c music. In fact, de F o r e s t himself made an electronic m u s i c a l instrument in which a varia b l e capacitance controlled an ■ o s c i l l a t i n g valve [35]. In America, Laurens Hammond and Benja m i n F. M i e s s n e r s h o w e d the capabilities of the new amplifiers by using the m in m a n y p a t e n t e d musical i n s t r u m e n t s , including various "electric" p i a n o s and organs. ... -- • •• ---- . In Europe, after the v a c u u m tube was discovered, it was u sed as b o t h an o s c i llator and amplifier in many of their single-melody i n ­struments. In these instruments one oscillator was controlled to p r o ­duce all the output frequencies. For economical reasons, the attempts to p r o d u c e instruments w ith large numbers of tubes failed. Descriptions of some of the instruments from the second period follow. The T h e remin was the early w ell publicized sound-effects i n s t r u ­me n t invented b y the Russian Leon T h e remin in 1919. It was capable of e m i t t i n g ho r r i f y i n g wails and e a r - p iercing shrieks to smooth gliding tones. It w o r k e d on the h e t e r o d y n i n g of two radio frequency oscillators w h i c h oscillate at different frequencies. There was a fixed o s c i llator and a variable one wh i c h was controlled by a varying capacitance. To b e g i n with, the variable o s c i llator was adjusted to match the frequency o f the fixed oscillator so that no b e a t i n g occurred. The high fre­q u e n c y of these oscillators, of course, was out of the audio range. As capacitance was added to the t u n e d circuit which d e t e r m i n e d the o s c i l l a t i o n frequency of the variable oscillator, the frequencies of 45 the two oscillators interfered with each other, p roducing a frequency wh i c h was the difference of the two and w h i c h was in the audio range. The v a r iable capacitance was added by the performer's h and which • m o v e d to and fro near an antenna. The other h a n d also introduced a va r iable capacitance which controlled another oscillator. This second osc i l l a t o r was c onnected to the gain control of an amplifier and the closer the h a n d to the antenna, the less the output from the oscillator. This r e d u c e d the bias on the amplifier and thus increased the audio volume. In the space of about three feet the p i t c h range of six octaves could be covered. Tone circuits c onsisted of some isolating condensers and radio frequency transformers, w h i c h p a s s e d on or reject­ed certain h armonics p r o d u c e d by the oscillators-. These high frequency h armonics p r o d u c e d corresponding overtones in the bea t n o t e , and the control of these harmonics facilitated the imitation of m a n y tones of diff e r e n t m u s i c a l i n s t r u m e n t s . One of the most realistic imitations was that of the cello [130, 14, 109, 106, 44]. A n o t h e r instrument, the Trautonium, d eveloped by Friedrich Trau t w e i n in Berlin, Germany, produced perio d i c waveforms by means of o s c i l l a t i n g neo n tubes. These tubes w e r e used to generate sawtooth waveforms, rich in harmonics. The frequency of the oscillator was d e t e r m i n e d by a tuned R-C Circuit in wh i c h the resistance was variably c o n t rolled by the p o s ition at which a taut m e t a l wire was pressed onto an u n d e rlying metal rail. The rail was a resistor, and the point at which the g r o unded me t a l string contacted the r ail fixed the resistance value in the oscil l a t i n g circuit. Dummy rubber keys were mounted over the wir e to indicate the positions o f the tempered scale. The keys 46 could be m o v e d to different positions along the wire to provide varied scales. The first version of this instrument was marke t e d by the T e l e f u n k e n Co. in 1930. The early models contained a few formant circuits through w h i c h a limited assortment of tone qualities was produced. The use of formant circuits in this machine was an inter­e s t i n g a p plication of the German d e v e l o p e d formant theory. The formant circuits were essentially r e s o nating circuits w h i c h introduced an e m p hasis to certain frequencies of the waveforms being modified. Va r i o u s o r d i n a r y instrumental sounds could be approximated by the Tr a u t o n i u m ' s formant filtering [12, 79, 97]. A n i n s t rument resembling the T r a u t o n i u m w a s developed by Helb e r g e r and Lertes called the Hellertion. It was a single tube 'instrument w i t h a p i t c h range of five octaves. The o s c i llator frequency was controlled b y a v a r i a b l e grid voltage which in turn was determined at the k e y ­b o a r d as in the Trautonium. The k e y b o a r d was responsive to touch, the fi nger d e p r e s s i o n deter m i n i n g the loudness of the tone. Touch r e spon­sive k e y b o a r d s b e came common on man y European instruments and after the war, elaborate mechanisms were created w h i c h controlled the attack rate and volume of the sounds [88,41]. One such control is described later in connection w ith Sala's Mixturtrautonium. One of the troubles wit h the T h e r e m i n was that a new technique, h a n d m o vement, h a d to be employed to "play" the instrument. The T r a u t o n i u m took a step in the direction of keyboards wit h its dummy keys, and the Hellertion followed s uit by h a v i n g transverse welts in i n tervals along long thin strips of leather. The welts indicated the p o s i t i o n of the equal tempered notes. A wire running under the 47 48 l e a t h e r s t r i p s c o r r e s p o n d e d to the t a u t w i r e o f the Trauto n i u m . A n o t h e r p r o b l e m w i t h the T h e r e m i n w a s t h a t it c o u l d b e p l a y e d o n l y in a g l i s s a n d o mode, w h e r e all i n t e r m e d i a t e p i t c h e s w o u l d be s o u n d e d as the p e r f o r m e r m o v e d hi s h a n d fro m on e n o t e to another. J o r g M a g e r e l i m i n a t e d this r e s t r i c t i o n in his S p h a e r o p h o n w h i c h like the Th e r e m i n , u s e d the b e a t i n g o f t w o h i g h - f r e q u e n c y o s c i l l a t o r s to p r o d u c e the a u d i o tones. M a g e r ' s i d e a wa s to s w i t c h v a r i o u s a m o u n t s of c a p a c i t a n c e a c r o s s the c i r c u i t c o n t r o l l i n g the v a r i a b l e r.f. o s c i l l a t o r . The t u n i n g c a p a c i t o r s w e r e a r r a n g e d in p a r a l l e l w i t h a k e y - o p e r a t e d s w i t c h b e t w e e n e a c h section. t ube F i g u r e 2-1 V a r i a b l e C a p a c i t a n c e S w i t c h i n g in the S p h a e r o p h o n D e p r e s s i o n o f the k e y o p e n e d the switch, thus v a r y i n g th e a m o u n t of c a p a c i t a n c e in the circuit. The S p h a e r o p h o n w a s f irst d e m o n s t r a t e d in the D o n a u e s h i n g e n F e s t i v a l in 1926. T h e r e w e r e f o u r m a n u a l s (strips o f leather) e a c h o f w h i c h was a m o n o p h o n i c " k e y b o a r d " , s o the p e r ­fo r m e r c o u l d p l a y up to 4 tones s i m u l t a n e o u s l y . T h e T i m b r e q u a l i t i e s r a t h e r t han b e i n g f o r m e d b y e l e c t r o n i c a u d i o f i l t e r s w e r e p r o d u c e d b y a v a r i e t y o f l o u d s p e a k e r s w i t h d i f f e r e n t l y s h a p e d d i a p h r a g m s , h o r n s a n d o t h e r r e s o n a t o r s . T h e s e r e s o n a t o r s in t u r n a d d e d f o r m a n t s to form d i f f e r e n t T i m b r e s [90,79]. A f u r t h e r i n s t r u m e n t o p e r a t i n g on the h e t e r o d y n i n g p r i n c i p l e was the F r e n c h Le O ndes M a r t e n o t b u i l t b y M a u r i c e Martenot. In thi s i n ­s t r u m e n t the c a p a c i t a n c e w a s v a r i e d b y the m o v e m e n t of a f i n g e r r i n g o v e r the key b o a r d . T h e r e w e r e als o som e s top keys w h i c h s e l e c t e d v a r i o u s filters. In 1928 the O n d e s M a r t e n o t a p p e a r e d in the P aris Opera, an d te n years l a t e r a s p e c i a l v e r s i o n w a s b u i l t to p r o d u c e m i c r o t o n e s r e q u i r e d in som e of H i n d e m i t h ' s m u s i c [78]. T h e e a r l y E u r o p e a n e l e c t r o n i c i n s t r u m e n t s w e r e m o s t l y m o n o p h o n i c b e c a u s e the p e o p l e w e r e r e l u c t a n t to b u i l d w i t h a large n u m b e r o f tubes. Th e F r e n c h G i v e l e t - C o u p l e a u x o r g a n wa s an e x c e p t i o n h o w ever. T h e large F r e n c h G i v e l e t - C o u p l e a u x Organ, p i o n e e r e d in 1930, was the first e x p e r i m e n t w i t h a l arge n u m b e r of o s c i l l a t o r s . It h a d h u n d r e d s of k e y e d o s c i l l a t o r s a n d its d e s i g n set a p r e c e d e n t f o l l o w e d b y the A l l e n and C o n n O r g a n m a n u f a c t u r e r s , t h a t is, a si n g l e o s c i l l a t o r p e r tone. The c o m p l e x w a v e f o r m s p r o d u c e d b y the o s c i l l a t o r s w e r e f i r s t a m p l i f i e d and t h e n s h a p e d b y filters and form a n t circuits. The i d e a w a s t h a t s ince m o r e c u r r e n t b e c a m e a v a i l a b l e a f t e r a m p l i f i c a t i o n , the t o n e s h a p i n g c i r c u i t s c o u l d p e r f o r m g r e a t e r p h a s e sh i f t s an d thus m o r e e a s i l y c olor the sound. Th e o r g a n t u r n e d ou t to b e b u l k y and it c o n s u m e d m u c h power. It w a s so e x p e n s i v e t hat the E u r o p e a n m a r k e t c o u l d n ' t a f f o r d this d e s i g n [12,40]. C a p t a i n R i c h a r d H. R a n g e r d i s p l a y e d h i s m a g n e t o - e l e c t r o n i c o r g a n in 1931. His system, w h i c h w o r k e d on the same p r i n c i p l e as C a h i l l ' s T e l e h a r m o n i u m w a s m u c h s m a l l e r due t o v a c u u m tube a m p l i f i c a t i o n o f the sound. A d d i t i v e s y n t h e s i s d e t e r m i n e d the tone color, a n d c e r t a i n of the c o m p l e x s o u n d c o m p o n e n t s c o u l d b e m o d u l a t e d in amplitude. D e c a y 49 shaping of the w a v e f o r m envelopes p r o duced percussive-like tones. There were twel v e separate sets of motor driven alternators whose speeds were controlled by tuning fork vibrations [102], R o t ating discs p r o vided a basis for periodicity, and b e fore World War II, e x periments were p e r f o r m e d to p roduce tonal waveforms with gramophone records. Equispaced tinfoil waveforms were attached to the records and fixed electrostatic p i c k u p electrodes were p l a c e d above the r o t a t i n g waveforms. Later, an all metal disc was substituted wit h e l e c t r o s t a t i c pickup, b u t these experiments never proved to be fruitful b e c a u s e of mechanical instabilities w h i c h caused m o d u lation and di s t o r t i o n of the tones [81], In the late 1920's Langer p r o d u c e d his Emicon which use d variable resistance s e l e c t e d by a keybo a r d to control the oscillating frequency of a neo n tube. Later in 1934 V i e rling and Kock p r o duced their "E l e c t r o a k u s t i c h e O r g e l" wh i c h generated complex w a v e s from n eon tube oscillators. The complex w a veforms wer e filtered b y an inductive discharge circuit capable of producing organ sounds ranging from flutes to strings [102]. A r o u n d the 1830's the Frenchman, Constant Martin, developed the C l a v ioline w h i c h was a melody instrument containing a u n ivibrator tone gene r a t o r and two frequency dividers of the Eccles-Jordan type. Vibrato was controlled by a m u l tivibrator which h a d variable o s cillation rates. This instrument could imitate m any orch e s t r a sounds with its tone sh a p i n g and pe r c u s s i v e circuits. Its appearance resembled the Hammond Solovox although the circuitry of the Clavioline was much simpler. It h a d a three octa v e keybo a r d and a five octave pi t c h range, so only a 50 • p o r t i o n of the instrument's range could be p l ayed at one time [40]. The N e o Bechstein p i a n o was a miniature grand havi n g relatively short strings which w ere struck by very small "micro hammers." There was no sounding board, b u t instead the string vibrations wer e picked u p electr o m a g n e t i c a l l y and e lectronically amplified. The lou d sounds • b e c a m e quite blurry, however, because of interference caused b y spurious m a g n e t i c fields. The p i c k u p device was connected directly to the v i b r a t o r and dampers could control t h e loudness o f pitch [77,102]. B e n j a m i n F. M i e s s n e r ejqperimented with m a n y ways of using electronic d evices to produce music. There are several patents under his name d i s p l a y i n g the sundry designs of electro-mechanical pianos, p h o t o ­e l e c t r i c organs, and ot h e r e l e c tronic musical instruments. Miess n e r d e v e l o p e d an amplified p i a n o which w o r k e d on the principle of the con­d e n s e r microphone. A large voltage (150 v.) was impressed across the p i a n o string and a p i c k u p screw. As the string vibrated, the c h a n g i n g capacitance v a r i e d a voltage which was fed to an amplifier. The v o lume of the sound was de t e r m i n e d by the distance of the pick u p s c r e w from the string. Eac h pickup was adjusted to give an overall b a l a n c e of the 88 p i a n o tones [103]. A n o t h e r s i m i l a r i n s t r u m e n t wa s t h e D y n a t o n e w h i c h h a d a k n e e l e v e r t h a t a l l o w e d the p l a y e r t o s w i t c h f r o m h a r p s i c h o r d t o p i a n o to c o n c e r t g r a n d p i a n o sounds. T h e e f f e c t s w e r e c a u s e d b y the d e g r e e o f a m p l i f i c a ­t i o n a n d t h e n u m b e r o f h a r m o n i c s a d d e d to p r o d u c e the sound. W i t h no a m p l i f i c a t i o n the s t r u c k n o t e s r e s e m b l e d a h a r p s i c h o r d s i n c e t h e r e w a s n o l a r g e s o u n d i n g bo a r d . It w a s a d v e r t i s e d as "an e l e c t r i c p i a n o w i t h r a d i o a n d p h o n o g r a p h r e p r o d u c t i o n [82]. A l o n g w i t h the p i a n q the . 51 console c o n t a i n e d a record player, a radio, a microphone and speaker system. . . The E l e c t o n e was similar to the Dynatone except that the Electone h a d three e l e c t r o s t a t i c pickup electrodes along each of the 88 strings ins t e a d of just one pickup. Since the p i c k u p screws were at different . p ositions a l o n g the strings, several modes of the string vibration were sensed. Each of the resulting wave f o r m s contained different harm o n i c s and various tone colors w e r e s y nthesized b y the selective m i x i n g of the waveforms.. In .other si m i l a r instruments selective h a r m o n i c con t e n t was achieved through the use of variable shaped p i c k u p plates. Large p i ckup electrodes wer e u s e d to filter out some of the h i g h harmonics and thus enhance the lower- notes. The first " e l e c t r o n i c pianos were introduced b e f o r e W o r l d War II, but wit h the b o m b i n g of P e a r l Harbor came the cessation of manufacturing. After ' the war, b e t t e r models w ere produced [97,140]. M i e s s n e r a l s o p r o d u c e d s e v e r a l i n s t r u m e n t s h a v i n g e l e c t r o s t a t i c p i c k u p o f w i n d b l o w n reeds. B y c o n t r o l l i n g t h e p o l a r i t i e s an d m a g n i t u d e s o f the c u r r e n t s in the p i c k u p e l e c t r o d e s , d i s t o r t e d , c o m p l e x w a v e f o r m s w e r e o b t a i n e d f r o m the v i b r a t i n g reeds. Thu s v a r i a b l e t i m b r e s c o u l d b e p r o d u c e d b y v o l t a g e r e g u l a t i n g p o t e n t i o m e t e r s . Th e r eeds als o o f f e r e d a d e l a y e d a t t a c k s i n c e it t a k e s t ime to ge t t h e m v i b r a t i n g . M i e s s n e r w a s v e r y h a p p y w i t h t hese i n s t r u m e n t s b e c a u s e the r e e d s w e r e cheap, e a s y t o u s e a n d c o u l d b e k e y e d b y s i m p l y o p e n i n g air valves. T h e r e w e r e a l s o n o p r o b l e m s of e x a c t s p e e d c o n t r o l o r c o m p l i c a t e d t i m i n g m e c h a n i s m s . M i e s s n e r w a s c o n s t a n t l y s e e k i n g a f t e r th e ideal i n s t r u m e n t w h i c h w a s fo r h i m on e t h a t " c a n m a k e any sound, kn o w n , 52 unknown, o r conceivable" [102], The L o a r Vivitone also used reeds as tone sources, b u t the v i b r ations were p i c k e d up m a g n e tically as in the RCA electric carillon and in various s t r inged instruments. In Euro p e V i e r l i n g p r o duced the Electrochord, an electronic • p i a n o w h i c h h a d e l e ctrostatic pickups similar to the M i e s s n e r pianos. O n each string several p i c k u p electrodes were used and they differed in size. Some of them w e r e placed un d e r the brid g e to p i c k up the attack transients. The formant tone control filters and envelope s ha p i n g circuits enabled the Electrochord to produce a var i e t y of tone colors from the sound of a xylophone to that of an organ [97]. J u s t as the Everett Pianotron produced music by electrostatic screw p i c k u p of v i brating s t r i n g s , the Orgatron generated sounds from the el e c t r o s t a t i c p i ckup of w i n d blown h a r m o n i u m reeds. Screw pickups wer e a d j usted over the five ranks of ordinarily shaped organ reeds to dev e l o p the p r o p e r volume. Each timbre h a d an independent ban k of reeds, w h i c h were coupled pneumatically. Differently shaped pickup screws w e r e e x p e r imented w ith and it was found that a flat h e a d screw p r o d u c e d the most capacitance. To produce complex waveforms the languids of the reeds wer e shaped as was common to reed organ builders. Figure 3 shows some examples of shaped reeds which vibrate in different modes [101]. 53 Figure 2-2 Shaped Reeds H a r o l d Bode, now living in New York, is the inventor of many electronic m u s i c instruments. He was e s p e cially active in building electronic i n struments in Germany just before and after Wo r l d War II. • Since the E u r o p e a n economy couldn't support large expensive instruments, inventors h a d to mak e several compromises in b u i l d i n g consumer products. One i n t e r e s t i n g design was attempted b y Bode in his Wa r b o Formant Organ. The o b j e c t i v e was to have a m ultitone organ with few oscillators. This was a c c o m p l i s h e d by sharing the four relaxation oscillators such that four notes w o u l d sound simultaneously. The switching system assigned the first oscillator to the h i g h e s t note, the second oscillator to the nex t h i g h e s t and so on. The m a n u a l h a d 44 keys and there were two sets o f filters and some envelope control circuits for percussive sounds. The ven t u r e was unsuccessful bec a u s e the switching tree was too expensive and the oscillators w e r e n ' t very stable [12]. In the late 1930's, Bod e developed his M elochord w h i c h contained complex tone and envelope shaping circuits. An inter e s t i n g idea employed in this instrument was a traveling formant circuit wh i c h was retuned as notes from d i f f e r e n t pi t c h ranges were played. The result was that a simi­lar timbre could be ma i n t a i n e d over the entire p i t c h range of the instrument. As an analytical device the M e l o c h o r d perm i t t e d the in ­vestig a t i o n o f different rates of attack and decay, imitative tonal synthesis, art i f i c i a l echo and reverberation. There were two manuals of which the lower s u p plied the pitch note and the upper p r o vided any other note w i t h or w i t h o u t harmonics [29]. ; In 1932 the Russian physicist Ivan Eremeef demonstrated an organ w h i c h d e r i v e d its tones from a system of electro-magnets which sensed 54 ' circle. Some discs cont a i n e d sine waves instead of recorded complex w a v e s , and additive synthesis was accomplished to produce various tone qualities [40]. As early as 1921 an idea came from France to rotate the scan s l i t s ^ i n s t e a d of the waveforms. In.this design Toulon used stationary complex and simple waveforms. There wer e four light sources, each of w h i c h s u p p l i e d two wheels. Shutters operating at the focal points of the light b e a m s w e r e controlled by the. organ keys. The mask disc con­s i s t e d o f several sectors, each of wh i c h contain a specific waveform. T h e r e w e r e twelve concentric, circles of these w aveforms so that an octa v e of tones a p p e a r e d in each sector [101]. (See Figure 2-3). 56 Figure 2-3 T o ulon Disks The other early full scale photoelectric organ besides the Toulon was the German We l t e wh i c h was produced in 19 30. In this design, the slits wer e stationary and the glass disc containing engraved waveforms w a s rotated. The w a v e f o r m s w e r e complete tonal images made from sound tracks of actual organ pipes. There w e r e twelve identical tone wheels w h i c h w e r e rot a t e d at diff e r e n t speeds to m a i ntain a'tempered scale. Keys o p e r a t e d m a g netic shutters which selected waveforms from one of magnetic fields set up by rotating t oothed magn e t o - t y p e phonic wheels. Earlier he h a d u n s u c c essfully experimented w ith his Photona, a ph o t o ­electric o r g a n us i n g p e r f orated discs d r iven b y a synchronous motor . [40]. There h ave b e e n man y attempts to use the p h o toelectric effect to p r o d u c e music. The ideas centered around a rotating wheel which ■ contained e v e n l y s p a c e d holes, periodic waveforms, or recorded sound tracks. . . One of the e a r l i e s t patented p h o t o e l e c t r i c organ designs was by B.J. 1-liessner in 1926, after four years of experimentation. This s y s ­tem c onsisted of a rotational scanning m e c h a n i s m wh i c h could generate sinusoidal a n d com p l e x waves. The intensity of the light b e a m was v a r i e d to control the volume. This p r o j e c t di'dn't come much farther than the d e s i g n state, probably because of i n h e r e n t difficulties en­countered in the p h o t o e l e c t r i c approach and bec a u s e Miessner was h a ving more success w i t h ot h e r forms of electronic m u s i c p r o d uction [102]. Ano t h e r ea r l y attempt at using the p h o t o e l e c t r i c effect was made in the H a r d y - G o l d t h w a i t e organ which used phot o g r a p h i c discs containing variable de n s i t y waves. The waves w ere r e p l icated from those made by original instruments, a n d each disc contained the pitch frequencies of 71 notes of the t e m p e r e d scale for a pa r t i c u l a r tone color. The ro­tating w a v e f o r m s w e r e scanned through optical slits which were controll­ed b y shutters. Since the waveforms didn't quite m a t c h up at the ends w hen laid out in a circle on the disc, there w a s a click corresponding to the ro t a t i o n a l s p e e d of the disc when the s h i f t i n g was done only at one place. In order to minimize this clicking effect, the shifting was d i v i d e d up and done in several equispaced intervals around the 55 the 18 concentric circles of engraved images. The organ was received in B e r l i n w ith muc h enthusiasm. Unfortunately, the Welte organ was d e s t r o y e d in W o r l d W a r II [25,32], . The Toulon idea of rotating slits was also used by Lesti and Sammis in their Polytone. The slit spacings w ere equal to the lengths of the waveforms, and thus the waveforms were repeated at the frequency d e t e r m i n e d b y the n u m b e r of slits pa s s i n g p e r second. The waveforms w ere stor e d in radial sectors on a disc, and waveshapes wer e selected b y rotating the appropriate sector of the disc to an o perating position. S e v e r a l rows of lights facilitated the retrieving of sounds from sev e r a l sectors simul t a n e o u s l y [40]. A l t h o u g h many attempts have b e e n made at using p h o t o e l e c t r i c ‘ generators, there wer e inherent difficulties that prev e n t e d this method from b e c o m i n g w i d e l y used: Early efforts use d 60 cycle alternating cur r e n t to run their lamps because of the high cost to rectify enough c urrent to operate the m a n y lights. Unfortunately, this alternating current introduced a 120 cps frequency which m o d u l a t e d the synthesized sound. The emulsions containing the waveforms on the discs wo u l d c o n ­tract after some time, leaving a d istorted waveform. Lamps b u r n e d out and h a d to be replaced frequently. Finally, the equitempered notes o f the scale could m e r e l y be roughly approximated, and then wit h tedious e f f o r t since the o nly exact frequencies possible w ith rotating discs are frequencies at integral multiples of the rotational speeds. The only organ d eveloped before W o r l d W a r II that still is be i n g w i d e l y aold today is the Ham m o n d Organ developed by Laurens Ham m o n d of Chi c a g o in the 1930's. He was a very successful inventor and h o lder of 57 • many patents even before he turned his talents towards the organ in­dustry. He d e v e l o p e d a synchronous motor wh i c h he used to drive clocks. One day, it is reported, he heard a hu m m i n g noise coming from one of his clocks and dec i d e d he wo u l d build an organ using his synchronous . motor. The tone generators consisted of 91 s calloped discs wh i c h r o ­tated near the e n d of a perm a n e n t m a g n e t w i t h a coil w r a p p e d around it. As the h i g h spots on the edge of a disc p a s s e d the magnet, its magnetic field was v a r i e d w h i c h induced a small current in the coil. The tone w h e e l dri v i n g m e c h a n i s m was so well constructed that p r e s e n t day models vising tone whee l s hav e the same design with little modification. The w h i r l i n g tone w h e e l s gene r a t e d near-sine waves w h i c h w e r e m i x e d to pr o d u c e sounds b y additive synthesis. The distinctive H a m m o n d sound comes from the tones being generated all in p h a s e and from the mixing o f tempe r e d harm o n i c s r a ther than pur e harmonics. Tone colors for each m a n u a l w ere c o n t rolled by a separate set of nine drawbars which d e t e r m i n e d the amplitudes of the fundamental and eight harmonics: the subfundamental, subthird, fundamental, second, third, fourth, fifth, sixth, and eighth. Every one of the mus i c a l frequencies in the entire range of the organ was p r o d u c e d by a d i s tinct tone wheel. Each gear on the driveshaft was coupled to two tone wheels [32,43]. 58 (See Figure 2-4). 59 tone wheels J N - w ir e coil f drive shaft Figure 2-4 H a m m o n d Tone Wheels The signals from the generators were mixed, amplified, and then se lectively sent to v i b r a t o circuits. . On the very early H a m m o n d organs the vibrato was a c h i e v e d by amplitude m o d u lation ( t remulant). Later an interesting ph a s e shift circuit was deve l o p e d w h i c h imparted both frequency and amplitude mo d u l a t i o n to the signal. A capacity pickup scanner t a p p e d off signals from d i f f e r e n t points al o n g an electrical time delay circuit consist­ing of a number of low pass filter sections. At each tap on the line the phase of the signal is retarded in relation to the previous tap. The scanner s equenced re p e a t e d l y up and down the line c ausing an u n ­du l a t i n g m o d u lation of frequency. The amount of v ibrato could be v a r i e d from narrow to w i d e by causing the sequencer to scan small to large portions of the delay line in a given amount of time. The development of a reverberation control for the H ammond was also interesting. This d e vice produced damped echos by reflections of wa v e s w i t h i n a network containing five lengths of s p r i n g s , four tubes filled w i t h oil, and two r o cker arms. The electrical signal was t r a n s d u c e d to a mechanical vibration which was pr o p a g a t e d and re­flected in the series of spring coils, some of which were immersed in oil to e f f e c t damping. The output v i bration was converted b y Rochelle p i e z o crystals to an electrical signal which was mi x e d in differing p r o p o r t i o n s w i t h the original u n r e v erberated signal [40, 104]. Lau r e n s Ham m o n d p r o d u c e d many o t h e r interesting designs in the or g a n industry. One successful invention in 1937 was the Nova c h o r d w h i c h use d some t h i n g like 163 v a cuum tubes to produce the tones. It was the first commerical p u r e l y electronic m usical instrument p o s s e s s ­ing a full k e y b o a r d on w h i c h chords could be played. Twelve v a c u u m tube o s c i l l a t o r s fed the signals to frequency divid i n g circuits which p r o d u c e d o utputs h a ving a rich content of harmonics. These complex •' w a v e signals w ere then sent through e n v e l o p e - s h a p i n g and tone-forming circuits w h i c h wer e controlled by dials above the single six-octave keyboard. The tone control dials h a d settings which allowed the p l a y e r to s e l e c t gradations of attack and de c a y from slow to fast, diff e r i n g re s o n a t i n g filters to p roduce formants, various high and low pass filters, four levels of volume, and va r y i n g rates of vibrato. As a result, tones ranging from those of a plu c k e d string to those of a b o w e d s t ring could be selected, and the tone could be "full", "brilliant", or "deep" [77, 98]. George Gershwin was so impressed wit h the p r o spect of this instrument that he b o u g h t the first one and, of course, gave the N o v a c h o r d muc h publ i c i t y [114]. A n o t h e r Ha m m o n d instrument havi n g tone controls similar to the No v a c h o r d was the Solovox wh i c h was a m o n o phonic instrument. It was d e s i g n e d to be attached to a pi a n o to provide an instrumental voice 6 0 along w i t h pi a n o accompaniment. Atta c k and decay rates were determined b y diff e r e n t sized capacitors which were switched in and out of the circuits b y tone control knobs. The five octaves of pitches were . g e n e r a t e d by a single tube o s c i llator and four stages of frequency dividers. A panel switch was used to select which pitch range (bass,' tenor, contralto, soprano) should sound. Any or all of these octaves co u l d b e used. Filter circuits ena b l e d the Solovox to imitate the sounds of stringed and w i n d instruments including violins, cellos, trumpets, flutes and tubas [32, 99, 137]. The principle of m a g n e t i c induction was used in other instruments be s i d e s those of Hammond. Joh n Goodell and Ellsworth S wedien develop­ed t h e i r Maste r s o n i c w h i c h u s e d various shaped-magnet ends to provide d i s t i n c t i v e tone colors. There wer e twelve shafts with seven pitch w h e e l s each which r otated n e a r the irregularly shaped magnets wo u n d w i t h coils. Each of the pitch wheels contained twice as m a n y r e c ­t a n g u l a r teeth as the p r e c e d i n g one, so seven octaves w ere p r o duced p e r shaft. Several d i fferently shap e d poles were dispersed radially a r o u n d each wheel. teeth Shaped wave Poles form flute Pole M a g n e t string diapa s o n Figure 2-5 Mastersonic Tone Generation As is see n from Figure 2-5, the different organ sounds were form­ed d i r e c t l y r a ther than by additive synthesis as on the Hammond. The aim of this M a s t e r s o n i c project was to produce p ipe organ sounds, and m uch o f the d e sign time was spent d e t e r m i n i n g the shapes of the m e c h ­a n i s m tha t simu l a t e d attack effects. The attack transients wer e o b ­t a i n e d b y e x t r a contactors during the initial travel of the key. The r e s u l t i n g tone qualities imitated p i p e organ sounds very w e l l [62]. In England, John Compton d e v e l o p e d a successful electronic o r g a n w h i c h p r o d u c e d sounds by rotating electr o s t a t i c generators. The ge n e r a t o r s c onsisted of two plates, o n e stationary and the other r o ­tating. The r o t ating member r esembled to some extent a spider's web, con s i s t i n g of a rim w ith radial spokes. These conductive radial lines, r o t a t e d o p p o s i t e the fixed stator w h i c h was a disc containing 5 c o n ­centric circles of engraven metal in the shapes of different complex waves. The inner annular ring cont a i n e d two such waves and each s u c c e e d i n g r i n g contained twice as m a n y as its predecessor. The w a v e ­forms w e r e e l e c trostatically p i cked o f f the stator b y the rotating dis c and fed to the grid of an amplifier. Fro m there the signal was sent thr o u g h the voicing filters. V i b r a t o was applied by slightly o s c i l l a t i n g the stators [32]. A n o t h e r organ, closely resembling the C ompton b u t m uch smaller, was the Melotone. The tone synthesis was similar to that done on the H a m m o n d since only sine waves were generated. Vibrato was applied by m o v i n g an e ccentric pulley back and forth. The pulley was attached to the b e l t w h i c h drove the rotors [25] . • 62 ' ■ Extensions of the E a r l y Developments. A f t e r W o r l d W a r II, improvements were mad e on the basic instru­mental designs just mentioned, and these modifications make up the third h i s t o r i c a l p e r i o d of electronic musical instruments. Oscar Sala e x t e n d e d the T r a u t o n i u m to a studio model, the M i x t u r t r a u t o n i u m , which had two sets of v a r i e d tone c i r c u i t s , several frequency d i v i d e r s ; a foot-o p e r a t e d m e c h a n i s m to provide g r a d u a l timbre changes and a , pressure s ensitive k e y b o a r d which c o n t rolled volume as well as pitch [74]. Sal a w a s a stu d e n t of Trautwein and H indemith at the Berlin H o c h schule ftlr Milsik in the m i d - 1 9 5 0 ' s. The touch responsive keyboard h a d a liquid resistor in the form of a long tube w h i c h was depressed by the keys to generate gradations of volume and the rate of attack [33] . The liquid resistance tube was inside a long cylindrical grid resistor w h i c h controlled the oscillation frequency of the sawtooth pr o d u c i n g thyratron tube. Around this grid resistor was a springy metal gauze w h i c h mad e contact with the grid resistor w hen depressed by a dummy rubber key [ 3 1 ] . 63 attack resistor Figure 2-6 Thus an u n d u l a t i n g m o v ement of the fingers on the key wo u l d produce a v ibrato c o n s i s t i n g of b oth pitch and volume variation. The harmonic content of each octave was sent to indep e n d e n t tone f i l t e r s , which m e a n t t hat the lower octaves could b e reproduced along w i t h the fundamental p i t c h but with different tone coloring. The studio T r a u t o n i u m was used to create the m u s i c of the ballet "Electronics" w h i c h m a d e its debut in 1960 with 31 loudspeakers. The frequency of the p e r f o r m a n c e r a nged from 10 cps. to 40,000 cps. and the p e a k output a p p r o a c h e d the t hreshold of pain. T h e mu s i c of the b a llet was d e s ­cribed b y critics as b e i n g conservative s y mphonic music. The Tra u t o n i u m is c a p a b l e , as a demonstration record s h o w s , of producing the sounds of a min e eaqplosion, s t e a m locomotive, small town b a n d (amusingly out of. tune) and a full milit a r y b a n d [85]. A f t e r the w a r H a rold Bode p r o d u c e d a variety of interesting e l e c t r o n i c m u s i c instruments including a large organ which used Hartly v a c u u m tube oscillators in conjunction with frequency dividers [12]. In 1953 he d e v e l o p e d a small portable p u rely electronic organ called the T u t t i v o x w h i c h used p h a s e shift frequency dividers. The Tittivox was later combined with a Clavioline under the name of Combichord and is m a n u f a c t u r e d by the German Hohner Corporation. M atthew Hohner, the f ounder of this company, specialized in maki n g harmonicas and later p r o d u c e d several electronic novelties. One of these was the Multimoni-k a w h i c h c o n s i s t e d of 2 manuals each with 41 keys. They also make the f oll o w i n g e l e c t r o n i c music instruments: the Electronium, a single m e l o d y instrument; the Hohnerola, a single manual reed electronic organ; the Cembalet, a h arpsichord-like i n s t rument w ith percussive tones; and 6 4 . the H o hner V o x a n d Bassophon, which are supplements for the Hohner accordians. The Bassophon, which h a d controllable attack circuitry, could simulate eith e r a string bass v i o l a or t u b a in a 2- 1 / 2 o c t a v e . range. These instruments contain various formant filters which em­phasize certain frequency bands and change the timbre of the accordian sound. In the E l e c t r o n i u m the bellows serve a double function. Pressure at o n e e n d controlled the volume, wh i l e pressure at the other m o d i f i e d the h a r m o n i c content and thus, the timbre of the sound [29]. A n o t h e r i n s t r u m e n t from Bode, the Univox, was an interesting s ingle-melody in s t r u m e n t that produced sawtooth waveforms by the rapid charging and the s low discharging of v a r i o u s - s i z e d capacitors which could be s w i t c h e d in and out of the oscil l a t i n g circuit.- Vibrato was furnished by a m u l t i v i b r a t o r which could be s w i t c h e d from four to eight cps. The v i b r a t o oscillator could also b e switched to trigger r epe a t - p e r c u s s i o n to give effects similar to a mandolin, etc. As on the Clavioline the U n i v o x h a d a three-octave k e y b o a r d and a five-octave pitch range. The keys wer e touch respondent in tha t a soft depression w o u l d yi e l d a g r a d u a l attack and a quick, h a r d depression w o u l d cause a staccato effect. Percussive delay circuits and formant circuits enabled the p r o d u c t i o n of various timbres [40]. Later a U n i v o x - l i k e instrument was added to an accordian to form the Hohner Multimonika. The upper manu a l co n t r o l l e d the monophonic melody wh i l e the lower operated the a c cordina a c c ompaniment which was p r o d u c e d by 3 sets of wind-blown reeds. One of the last electronic reed organs was the W u r l i t z e r which had several c a p a citive screw pickups above a w i n d b l o w n reed as shown 65 ' 66 in Figure 2-6. p i ckup screws - -air opening * air p a t h Figure 2-6 W i n d Bl o w n Reeds wit h Electro s t a t i c P i ckup The air flow h a d to be carefully controlled b e c a u s e an explosive surge w o u l d cause the ree d to strike the p i c k u p screws 132, 40]. After this model, W u r l i t z e r joined the el e c t r o n i c gr o u p and be g a n p roducing pure l y e l e c t r o n i c organs. They do, however, still mak e electric pianos w h i c h operate on capacitive p i c k u p and amplification of the vibrations of m e t a l l i c reeds which are struck w i t h k e y activated mallets. The o t h e r means of electronic m u s i c pr o d u c t i o n tha t h e l d on into the late 1950's was p h o t o e l e c t r i c generation of tones. M a n y experimental designs w i t h r o t ating disks were proposed. Figu r e 2-7 shows one such design. Five i dentical tone wheels are rotated, each twice as fast as its p r e d e c e s s o r to give five octaves of tones. There are twelve con­centric circles w h o s e spaced holes determine the frequencies of a 12- tone scale. The w h e e l shown approximates an e ven tempered scale. Various lamps are turned on behind the wh e e l and the p u l s a t i n g light beams are c onverted into oscillating voltages by the recipient p h o t o ­cells. V o lume is c o n t rolled by interposing a d i f f u s d r between the light source and the photocell. The diffuser is a grad u a t e d exposed Photocell m i r r o r■ ' ^ L - s / Amp motor lamp Figure 2-7 P h o t o e l e c t r i c Disk and Tone Production f ilm w h i c h lets varying amounts of light through. A slight periodic m o v e m e n t of the diffuser produces tremolo [19]. Another photoe l e c t r i c c o n f i g u r a t i o n u t i l i z e d tone wheels w h i c h contained the fundamental f requency and nine other harmonics in successive concentric rings. A d d itive synthesis of these sine w aveforms was accomplished b y • sele c t i v e t urning various lights on and off. A variable s p e e d motor was used, b u t the instabilities in the rotational speed caused con-, s i d e r a b i e audio d i s t o r t i o n [64], B a l d w i n p r o d u c e d a p h o t o e l e c t r i c organ in the 1950's, b u t quickly w i t h d r e w it. The last serious attempt at p h o t o e l e c t r i c r e p r oduction o f a u s i c was b y the Kimball Or g a n Company in 1958. In this design a k e y s e l e c t e d the appropriate tone pic t u r e to b e * s c a n n e d b y rotating slits. Each picture pl a t e contained 61 recorded copies of actual pipe o r g a n sounds [63]. In California, C h a m b e r l a i n p r o p o s e d an organ made up of a b ank o f tape recorders w h i c h p l a y e d pre - r e c o r d e d or g a n sounds, b a s s accom­p animent, and special effects. The Mellotron, as it was called, h a d s e v e n t y individual 3-track tapes, each forty-eight feet in length. The k e y b o a r d was arranged in two 35-key manuals, side by side. The left 35 keys selected b a s i c chord sequences and rhythmic patterns and the right k e y b o a r d p l a y e d the lead instrument. The choice of 18 instruments i n c luded clarinet, flute, trombone, p i a n o and organ [119]. The me c h a n i c a l r o t ating and vibrating electronic mus i c a l instru­ments are a thing of the p ast now, w i t h the exception-of the Hammond r o t a t i n g wheels. All large m a n u f a c t u r e r s , including Hammond, have turned to p ure electronic so u n d p r o d uction because of inexpensive solid state components. . 68 Some very intere s t i n g things have been done in the electronic organ industry to make their instruments appealing to the consumer. An early d e velopment after the w a r was Hammond's chord o r g a n w h i c h had a m a t r i x of 96 buttons w h i c h could so u n d the 6th, 9th major, minor, 7th, diminished, augmented, a n d m i n o r 7th chords of all the twelve keys. Each adjacent two or three notes shared the same oscillator, so the 32-note compass of the organ was covered by only 16 oscillators [26]. The first organ p r o d u c e d by Thomas in the 50's shared one oscillator for three notes. The sharing of oscillators in this m a n n e r is long ou t d a t e d now. The normal square, sine, and sawtooth waveforms were p r o d u c e d by a v ariety of ingenious techniques. Baldwin came up with the idea of staircasing square wa v e s to approximate sawtooths and o u t p hasing saw-tooths to p roduce square waves. Staircasing is p e r f o r m e d b y mixing t o g e t h e r square waves o f two or m ore frequencies w h i c h are octavely related, s y n c h r o n i z e d a n d have amplitudes inversely proportional to frequency. f 6 9 £ t 2 f Figure 2-8 Staircasing Figure 2-8 shows the simple staircasing of a fundamental tone and its first octave. The n ext note w h i c h could be s t a i rcased is 4f at o n e - f o u r t h the amplitude of f. As mor e octaves are added in this manner, the resulting w a v e f o r m contains a greater number of even harmonics. To see how the even harmonics are introduced consider the a d d i t i o n of f and 2f square waves: the fundamental contains harmonics f, 3f, 5f, If. ... and the octave h i g h e r note contains 2f, 6f, lOf, 14f ... O u t p h a s i n g consists of adding o n e - h a l f the amplitude of an in ­v e r t e d s a w tooth at twice the frequency to the fundamental sawtooth. The r e s u l t is a square wave. The i n t erest of producing sawtooth waves is tha t they are used for prod u c i n g diapasons and strings. Square w a v e s on the other h a n d , c o n t a i n only odd harmonics and are u s e d for h o l l o w w o o d y sounds like tha t of a clarinet [25,27], Squpxe waves are d i r e c t l y p r o d u c e d by flip flop divider stages, and stair c a s i n g is an e asy w a y to form a more u s eful waveform. Conn organs derive three f 1/2 (inverted 2f) __ _ _ J _______ | f + 1/2 (inverted 2f) Figure 2-9 Outphasing w a v e f o r m s directly from their o s c i llator circuits. Both square and sine wa v e s are ke y e d by transistors and the sawtooth is k