! ! A UNIFIED CURRICULUM FOR THE TEACHING OF ELECTRONIC SOUND SYNTHESIS by NATHAN ANDRE PETER VAN DER REST B.S., University of Colorado Denver, 2012 ! ! ! A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulllment of the requirements for the degree of Master of Science Recording Arts 2014
! ! This thesis for the Master of Science degree by Nathan Andre Peter van der Rest has been approved for the Recording Arts Program by ! Samuel McGuire, Chair ! Pete Buchwald ! Lorne Bregitzer ! ! ! ! November 20, 2014 ii
van der Rest, Nathan Andre Peter (M.A, Recording Arts) A Unied Curriculum for the Teaching of Electronic Sound Synthesis Thesis directed by Professor Samuel McGuire ! ABSTRACT The purpose of this thesis is to create a universal curriculum that can be used to teach electronic sound synthesis. Synthesizers have been a staple in modern music since the 1960s. The number of individual synthesizers on the market has grown immensely since their introduction, resulting in a climate that oftentimes leaves the consumer overwhelmed with options. Due to the large number of synthesizers and their unique layouts and naming schemes, many new users shy away from advancing their knowledge of synthesis. Although there are so many individual instruments, the number of synthesis formats, or way in which the instruments generate sound, is much fewer. What is more, the parameters found on individual synthesizers are often times universal and can be found on most instruments despite having different names and utilizing different formats. Due to the universality of synthesizers and their parameters, it is possible for a user to learn synthesis as a concept rather than learning an individual instrument. This allows a user to be procient on any given synthesizer despite its looks, layout, and format. This study was conducted in order to compile a cumulative list of every synthesizer available along with their parameters in order to determine what the average synthesizer is capable of. Once this list was complied, it was then used to generate a curriculum in which to # iii
teach synthesis to new users regardless of the make and model of the synthesizer they have access to. By creating a universal curriculum that encompasses all synthesizer formats, it is then possible to teach synthesis as a concept that can be applied to any individual instrument. As no course or text currently exists which teaches synthesis in this way, it is the purpose of this study to introduce a learning model which can be adopted at the undergraduate university level. ! ! ! ! ! ! ! The form and content of this abstract are approved. I recommend its publication. Approved: Samuel McGuire iv
DEDICATION ! This thesis is dedicated to my mother and father. Throughout my life, they have urged me to pursue my dreams and without them, this thesis surely could not have been done. The love and support they have provided me has been unconditional and has inuenced the course of my entire life. v
ACKNOWLEDGMENTS ! I would like to thank Samuel McGuire for mentoring me through this thesis process and believing in me throughout my undergraduate and graduate careers. I would also like to thank Pete Buchwald and Lorne Bregitzer for agreeing to be on my thesis committee, their feedback and encouragement have been astounding. vi
TABLE OF CONTENTS ! CHAPTER I. INTRODUCTION ...1 ! 1.1 What is Synthesis ..1 1.2 A Brief History of Synthesis ....2 ! 1.2.1 Early Electronic Tone Generation ..2 ! 1.2.2 Teleharmonium .....3 ! 1.2.3 Theremin ....4 ! 1.2.4 Ondes Martenot ....5 ! 1.2.5 Trautonium .....5 ! 1.2.6 Hammond Novachord ..5 ! 1.2.7 Ondioline ...6 ! 1.2.8 Electronic Sackbut ...7 ! 1.2.9 RCA Electronic Sound Synthesizer MKII ..7 ! 1.2.10 Buchla 100 Series Modular Electronic Music System ..8 1.2.11 Moog Modular Synthesizer ...9 ! 1.2.12 Harmonic Tone Generator .....10 ! 1.2.13 Other Early Modular Synthesizers .10 ! 1.2.14 The Rise of the Compact Analog Monophonic Synthesizer ..11 ! 1.2.15 The Rise of the Polyphonic Synthesizer ..12 ! 1.2.16 The Prophet 600 and the MIDI Revolution ..13 ! 1.2.17 The Digital Revolution .14 ! 1.2.18 Software Synthesizers .15 ! 1.2.19 The Analog Resurgence ..15 vii
II. TYPES OF SYNTHESIS ...17 ! 2.1 Subtractive Synthesis .....17 ! 2.1.1 Oscillators ....18 ! 2.1.2 Filters ....20 ! 2.1.3 Ampliers .....22 ! 2.1.4 Envelope Generators .....23 ! 2.1.5 LFOs .....25 ! 2.1.6 Controllers ...25 ! 2.2 Additive Synthesis ......26 ! 2.2.1 Early Additive Synthesizers ..27 ! 2.2.2 Electronic Additive Synthesis ...28 ! 2.2.3 Combination Additive Synthesis ...29 ! 2.3 Frequency Modulation (FM) Synthesis 29 ! 2.3.1 Operators .....30 ! 2.3.2 Modulators & Carriers ....31 ! 2.3.3 Algorithms .......31 ! 2.3.4 Envelope Generators and LFOs ..31 ! 2.3.5 Combination FM Synthesis ...32 ! 2.4 Phase Distortion Synthesis ...32 ! 2.5 Wavetable Synthesis ..33 ! 2.6 Vector Synthesis .33 ! 2.7 Sample Based Synthesis ..34 ! 2.8 Granular Synthesis ..35 ! 2.9 Rompler Synthesis ..36 ! 2.10 Physical Modeling Synthesis ...36 viii
! 2.11 Formant Synthesis .37 ! 2.12 Combination Synthesis .37 III. SIMILAR FUNCTIONS OF THE VARIOUS SYNTHESIS FORMATS ..38 ! 3.1 Oscillators (or Tone Generation Sources) ...39 ! 3.2 Filtering (or Individual Harmonic Amplitude Adjustments) ....40 ! 3.3 Envelope Generators (or Amplitude and Timbre Adjustments Over Time) .42 ! 3.4 Modulation (or Periodic Timbre and Amplitude Variation) .43 ! 3.5 Conclusions ..44 IV. COMPILING THE SYNTHESIZER DATABASE ...45 ! 4.1 Breaking up the Database ..45 ! 4.2 General Section ...45 ! 4.2.1 Synthesis Type ....46 ! 4.2.2 Format ..46 ! 4.2.3 Overall Tune ....46 ! 4.2.4 Glide ..47 ! 4.2.5 Pitch Wheel .....47 ! 4.2.6 Modulation Wheel ...47 ! 4.2.7 Noise .....48 ! 4.2.8 Ring Modulator ....48 ! 4.2.9 Arpeggiator ...48 ! 4.2.10 Sequencer ..49 ! 4.2.11 C.V. Connectivity ...49 ! 4.2.12 MIDI .49 ! 4.2.13 External Signal Processing ..49 ! 4.3 Oscillator Section .....50 ix
! 4.3.1 Number of Oscillators ....51 ! 4.3.2 Sub Oscillator ..51 ! 4.3.3 Sine ...52 ! 4.3.4 Triangle ....52 ! 4.3.5 Sawtooth ..52 ! 4.3.6 Reverse Sawtooth (Ramp) .52 ! 4.3.7 Pulse/ Square ..52 ! 4.3.8 Oscillator Sync ....53 ! 4.3.9 Independent Oscillator Tuning ..53 ! 4.4 Modulation Section ..53 ! 4.4.1 Number of LFOs ..54 ! 4.4.2 LFO(s) Independent of Oscillators ...54 ! 4.4.3 Sine ...54 ! 4.4.4 Triangle .....55 ! 4.4.5 Sawtooth ...55 ! 4.4.6 Ramp .....55 ! 4.4.7 Pulse/ Square ..55 ! 4.4.8 Pitch Routing ...56 ! 4.4.9 Filter Routing ...56 ! 4.4.10 Pulse Width Modulation ...56 ! 4.4.11 Sample and Hold ...56 ! 4.5 Filter Section ....57 ! 4.5.1 Number of Filters ....57 ! 4.5.2 Low Pass Filter ....58 ! 4.5.3 High Pass Filter ...58 x
! 4.5.4 Notch Filter ...58 ! 4.5.5 Band Pass Filter ..58 ! 4.5.6 Resonance ...58 ! 4.5.7 Independent Low Pas Filter Resonance .....59 ! 4.5.8 Independent High Pass Filter Resonance ..59 ! 4.5.9 Designated Envelope .....59 ! 4.5.10 Reverse Envelope Capability ..60 ! 4.5.11 Envelope Amount ..60 ! 4.5.12 Keyboard Track Amount ..60 ! 4.5.13 Pole Select .61 ! 4.6 Envelope Section ....61 ! 4.6.1 Number of Envelopes ....61 ! 4.6.2 Attack ...61 ! 4.6.3 Decay ....62 ! 4.6.4 Sustain ..62 ! 4.6.5 Release ....62 V. INTERPRETING THE SYNTHESIZER DATABASE .63 ! 5.1 General Section ...64 ! 5.2 Oscillator Section .....64 ! 5.3 Modulation Section ..65 ! 5.4 Filter Section ....66 ! 5.5 Envelope Section ....66 ! 5.6 The Average Synthesizer ...67 ! 5.7 Rounding and Outliers ....68 ! 5.8 Reections on the Average Synthesizer ..69 xi
! 5.9 Applying the Average Synthesizer Model in the Classroom ....69 VI. CURRENT SYNTHESIS EDUCATION .71 ! 6.1 Synthesis Literature 72 ! 6.1.1 The Synthesizer by Mark Vail ...72 ! 6.1.2 Rening Sound by Brian K. Shepard ...73 ! 6.1.3 How to Make a Noise Series by Simon Cann ....74 ! 6.1.4 Power Tools for Synthesizer Programming by Jim Akin 75 ! 6.1.5 Becoming a Synthesizer Wizard by Simon Cann ..76 ! 6.1.6 Analog Synthesizers by Mark Jenkins ....77 ! 6.1.7 Sound Synthesis and Sampling by Martin Russ ...77 ! 6.2 Courses in Synthesis ......78 ! 6.3 The Need for a Unied Curriculum ...79 VII. PROPOSED SYNTHESIS COURSE ...80 ! 7.1 Course Lessons ...80 ! 7.1.1 Week 1: Sound & Hearing Review ...81 ! 7.1.2 Week 2: What is Synthesis/History of Synthesis ...82 ! 7.1.3 Week 3: Sound Sources and Combinations 83 ! 7.1.4 Week 4: Amplitude Control .84 ! 7.1.5 Week 5: Harmonic Manipulation ..85 ! 7.1.6 Week 6: Harmonic Control/Filtering Demos ...86 ! 7.1.7 Week 7: Modulation & LFOs ..87 ! 7.1.8 Week 8: Audio Rate Modulation/Midterm Review ..88 ! 7.1.9 Week 9: Analog & Digital Design/Physical Control 89 ! 7.1.10 Week 10: Synthesis Signal Flow/Putting it all Together ..90 ! 7.1.11 Week 11: Synthesis Signal Flow Boot Camp ...91 xii
! 7.1.12 Week 12: Creating PatchesDemonstrations/Exercises 92 ! 7.1.13 Week 13: Sound Recreation Theory/Demos ...93 ! 7.1.14 Week 14: Getting it to Fit in the Mix ...95 ! 7.1.15 Week 15: Synthesis Exploration in Commercial Music ..95 ! 7.1.16 Week 16: Final/Final Project ..96 ! 7.2 Course Assignments/Projects 96 ! 7.2.1 Final Project ....97 ! 7.2.2 Individual Lesson Assignments .97 ! 7.3 Complimentary Text .98 ! 7.4 Course Outcomes 98 VIII. CONCLUSION 100 BIBLIOGRAPHY 102 APPENDIX A: Synthesizer DatabaseGeneral Section ....103 B: Synthesizer DatabaseOscillator Section ..120 C: Synthesizer DatabaseModulation Section ...150 D: Synthesizer DatabaseFilter Section .173 E: Synthesizer DatabaseEnvelope Section ..190 ! ! ! ! xiii
LIST OF FIGURES ! FIGURES 2.1: Subtractive Synthesis Wave Shapes .19 2.2: Subtractive Synthesis Filter Shapes ..21 2.3: ADSR Envelope Generator ..23 2.4: Vector Plane ...34 4.1: Synthesizer Database General Section .46 4.2: Synthesizer Database Oscillator Section ..51 4.3: Synthesizer Database Modulation Section ...54 4.4: Synthesizer Database Filter Section ..57 4.5: Synthesizer Database Envelope Section ..61 5.1: Synthesizer Database Before and After Coding ...63 5.2: Average Synthesizer Block Diagram ..68 ! ! ! ! ! ! xiv
LIST OF TABLES ! TABLES 3.1: Tone Generation Sources of Different Synthesis Formats ..40 3.2: Harmonic Amplitude Control Sources of Different Synthesis Formats .41 3.3: Time Based Timbre and Amplitude Control of Synthesis Format ..42 3.4: Modulation Sources of Different Synthesis Formats 43 5.1: Synthesizer Database General Section Averages ..64 5.2: Synthesizer Database Oscillator Section Averages .65 5.3: Synthesizer Database Modulation Section Averages .65 5.4: Synthesizer Database Filter Section Averages 66 5.5: Synthesizer Database Envelope Section Averages .67 ! ! !!! ! !! xv
CHAPTER I INTRODUCTION !! This thesis aims to create a unied curriculum for teaching all forms of synthesis. In the current environment of synthesis education, each synthesis format is typically taught independent of one another. This method of separating each synthesis format not only instills a narrow lens of synthesis as a whole onto a student, but often leads to each synthesis format being met with dread as it is introduced as an entirely separate technology. ! In reality, each synthesis format is extremely similar to one another and effectively does the same thing through different means. Each synthesis format creates sound and allows users to manipulate it in varying ways. Although the ways in which sound is manipulated varies from format to format, the results are often times the same. With the information presented in this thesis, it becomes apparent that each synthesis format borrows from and compliments the other synthesis formats rather than existing independently from one another. Therefore, it becomes logical to teach synthesis as a concept, independent of the various formats, in order to instill deeper understanding of synthesis onto a student that could not be achieved by teaching each format in succession. ! 1.1 What is Synthesis? ! Synthesis is a means of generating and manipulating sound. At its core, synthesis is the addition or omission of higher order harmonics to a tone in both the amplitude and time spectrums. All synthesis formats do this despite the varying ways in which they accomplish it. In its earliest stages, synthesis was something done in 1
university laboratories and was seen as an esoteric means of sound creation with not much commercial appeal. As time has gone on however, synthesis has left the dark laboratories and has become a common staple in the commercial music world. ! A synthesizer on the other hand, is an instrument capable of producing sound electronically, using one or more of the various synthesis formats The caveat of a 1 synthesizer producing sound electronically is important to note because it is possible for an instrument to create sound using synthesis but without being technically called a synthesizer. ! 1.2 A Brief History of Synthesis ! Synthesis is a fairly old technology dating back to the turn of the twentieth century. The coming of the electric age brought with it new and exciting means of producing sound. The history of synthesis is hard to trace because it is synonymous with the history of electricity. Furthermore, many early advancements of synthesis technology happened by accident and were not recognized as signicant milestones for electronic musical instruments until reected upon many years later. ! 1.2.1 Early Electronic Tone Generation ! One of the rst steps in synthesis history was taken by Elisha Grey in 1876 when he built the musical telegraph (Jenkins, 2007). In essence, the telegraph was a single note tone generator that could be transmitted through telegraph lines. Around the same time, Elihu Thomson invented what became known as the oscillating light arc. Although not intended as a tone generator, the oscillating light arc was found to hum when used. 2 The various synthesis formats are covered in depth in chapter 2. 1
! Recognizing that the oscillating light arc could be used to produced tones, William Duddell wired several light arcs together and performed "God Save the Queen" (Jenkins, 2007). Although Dudell's performance was meant as a novelty, it has become the rst known instance of performance on an electronic instrument. These inventions, although signicant, were not necessarily designed as musical instruments. ! 1.2.2 Teleharmonium ! Arguably the rst electronic instrument designed solely to produce music was the Teleharmonium, built around 1896 (Chadabe, 1997). Thaddeus Cahill designed and built the Teleharmonium while designing an electronic typewriter. Working with the newly invented telephone, Cahill envisioned an instrument that produced sound through electronic means that was capable of being played across telephone lines in order to bring music to hotels and restaurants. The Teleharmonium was a large, multi-room 2 instrument that was controlled via a keyboard. The Teleharmonium had a series of cylinders, divided into sections, that each contained a number of tone wheels. A 3 magnetic coil was placed in close proximity to the tone wheels and as the tone wheels spun, the raised bumps on their surface would pass by the magnetic coil and produce electricity. The space in between the bumps on the tone wheels would yield little, or no electricity. This alternating current produced sound. The cylinders and tone wheels were able to be rotated at various speeds causing a large amount of frequencies to be 3 It is important to note that at the time of the Teleharmonium's invention, there 2 was no means of sound amplication. Cahill theorized that if his instrument produced enough electricity, it would be loud enough to be heard over a telephone headset that could then be connected to a cone like apparatus for amplication. Cahill's studies of sound amplication paved the way for loud speaker design. A tone wheel is a disc like apparatus that contains a series of bumps on its 3 surface.
produced. Multiple frequencies would then be combined together in order to create rich and pleasing tones. Combining multiple frequencies in this way is referred to as additive synthesis and will be covered in depth in section 2.2. ! The rst Teleharmonium Cahill built was slightly smaller than he originally envisioned and contained thirty ve cylinders. After the success of the rst Teleharmonium, Cahill built a much larger and more advanced second Teleharmonium. This second instrument weighed upwards of two hundred tons and featured one hundred and forty-ve cylinders. Cahill eventually built a third larger Teleharmonium but was not met with much enthusiasm as the novelty of the instrument had worn off (Chadabe, 4 1997). ! 1.2.3 Theremin ! Although small advancements were made in the eld of electronic musical instruments after the Teleharmonium, the next large evolution was the Theremin. Invented by Leon Theremin in 1920, the Theremin was the rst, and still one of the only, non-contact electronic musical instruments (Yurji, 2006). The Theremin has two antennas, one for amplitude and one for pitch. The user places theirs hands in the space between the antennas in order to designate the pitch and amplitude of the instrument. The Theremin utilizes two generators of constant and changing frequency which are then transferred to a frequency detector. The resulting signal is a low frequency oscillation which is fed to the amplier. The principles of frequency beating are used in order to create a full range of frequencies without any jumps between individual frequency ranges. 4 The Hammond Tone Wheel Organ, invented in 1935, utilized a similar design to 4 the Teleharmonium. The Hammond organ is also recognized as an early electromechanical additive synthesizer.
1.2.4 Ondes Martenot ! Invented in 1928, the Ondes Martenot was an early electronic musical instrument with a similar sound to the Theremin. Originally invented by Maurice Martenot, the Ondes Martenot was a monophonic single oscillator synthesizer with continuos pitch control (Chadabe, 1997). Although the Ondes Martenot featured a six-octave keyboard, the instrument was most often played via a ring, suspended by wire, that the user would insert their nger into and slide up and down the length of the keyboard. An expression block was used in order to determine amplitude. ! 1.2.5 Trautonium ! The Trautonium was invented in 1929 by Friedrich Trautwein in Berlin (Chadabe, 1997). Like the Ondes Martenot and Theremin, the Trautonium was a monophonic synthesizer that produced sawtooth like sounds via neon-tube relaxation oscillators. The Trautonium was controlled via resistor wire strung over a metal plate. The user would press this wire to generate sound. The harder one pressed, the louder the sound became. Vibrato could be achieved by implementing small nger movements. ! 1.2.6 Hammond Novachord ! The Hammond Novachord is often considered to be the rst commercially successful polyphonic synthesizer. Invented in 1938 by John M. Hanert, C.N. Williams, 5 and Laurens Hammond, the Novachord produced its sound via divide-down oscillator 5 Harold Bode in fact created a predecessor to the Novachord called the Warbo 5 Formant Organ. The Warbo Formant Organ was a four voice keyboard containing two formant lters with advanced envelope control. Bode invented his organ in 1937. It was not adopted commercially until later.
technology and contained over one thousand capacitors as well as one hundred and 6 sixty-three vacuum tubes (Murphy and Kupp, 2013). Besides being the rst successful polyphonic synthesizer, the Novachord is considered to have the rst implementations of envelope generator control (refer to section 2.1.4). The Novachord featured seven discreet envelope shapes that could be selected via a rotary switch on the front plate. The release parameter of the onboard envelope generator was adjustable via a foot pedal controller. The Novachord also contained a three stage band pass lter and an early implementation of an LFO (refer to section 2.1.5) in its vibrato unit. ! 1.2.7 Ondioline ! The Ondioline was invented by Georges Jenny in 1941 (Chadabe, 1997). The Ondioline's circuitry contained vacuum tubes and featured an advanced lter bank with fteen individual sliders in order to create a wealth of complex sounds. The Ondioline's keyboard was suspended on strings which allowed the user to physically shake the keyboard back and forth in order to create vibrato effects. The Ondioline was extremely well suited to create not only a variety of orchestral sounds, but new, never before heard sounds as well. The sounds of the Ondioline were far more varied than with the 7 Theremin, Trautonium, and Ondes Martenot, and it proved quite successful on a number of pop records in the 1940's-1960's. ! 6 Divide-down oscillator technology would later be used in famous analog 6 polyphonic synthesizers in order to achieve a high number of polyphony. The Novachord was among the rst instruments to implement this technology. Another early electronic music instrument, called the Clavioline, was invented in 7 1947 and is extremely similar in function and sound to the Ondioline.
1.2.8 Electronic Sackbut ! Invented by Hugh Le Caine sometime in the 1940's, the Electronic Sackbut was an early electronic subtractive synthesizer (Oxford Handbook of Computer Music, 2009). The Electronic Sackbut was one of the rst synthesizers to offer continuous control over various parameters of the sound. The synthesizer also offered a predecessor to the modulation wheel. Although the Electronic Sackbut never saw commercial production, it is an important evolution in electronic musical instruments. ! 1.2.9 RCA Electronic Sound Synthesizer MKII ! The rst RCA Electronic Sound Synthesizer was created by Herbert Belar and Harry Olson around 1955 (Ernst, 1977). However, The RCA MK II, created in 1957, is much more well known. The RCA synthesizers were the rst programable synthesizers meaning, a user did not need to physically play the instrument. Instead, the RCA synthesizer featured a binary sequencer which used perforated sheets of paper similar to a player piano. The binary sequencer would send instructions to the synthesizer 8 resulting in fully automated playback. The RCA MK II contained twenty-four variable oscillators as well as noise generators. The RCA MK II took up an entire room and was extremely complex to use. Early electronic musicians such as Vladimir Ussachevsky and Milton Babbitt made great use of the RCA MK II but not many composers were granted access to the synthesizer. Although the RCA MK II was never commercially produced, it is recognized as the rst all-encompassing synthesizer that paved the way for synthesizers as we know them today. ! 7 It should be noted that although the perforated paper used to control the RCA 8 synthesizer resembled player piano paper, the functionality was completely different.
1.2.10 Buchla 100 Series Modular Electronic Music System 9 In the early 1960's, electronic music pioneers Morton Subotnick and Ramon Sender placed an ad for someone to create an all new electronic instrument to t the needs of the newly formed San Francisco Tape Music Center. Don Buchla answered the ad and built what would become the Buchla 100 Series Modular Electronic Music System (Manning, 2004). The Buchla system incorporated individual modules for oscillators, lters, envelope generators, ampliers, and sequencers and housed them in a case. The user would then connect various modules together via patch cables. The Buchla system of patching, consisted of both banana cables and mini jack cables which were used for control signals and audio signals respectively. Responding to the fact that electronic instruments such as the Ondioline and Ondes Martenot utilized traditional keyboards, Subotnick made very clear that he wanted a whole new means of control in order to create music that was not inuenced by western tonal music. Buchla then built alternative controllers such as touch plate controllers and sequencers in order to appease Subotnick. Buchla synthesizers have become known as the west coast style of synthesis, meaning sounds are created more organically, using wave shaping technology more than ltering, as well as control via sequencers and random voltage generators. Buchla's synthesizers revolutionized synthesis in that it brought electronic 10 8 Many heated arguments have resulted from trying to determine whether Don 9 Buchla or Robert Moog created the rst voltage controlled analog modular synthesizer. Both Buchla and Moog were working on their respective designs at or around the same time as one another. However, Robert Moog is rumored to have stated that although he might have built individual synthesizer modules rst, Buchla was the rst to put modules together into one encompassing unit. Therefore, I have placed Buchla prior to Moog on my list. The designation of west coast synthesis is due to the fact that Buchla was 10 living in California while Moog, who utilized more standard synthesis techniques, was living in New York. Likewise, Moog's synthesizers were considered to be producing east coast means of synthesis.
music instruments out of laboratories with much more success than the earlier stated electronic music instruments. ! 1.2.11 R.A. Moog Modular Synthesizer ! Dr. Robert Moog began his career in electronics building and selling Theremin kits. Herb Deutsch, a prominent early electronic musician purchased one of these Theremin kits and began talking with Moog about electronic music and the need for electronic musical instruments. Moog then began working on a series of tone producing and modifying modules with the help of Deutsch. In 1964, Moog unveiled his rst modular synthesizer system at the AES convention (Chadabe, 1997). Moog's modular 11 systems are often cited as the rst voltage controlled subtractive synthesizer. The Moog system was completely modular in the sense that systems were custom made to include any module the user wanted, modules could be moved around inside the system, and nally, each module had to be physically connected together via quarter-inch patch cables. ! Throughout Moog's modular years, the company produced a wealth of individual modules including oscillators, lters, envelope generators, ampliers, sequencers, reverb units, and attenuators. The Moog modular was hugely inuential. Walter (now Wendy) Carlos used her Moog modular to record the hugely successful album "Switched On Bach" in 1968. "Switched On Bach" was an album containing a variety of Bach's pieces re-worked in order to be solely performed on the synthesizer. The album not 12 9 AES stands for Audio Engineering S ociety. 11 It should be noted that many electronic albums incorporating synthesizers had 12 been released prior to "Switched On Bach". Although these albums are hugely inuential in the history of electronic music and synthesizers, "Switched On Bach" was the rst to have wide success outside of the world of electronic music enthusiasts.
only proved to be an audio engineering marvel, but went on to have wide acclaim outside of the avant-garde electronic music scene. "Switched On Bach" is often credited as the album that brought synthesizers and synthesis into the spotlight. After the success of "Switched On Bach", a large number of "Switched On" albums were released varying from synthesizer re-workings of pop songs, all the way to synthesizer Christmas albums. Due to the ooding of the market of novelty synthesizer albums, "Switched On Bach" is sometimes viewed as a novelty album itself, but nonetheless, still holds a huge place in synthesis history. ! 1.2.12 Harmonic Tone Generator ! Designed by James Beauchamp at the University of Illinois, the Harmonic Tone Generator is often considered to be the rst all electronic additive synthesizer 13 (Beauchamp, 1966). The Harmonic Tone Generator was capable of generating six harmonics up to two thousand Hz. Although the Harmonic Tone Generator saw no commercial production and was only used at the Experimental Music Studio at the University of Illinois, it is an important milestone for additive synthesis and electronic music. ! 1.2.13 Other Early Modular Synthesizers ! Both Buchla and Moog had a virtual monopoly on the modular synthesizer market throughout the 1960's. This virtual monopoly however, was short lived. The success of both the Buchla and Moog modular systems led to a number of companies producing modular systems of their own in the 1970's. Companies like EMS, EML, 10 An additive synthesizer designed by E.L. Kent in the 1940' s as well as an 13 experimental additive synth designed at Bell Labs preceded the Harmonic Tone Generator but little literature is available on these two machines.
Serge and ARP released modular systems that went on to have large success. The 14 Arp 2500, EMS Synthi 100, EML Electro Comp 200, and Serge modular systems were all widely adopted by musicians and helped secure the synthesizer's place as a respectable musical instrument. 15 1.2.14 The Rise of the Compact Analog Monophonic Synthesizer ! Although the large modular synthesizers of the 1960's and early 1970's were largely successful, their size, cost, and steep learning curves, restrained synthesizers from becoming as widespread as they would eventually become. Because these early synthesizers required users to physically connect each module via patch cables, users had to not only learn synthesis theory, but signal routing as well. Due to this, many users felt the steep learning curve outweighed the advantages these synthesizers promised. The high cost of most of these systems also prohibited the average user from buying a synthesizer. In order to combat these common complaints, Moog began work on what 16 would arguably become the most famous synthesizer of all time; The MiniMoog Model D (Vail, 2014). ! Dr. Robert Moog observed the common modules that users were purchasing for their Moog systems as well as how they were typically patching them together. Using this knowledge, Moog invented a compact synthesizer that had its various modules internally wired together. For the rst time, the user did not have to connect modules together in order to produce sound. The MiniMoog was physically connected to a 11 Roland also released two modular systems of their own in the late 1970' s. 14 It should be noted that the EMS Synthi 100 had an extremely limited run, due 15 to it s immense cost and size, resulting in very few instruments being made. The Synthi 100 however, is extremely sought after and inuenced many future synthesizers. As an example, a Moog modular system could potentially cost as much as a 16 modest house
compact keyboard and introduced the world of synthesizers to the pitch and modulation wheels. Due to the easier learning curve, as well as the much lower price, the MiniMoog became a huge success and nally brought synthesizers to popular culture ! Due to the huge success of the MiniMoog synthesizer, a large amount of companies began releasing compact analog synthesizers. Some of the more successful examples include the ARP 2600, ARP Odyssey, Korg MS-10, Korg MS-20, Roland SH line, Oberheim Two Voice, Octave Cat, Octave Kitten, EDP Wasp, and Yamaha CS line. 17 Throughout the 1970s, compact analog monophonic synthesizers became commonplace. In fact, most popular bands at the time have at least one instance of using a synthesizer in one means or another. Popular music was starting to become inuenced by the synthesizer and leading to a paradigm shift in the music industry. ! 1.2.15 The Rise of the Polyphonic Synthesizer ! Typically, the synthesizers mentioned above were monophonic instruments. Monophonic means that only one note can be played at a time. As synthesizers became more and more popular however, many artists desired the ability to play more than one note simultaneously. Therefore, companies were forced to introduce polyphonic synthesizers into their line of instruments. Throughout the late 1970's and early 1980's, companies responded with the their own polyphonic synthesizers. Some famous examples include the Moog PolyMoog, Sequential Circuits Prophet 5, Yamaha GX-1, Oberheim 4 voice, and Korg's PS line, just to name a few. Due to the unstable nature of 12 This is only a partial list, the actual number of compact analog monophonic 17 synthesizers released in the 1970's numbers in the many hundreds.
analog circuitry, certain manufactures, most notably sequential circuits, started using various digital circuitry in their synthesizers in order to achieve stable polyphony. 18 1.2.16 The Prophet 600 and the MIDI Revolution ! Analog synthesizers were controlled via voltages known as control voltage. Using control voltage, it was possible to control various synthesizers with a single keyboard. The problem with control voltage however, was that not all companies adhered to the 1 volt/octave standard most commonly used. Not adhering to the 1 volt/octave standard caused problems when trying to connect multiple synthesizers together. For example, if one were to try and control a Hz/octave synth such as the Korg MS-20 using a 1volt/ octave synth like the MiniMoog, nothing would work as expected. Therefore, the desire to create a unied communication format amongst synthesizer manufacturers was large. Tom Oberheim and Dave Smith were among the biggest names championing for a unied control format along with companies such as Roland. ! The 1983 NAMM show proved to be the arena of introduction for this type of control format. This format is known as MIDI, or musical instrument digital interface (Manning 2004). MIDI is a digital control format that sends and receives MIDI "messages" in order to control various aspects of a synthesizer. Sequential Circuit's Prophet 600 was the rst synth created which utilized this new control format (Manning, 2004). Although people had their doubts about MIDI, they soon were overshadowed by the vast amount of synthesizers being released that included the MIDI protocol. For the rst time, users were not only able to control various synthesizers using one keyboard, but synthesizer controls such as the lter cutoff knob were soon able to be controlled 13 Also around this time, synthesizer manufacturers were beginning to include 18 preset and patch storage into their synthesizers in order for users to recall patches without resetting every parameter.
using MIDI. MIDI has survived since its introduction in 1983 and is still found on virtually every synthesizer produced today. ! 1.2.17 The Digital Revolution ! After the analog monophonic and polyphonic synthesizers of the 1960's and 1970's had gained monumental status, more complaints were beginning to surface about these instruments. Some of these complaints included the instability of analog circuitry causing users to have to stop playing and tune their synths, as well as the desire to have pre-programmed presets available The answer to these complaints would come in the form of digital synthesizers. ! Although certain analog synthesizers utilized some digital circuitry in their design, the Yamaha DX-7 is most often credited with shifting the synthesizer industry from analog to digital technologies (Vail, 2014). With the introduction of the Yamaha DX-7, 19 users were nally able to plug in their synthesizers and recall a realistic sounding patch instantly without needing to tweak any parameters at all. Due to its digital circuitry, the DX-7 was completely stable and was not susceptible to temperature and pressure changes like its analog counterparts. Although the DX-7 featured advanced synthesis options that could be programmed to create new and exciting sounds, most users only used the onboard presets, effectively using the DX-7 as a rompler synth (refer to section 2.9). After the success of the DX7 most synthesizer companies began introducing their own digital synthesizers. A variety of synthesis formats were introduced during the digital era that include vector, physical modeling, sampling, and rompler. Throughout the 1980's 14 The Yamaha DX-7 utilized FM synthesis (refer to section 2.3). It should be 19 noted that the Yamaha DX-7 was by no means the rst digital synthesizer or the rst FM synthesizer. The DX-7 was however, the most successful digital synthesizer.
and early 1990's, synthesizers were almost exclusively digital instruments despite their changing synthesis formats. 20 1.2.18 Software Synthesizers ! As the modern recording studio moved from analog tape to digital audio workstations, synthesizers went through a similar transformation. Rather than building new hardware synthesizers, companies began introducing software synths that could be used in conjunction with user's digital audio workstations. Software synthesizers gained a lot of traction due to the lower cost as compared to hardware synthesizers. Using software synthesizers allowed users to load all of their synths onto a laptop and take them on the road with them much more easily than if they were to pack up their hardware synthesizers. ! 1.2.19 The Analog Resurgence ! Starting in the late 1990's and early 2000's, many users began buying old analog synthesizers through the second hand market. Despite the problems associated with analog circuitry, many users felt that analog synthesizers are "warmer", and more pleasing. In response to this analog resurgence, certain synthesizer companies began reintroducing all-analog synthesizers back into their product lines. Moog released the MiniMoog Voyager, while Dave Smith of Sequential Circuits released the Prophet '08. New advancements in analog circuitry resulted in new analog synthesizers that were much more stable than their vintage counterparts. As the analog resurgence continues to grow, many more companies are beginning to introduce new analog synthesizers or reissue their analog classic synthesizers. For example, Korg recently released an exact 15 Refer to chapter 2 for a complete list of synthesis formats. 20
recreation of their famous MS-20 analog monophonic synthesizer while Arturia, a company made famous by creating software synthesizers, introduced the MiniBrute and MicroBrute analog hardware synthesizers. In an almost ironic turn of events, new analog modular synthesizers have begun gaining traction as well. ! ! ! ! ! ! ! ! ! ! ! 16
CHAPTER II TYPES OF SYNTHESIS !! Synthesizers produce sound through a variety of different means. Throughout the history of electronic sound synthesis, a number of formats have emerged which in turn, have been adopted by a wide variety of synthesizer companies. Many synthesizers are not limited to just one format and in fact, create sound through a combination of different technologies. As the synthesizer marketplace continues to grow, it is increasingly more important to understand each of the synthesis types one can expect to encounter. ! 2.1 Subtractive Synthesis ! Subtractive synthesis was arguably the rst successful commercial synthesis format. Subtractive synthesis works by removing various aspects of a sound in order to create a desired end sound. The synthesist will start with a rich tone, at full amplitude and will remove different aspects of the tone through the use of lters, envelope generators, ampliers, and modulators. ! Early subtractive synthesizers were created with analog circuitry and were typically large cumbersome machines. As technology improved, subtractive synthesizers began utilizing digital oscillators and then lters along with their analog circuitry in order to give the user more stability and control. Many modern subtractive synthesizers are completely digital and exist as software plug-ins. An analog resurgence has taken place in modern years resulting in a wealth of new analog subtractive synthesizers that utilize new technological advancements in analog circuitry. Although the look and layout of subtractive synthesizers continue to evolve, subtractive synthesis theory remains the same. 17
2.1.1 Oscillators ! In a subtractive synthesizer, sound starts with an oscillator. An oscillator is a circuit which produces sound at a set pitch. Early analog oscillators were controlled using voltages known as control voltage, or CV. Control voltage works by dividing a single volt into twelve equal parts in order to accommodate the twelve tones in an octave. When using CV, a one octave shift then corresponds to a one volt change, while a half step shift results in a 1/12th volt change. Dr. Robert Moog is most often credited with introducing the volt/octave standard found in most analog synthesizers. In order to 21 fully understand what an oscillator does, it is necessary to give a brief description of what makes up sound. ! All sound is made up of a number of separate, related frequencies called harmonics. The way in which these harmonics relate to each other, both in frequency and in amplitude, is called the harmonic series and it determines the resulting sound. The overall sound, when viewed on a visual device such as an oscilloscope, imparts a pattern referred to as a waveform. Although most natural waveforms are rather complex, oscillators are capable of creating a number of basic waveforms that will be used as starting blocks when programming a subtractive synthesizer. Based on the research conducted for this thesis, the most common wave shapes found in subtractive synthesis are sine, triangle, sawtooth, square, and pulse waves. It is important to examine each of these waveforms before moving on. ! The most basic waveform found in subtractive synthesis is the sine wave. A sine wave contains only the rst harmonic, or fundamental frequency. No higher-order 18 Although the volt/octave standard is the most commonly used format for 21 controlling an oscillator, certain synthesizers such as the Korg Ms-10 and MS-20 as well as Buchla synthesizers use their own proprietary means of control with a different Hz/ octave standard.
harmonics are present in a sine wave. A sine wave is impossible to create without using electronic means. Figure 2.1: Subtractive Synthesis Wave Shapes !!! Triangle waves are made up of the fundamental frequency as well as all odd, higher-order harmonics. Therefore, a triangle wave consists of the rst harmonic as well as the third, fth, seventh, ninth, and so on. All harmonics above the fundamental frequency drop in amplitude at a rate that is proportionate to the inverse square of its harmonic number. This means that the third harmonic is one-sixth the amplitude of the fundamental frequency, while the fth harmonic is one-tenth the amplitude of the fundamental Such a rapid decline in amplitude causes very few harmonics above the fundamental to be audible. ! Sawtooth waves consist of all harmonics above the fundamental frequency. A sawtooth wave's harmonics drop in amplitude at an inversely proportionate rate to their harmonic number. In other words, the second harmonic is half the amplitude of the fundamental while the third harmonic is one-third the amplitude. 19
! Similar to a triangle wave, a square wave contains only odd harmonics. The harmonics of a square wave drop in amplitude at a rate that is inversely proportionate to their harmonic number, which results in the third harmonic being one-third the amplitude of the fundamental, while the fth harmonic is one-fth the amplitude. ! A pulse wave is, in essence, a variable square wave. A pulse wave is made up of the same harmonic series as is found in a square wave. The only difference between a square wave and a pulse wave is that a pulse wave' s width in the negative and positive domains is variable. ! Many subtractive synthesizers contain more than one oscillator. The research conducted for this thesis determined that on average, a subtractive synthesizer will contain four oscillators. The user mixes the signals from the onboard oscillators in a mixer section before the resulting sound is routed into the synthesizer's lter. The user can create complex and interesting wave shapes by combining different waveforms at different frequencies from the onboard oscillators. ! 2.1.2 Filters ! Filters are responsible for attenuating frequencies. In most subtractive synthesizers, the lter will be attenuating frequencies from the combined signals of the onboard oscillators. A variety of lter types exist in subtractive synthesis therefore it is necessary to explain each in depth. ! ! ! 20
Figure 2.2: Subtractive Synthesis Filter Shapes !! The most common lter type found in subtractive synthesis is the low pass lter, or LPF. Low pass lters attenuate all frequencies above a user specied point known as the cutoff frequency. The rate at which the lter will attenuate these frequencies is known as its lter response; often times referred to as its slope. The classic Moog lter for example, attenuates frequencies at a rate of 24dB/octave. Another common lter slope is an attenuation of 12dB/octave. Rather than stating the dB/octave lter response, many lters will label their slopes in reference to poles (one pole loosely relating to 6dB/octave attenuation). Using poles as a reference, a 24dB/octave attenuation is known as a four pole lter while a 12dB/octave attenuation is known as a two pole lter. Certain lters allow the user to switch between different lter responses. ! The second most common lter type found in subtractive synthesis is a high pass lter, or HPF. A high pass lter acts just the opposite of a low pass lter in that it 21
attenuates all frequencies below the cutoff frequency. An HPF can have the same lter responses as an LPF only they are not typically user selectable. ! The third lter type found in subtractive synthesis is known as a band pass lter, or BPF. A band pass lter attenuates frequencies above and below a set band of frequencies around the cutoff frequency. ! A notch, or band reject lter, is the exact opposite of a band pass lter, in that it attenuates the center band of frequencies around the cutoff while leaving the frequencies above and below intact. 22 Most lters have a feature called resonance (sometimes referred to as peak or emphasis). A resonance circuit feeds the cutoff frequency back into the lter, effectively boosting the cutoff frequency, and immediate surrounding frequencies, causing the lter to ring. Dr. Robert Moog is often credited with introducing the resonant circuit to subtractive synthesis. ! 2.1.3 Ampliers ! An oscillator only responds to pitch information. Because of this, an oscillator continuously produces sound whether a user wants it to or not. In order to have a synthesizer output sound only when a key is pressed, an amplier circuit must be present. An amplier acts as a gate by either allowing sound to pass or stopping it from passing. In an analog circuit, an amplier responds to a voltage known as a gate. Typically, this voltage is 5v and therefore, the amplier will only allow sound to pass when a +5v gate signal is present. ! 22 It should be noted that many lters allow the user to select the lter type 22 allowing the user to have an LPF, HPF, BPF, and notch lter without the need for multiple lters. It is also not uncommon to see a designated HPF alongside the main lter.
2.1.4 Envelope Generators ! An envelope generator is used as a modulator and most typically is routed to the synthesizer's amplier. An envelope generator has a variety of user adjustable 23 parameters that affect whatever the envelope is routed to in a time based manner. The most common type of envelope generator is known as an ADSR, or attack, decay, sustain, release envelope generators. Some envelope generators may have more, or sometimes less, parameters than the standard ADSR. ! Figure 2.3: ADSR Envelope Generator !! !! The attack control of an envelope generator controls the amount of time it takes for a parameter to reach its user-specied point. When controlling an amplier, attack would determine the amount of time it takes the amplier to rise in amplitude until it reaches the user specied amplitude of the sound. When controlling a lter, attack will determine the amount of time it takes the lter cutoff to reach the user specied 23 Envelope generators are by no means limited to only controlling ampliers. 23 Most synthesizers will feature a designated envelope in which to control the lter's cutoff frequency. Some synthesizers will even feature a third or sometimes fourth envelope generator that can control any number of parameters on the synthesizer.
frequency. The attack parameter is measured in time and is typically adjustable from milliseconds to seconds. ! The decay control of an envelope generator determines the amount of time it takes a parameter to fall from its user specied maximum point, right after the attack, down to its user specied sustain point. When controlling an amplier, the decay parameter will determine the time it takes for the amplitude to drop from its highest point, right after the attack, to its user set sustain point. When controlling a lter, the decay parameter will determine how long it takes the lter to close from its highest point to its user set sustain point. Like the attack parameter, the decay time is typically adjustable between milliseconds and seconds. ! The sustain control sets an amount that a parameter will remain, for the duration of a key being depressed, once the initial attack and decay times have run their course. When controlling an amplier, the sustain control will set a new amplitude that the synthesizer will remain at while a key is being held. When controlling a lter, the sustain control will set a new cutoff frequency that the lter will stay at while a key is being depressed after the attack and decay times are through. ! The release control of an envelope generator determines the amount of time it takes a parameter to fall after a key is released. When controlling an amplier, the release determines the amount of time it takes the sound to fall in amplitude, until it is inaudible, once a key is released. When controlling a lter, the release control determines the amount of time it takes the lter to close completely once a key is released. Like the attack and decay parameters, the release control is typically adjustable between milliseconds and seconds. ! 24
2.1.5 LFOs ! An LFO, or low frequency oscillator, is the main source of modulation in a subtractive synthesizer. An LFO is an oscillator that produces frequencies below the audible range that can be used as a control source. LFOs are often times routed to oscillators or ampliers creating vibrato and tremolo effects respectively. Because an 24 LFO is really just an oscillator, it is capable of producing the typical oscillator waveforms which then can be used as modulation sources. ! 2.1.6 Controllers ! Synthesizers are often thought of as keyboard instruments. It is true that the vast majority of subtractive synthesizers do have a traditional black and white keyboard attached to them, but many other means of control exist in order to control a subtractive synthesizer. Morton Subotnick, a prominent early adopter of synthesis, is often credited with inuencing esoteric controllers for synthesizers. While being a founding member of The San Francisco Tape Music Center, Subotnick commissioned Don Buchla to create what eventually became known as the Buchla Synthesizer. Subotnick urged Buchla to create a means of controlling the instrument in a way that would free the musician from thinking about traditional twelve tone music. In response, Buchla created controllers that included sequencers, touch plates, and push buttons. When Dr. Robert Moog was thinking about using traditional keyboards to control his modular synthesizers, Vladimir Ussachevsky, a prominent early electronic musician, urged Moog not to use a keyboard as a means of 25 LFOs can be routed to a variety of destinations besides just ampliers and 24 oscillators. Two other common destinations are the lter and pulse width of an oscillator. The amount of LFOs available on any given synthesizer, as well as their destinations, are determined by the company which produces the individual instrument.
control as he thought it would force the user into playing piano music. Despite this, Moog decided on using a traditional keyboard and because of this decision, keyboards have become the standard for controlling subtractive synthesizers. However, many other forms of control exist, such as touch plates, ribbons, joysticks, sequencers, push buttons, membrane surfaces, and antennas, which can be used to control a subtractive synthesizer. ! 2.2 Additive Synthesis ! Additive synthesis is as old, or older, than subtractive synthesis. Due to the usability and slightly smaller learning curve, subtractive synthesis became much more prevalent in the early years of synthesis. Additive synthesis works by combining harmonic or inharmonic partials, most commonly sine waves, in order to make a desired tone. As stated earlier, sound is made up of a fundamental frequency and a series of overtones called harmonics. The amplitude at which these harmonics are heard, as well as their amplitude changes over time, is what we perceive as timbre, or the unique quality of an instrument or sound that differentiates itself from other instruments or sounds. Additive synthesis effectively allows a user to build a sound from its most basic foundations; individual frequencies. Additive synthesis has its roots in the Fourier Series, which is a way to represent a wave-like function as a combination of frequencies. The Fourier Series is named after Jean-Baptiste Joseph Fourier, whose studies in the trigonometric series relating to the heat equation, led to the understanding of sound being made up of individual frequencies or harmonics. ! ! 26
2.2.1 Early Additive Synthesizers ! In the late 1800's Thaddeus Cahill was experimenting with what would become known as the Teleharmonium, otherwise known as the world's rst additive synthesizer. Cahill was trying to achieve a way in which to amplify music in a way that could be heard over a telephone line. The Teleharmonium contained thirty-ve cylinders, or tone wheels with raised bumps along the edges. Magnetic coils were then held close to the cylinders. In between the bumps on the cylinder, almost no electricity was produced. Due to the alternation of current, when the cylinders spun, a steady pitch was produced (Chadabe, 1997). The closer the coils were to the cylinders, the louder the pitch would become. Each of the thirty-ve cylinders on the Teleharmonium had a different amount of bumps around the cylinder, causing them to produce different pitches. The Teleharmonium is considered to be the rst additive synthesizer because it was the rst electronic device that added individual frequencies together in order to create a nal sound. ! While the Teleharmonium is considered to be the rst additive synthesizer, the rst successful additive synthesizer is the Hammond Tonewheel Organ. The Hammond Organ is not usually considered to be a synthesizer in the traditional sense, but it uses additive synthesis to create its sounds and therefore it must be discussed. The Hammond Organ uses a series of tone wheels, pre-determined to be harmonic multiples of the fundamental frequency being played, in order to generate its sounds. Each drawbar on the organ corresponds to the next harmonic in the harmonic series. Therefore, by adding in more draw-bars, a user can harmonically richen the sound. ! ! 27
2.2.2 Electronic Additive Synthesis ! Although the Teleharmonium and the Hammond Tonewheel Organ are early examples of additive synthesis, they used mechanical parts in order to create their sounds instead of electronic means that are typically used for synthesis. Many early analog subtractive synthesizers, such as Roland's SH line, utilized a small amount of additive synthesis by allowing users to combine waveforms on a single oscillator in order to create more complex sounds. But it was not until digital signal processing came about that additive synthesis became desirable as a means of sound creation. ! Additive synthesis bodes well in the digital domain due to the high number of tone generation sources needed in order to produce a desirable sound. In the digital domain, a sine wave generator is a relatively simple thing to code allowing additive synthesis to utilize the most amount of harmonics possible; something that would not be capable with traditional oscillator circuits. An additive synthesizer that is only capable of producing harmonics however, is not very useful to a user. Natural sounds not only contain a fundamental frequency and a variety of harmonics, their harmonics change in amplitude overtime. Therefore, individual envelope generators are needed for each harmonic in order to create interesting sounds that do not remain stagnant. A basic useful additive synthesizer will have anywhere from 32 to 64 available harmonics, each with a designated envelope generator and amplier. Additional ampliers will be needed in order to control overall pitch and amplitude. Finally, a number of LFOs would be included in order to modulate individual harmonics or the sound as a whole. By examining the parameters needed to create just a basic useful additive synthesizer, it can be seen that analog hardware is almost out of the question. Therefore, software additive synthesis has grown in popularity and the user is therefore only limited to their computer's CPU power. 28
! When analyzing the sound of a ute or plucked string for example, a series of harmonics changing in amplitude is not the only sound audible. Most sounds have some amount of noise associated with them. Although it is possible to break noise down to its individual frequencies, the amount of tone generators necessary would be enormous. Therefore, many additive synthesizers feature noise generators that can be added into the sound. ! 2.2.3 Combination Additive Synthesis ! Programming an additive synthesizer takes a wealth of time and patience. Creating a sound like a lter sweep can take upwards of thirty minutes for the most experienced of additive synthesists. Therefore, many additive synthesizers offer the user the ability to use lters in conjunction with additive synthesis technology. Incorporating subtractive synthesis techniques into an additive synthesizer makes for a more powerful instrument in the end. ! 2.3 Frequency Modulation (FM) Synthesis 25 Frequency modulation, or FM, synthesis works by taking one frequency and modulating it by another frequency in order to create a desired sound. Although LFOs are used in subtractive synthesis to modulate frequency, FM synthesis uses audio range frequencies for modulation, resulting in new complex tones. Some early analog subtractive synthesizers had the ability to modulate one oscillator with a separate oscillator to produce a complex tone, but this is a very primitive form of FM synthesis. Although FM synthesis is technically possible in the analog domain, the unstable nature 29 Although this form of synthesis is called frequency modulation, its 25 mathematical basis is actually phase modulation. Due to Yamaha's branding of the term FM, the name stuck (Russ, 2008).
of analog circuits would result in pitch instability at normal FM modulation amounts. Because of this, FM synthesis is almost exclusively utilized in the digital domain. ! Digital Implementation of FM synthesis was developed by John Chowning at Stanford University. Chowning patented this technology in 1975 (Russ, 2008). Arguably, the most successful FM synthesizer was the Yamaha DX7. The DX7 revolutionized the synthesizer industry and almost single handedly brought an end to analog subtractive synthesis until their resurgence in the late 1990's. Yamaha adopted FM synthesis soon after its creation and was granted exclusive licensing rights, which caused them to be the sole producers of FM synthesizers until their patent expired in 1995. Other synthesizer companies began introducing FM synthesizers into the market prior to 1995, but due to the high price they had to pay Yamaha, many companies were hesitant to fully adopt the technology, thus resulting in Yamaha having a monopoly on FM synthesis. Due to Yamaha's virtual monopoly on the market, this thesis will use Yamaha designated terminology when dealing with FM synthesis. ! 2.3.1 Operators ! FM synthesis uses a type of digital oscillator known as an operator. An operator is a basic waveform generator that can be used as either a signal source, or a source of modulation in the FM sound synthesis engine. FM synthesizers can have any number of operators but four and six operator models are most common. Typically, operators are only capable of producing sine waves but in more recent years, composite waveforms such as sawtooth and square waves have become more frequently used. ! ! 30
2.3.2 Modulators & Carriers ! In FM synthesis, operators can be used as carriers or modulators. A modulator's frequency will be used to change the sound of the carrier before being outputted. In essence, the carrier is routed directly to the instruments output while the modulator imparts a sonic change to the carrier, changing the overall tonal quality. ! 2.3.3 Algorithms ! Due to the digital implementation of FM synthesis, modulators and carriers can be congured in a multitude of ways. For example, in a four operator synth, one carrier could potentially be modulated by three modulators, each with different frequencies. The various modulators can be routed in different ways causing certain frequencies to modulate a carrier rst, then the new modulated sound be modulated further by the next modulator and so on. The way in which these modulators and carriers are organized is referred to as an algorithm. Most FM synthesizers will offer a variety of onboard algorithms for the user to choo se from. ! 2.3.4 Envelope Generators and LFOs ! Most FM synthesizers will feature a number of LFOs and envelope generators in which to further affect the sound. An overall LFO and envelope generator will typically be included in which to modulate the overall pitch and amplier respectively. It is not uncommon to see separate envelopes and LFOs for each operator in order to provide the user with the most amount of sonic options. ! ! 31
2.3.5 Combination FM Synthesis ! Similar to additive synthesis, many FM synthesizers will feature traditional subtractive synthesis parameters in conjunction with the FM synthesis engine in order to create a more versatile instrument. The most common subtractive parameter found on FM synthesizers is a resonant lter. ! 2.4 Phase Distortion Synthesis ! Phase distortion synthesis is extremely similar to frequency modulation synthesis. Casio introduced phase distortion, or PD, in their CZ line of synthesizers. PD synthesis was conceived as a type of work-around to Yamaha holding the patents for FM synthesis (Russ, 2008). As with FM synthesis, phase distortion uses a modulator and carrier signal. However, PD synthesis utilizes a single modulator and carrier rather than multiple modulators and carriers. While Yamaha's early FM operators were only capable of producing sine waves, PD tone generators create composite waveforms such as sawtooth and square waves, which results in a wide variety of harmonics in the nal sound. ! Besides traditional composite waveforms, Casio introduced a variety of other waveforms such as impulse, half sine, and double impulse waves which provided the user with even more sonic possibilities. Due to the popularity of the Casio CZ line of phase distortion synthesizers, Yamaha began including composite waveforms into their line of FM synthesizers. ! ! 32
2.5 Wavetable Synthesis ! Wavetable synthesis was rst developed by Wolfgang Palm of PPG in the 1970's (Manning, 2004). Wavetable synthesis utilizes various periodic waveforms as a means of sound creation. The amount of waveforms available in a wavetable synthesizer varies from instrument to instrument but waveforms numbering in the few hundreds is not uncommon. The waveforms are then organized with similar harmonic structured waveforms adjacent to one another. Various methods are then used to vary the wave shape overtime, effectively cycling through the various wave shapes. By changing the wave shape overtime, the resulting sound changes smoothly in timbre, at a rate specied by the user, creating sonically pleasing tones. Envelope generators, LFOs and ramp waves are commonly used as the source of movement through the various waveforms. 26 2.6 Vector Synthesis ! Although many companies have produced vector synthesizers, vector synthesis was rst introduced by Sequential Circuits in 1986 for their Prophet VS synthesizer (Manning, 2004). Vector synthesis effectively moves sound around an arbitrary plane called the vector plane. Typically, four sound sources will be placed on the plane, at the extreme corners, and then a source of movement will determine which sound source, or mix of sound sources, will be heard depending on the movement around the vector plane. Usually a joystick is used to provide the cross fading between sound sources. 27 33 Many sample playback synthesizers incorrectly state that their synthesizers 26 utilize wavetable synthesis. Using the term wavetable for these types of synthesizers was introduced as a marketing scheme and is not related to traditional wavetable synthesis. It should be noted that LFOs and envelope generators can also be used as a 27 source of movement in vector synthesis.
Figure 2.4: Vector Plane !! The sound sources used in vector synthesis vary from instrument to instrument. The Prophet VS synthesizer for example uses four wavetable oscillators as its sound sources, while Yamaha vector synthesizers provide both FM and sample playback sources. ! 2.7 Sample Based Synthesis ! Sample based synthesis typically uses standard subtractive or additive synthesis technology in its sound creation. The main difference between sample based synthesis and subtractive or additive synthesis is that the core starting tones are sampled sounds rather than traditional waveforms. ! One of the earliest examples of a sample based synthesizer is the Mellotron (Manning, 2004). The Mellotron is a polyphonic keyboard that uses physical analog tapes of recorded sounds. The tapes move at varying speeds when the user plays up and down the keyboard in order to pitch shift the sounds up or down. As revolutionary as 34
the Mellotron was however, it did not allow the user any accessible means of recording their own sounds to play on the instrument. ! Although instruments like the Mellotron paved the way for sample based synthesis, it was not until New England Digital released their Synclavier systems, starting in the late 1970's, that sample based synthesis came into its own (Chadabe, 1997). The Synclavier allowed users to record their own samples into the system and then change the pitch of the sounds by altering the sampler rate. The Synclavier also had onboard synthesis engines in order to further alter the sound. It is worth mentioning that while the Synclavier was on the market, another digital sampling instrument, the Fairlight CMI, was also available and used similar technology. The Synclavier and Fairlight both helped advance not only sample based synthesis, but digital recording technology as a whole. ! Modern sample based synthesizers allow users to digitally record and edit samples which can then be used in the synth. Users will often be able to assign samples to individual keys or enable pitch shifting technology to track the samples up and down the keyboard. Most Sample based synthesizers will feature a wide array of synthesis parameters such as lters, envelope generators, and LFOs to further alter the sound. Although there are many modern hardware sample based synthesizers, software is more commonly used when dealing with sample based synthesis. ! 2.8 Granular Synthesis ! Granular synthesis utilizes small splices of audio, called grains, in order to generate sound. Each grain of sound is typically one to fty milliseconds in length. Granular synthesis can be compared to sampling, only granular reproduces its samples much differently. Multiple grains can be combined together and played at various 35
amplitudes, phases, speeds, and frequencies. Granular synthesis technology is also used in a number of pitch shifting and time stretching devices. ! 2.9 Rompler Synthesis ! Although rompler synthesis is not, in itself, a form of synthesis, many synthesizers utilize rompler technology. The word rompler is in reference to ROM which stands for read-only memory. A rompler synthesizer will have pre-loaded samples or digital representations of waveforms which can then be manipulated by traditional synthesis parameters such as lters, envelopes and LFOs. ! 2.10 Physical Modeling Synthesis ! A physical modeling synthesizer uses a complex mathematical model in order to recreate the sounds of acoustic instruments. Yamaha worked closely with Stanford University in the late 1980's on the technology behind waveguide synthesis and subsequently released the rst physical modeling synthesizer; the Yamaha VL1 (Oxford Handbook of Computer Music, 2009). ! A physical modeling synthesizer uses a variety of equations and algorithms in order to simulate acoustic instruments. Digital waveguide synthesis is the most common sound generation source found in physical modeling synthesizers. A digital waveguide is a computational model of physical media that when used in a synthesizer, generates a realistic representation of a physical instrument. Although physical modeling synthesizers existed in theory long before Yamaha partnered with Stanford University, it was not until CPU power reached sufcient levels, in the late 1980's, that digital waveguide synthesis was able to be properly executed. 36
! Often times, a physical modeling synthesizer will allow the user to adjust certain parameters that mimic the parameters of the instrument the synthesizer is emulating. A user can sometimes adjust the physical materials of dimensions of the instrument their emulating as well as the plucking of strings and covering of tone holes. 28 2.11 Formant Synthesis ! Formant synthesis is not typically considered a musical synthesis technique but it is important nonetheless. Formant synthesis uses a combination of physical modeling and additive synthesis in order to recreate speech. Formant synthesis has been used in various arcade games, as well as speak and spell type devices, in order to generate speech type sounds. Some synthesizers such as the Yamaha FS1R have 29 incorporated some elements of formant synthesis in order to add vowel or speech type qualities to their sounds. ! 2.12 Combination Synthesis ! Many synthesizers utilize a number of the synthesis formats listed above. In fact, most modern synthesizers, especially software based synths, will almost always feature at least a small amount of features from various synthesis formats. Because of this inclusion of multiple synthesis formats, many companies refer to their synths as combinations synthesizers. Combination synthesis is extremely prevalent amongst large workstation synthesizers. 37 The parameters available to a user are dependent on the synthesizer itself. 28 Some physical modeling synthesizers will also include classic synthesis parameters for the user to adjust, such as lters, envelopes, and LFOs. The speech generated using formant synthesis is often robotic sounding and 29 not very useful for trying to mimic the human voice.
CHAPTER III SIMILAR FUNCTIONS OF THE VARIOUS SYNTHESIS FORMATS !! All synthesizers, despite their different looks, naming schemes, price ranges and synthesis engines, do the same thing; produce sound with continuously adjustable timbres. The way in which timbre is adjusted varies from synthesizer to synthesizer, the core concepts of using synthesis to produce new sounds however, is universal amongst synthesizers of all formats. At its most deconstructed state, synthesis is a means of creating sound by changing the relationships of harmonics in both the amplitude and time spectrums. This is true for each of the synthesis formats. ! In subtractive synthesis, harmonics are adjusted in the amplitude domain via wave shapes and lters and in the time domain via envelope generators and LFOs. In additive synthesis, harmonics are adjusted in the amplitude domain via user set amounts and the time domain via envelope generators and LFOs. In FM and phase distortion synthesis, harmonics are adjusted in the amplitude domain via phase modulation and the time domain via phase modulation as well as envelope generators and LFOs. In wavetable synthesis, harmonics are adjusted in the amplitude domain via wave shapes and lters and in the time domain via envelope generators and LFOs. In vector synthesis, harmonics are adjusted in the amplitude domain via vector point modulation and in the time domain via envelope generators and LFOs. In sample based synthesis, harmonics are adjusted in the amplitude domain via lters and in the time domain via envelope generators and LFOs. In granular synthesis, harmonics are adjusted in the amplitude domain via starting tones and lters and in the time domain via envelope generators and LFOs. In Rompler synthesis, harmonics are adjusted in the amplitude domain via starting tones and lters and in the time domain via envelope generators and 38
LFOs. In modeling synthesis, harmonics are adjusted in the amplitude domain via algorithms and lters and in the time domain via envelope generators and LFOs. And nally in formant synthesis, harmonics are adjusted in the amplitude domain via lters and the time domain via envelope generators and LFOs. ! When examining synthesis in this way, it quickly becomes apparent how similar each of the synthesis formats are with one another. Therefore, an examination of each synthesis format, in relation to the synthesizer database created for this thesis, will be 30 provided in order to fully demonstrate the similarities these various formats hold with one another. ! 3.1 Oscillators (or Tone Generation Sources) ! Each of the synthesis formats generate sound by rst creating tones. Traditionally, the term oscillator was given to the tone producing circuits in the rst synthesizers. Although each format may name their tone generation sources differently, the term oscillator is used in this study in reference to each of the format's tone generation sources in order to create a consistent and universal nomenclature. Table 3.1 illustrates the various types of tone generation utilized in the different synthesis formats. ! ! ! 39 The synthesizer database created for this thesis is examined in depth in 30 chapter 4 and is printed in its entirety in Appendices A-E.
Table 3.1: Tone Generation Sources of Different Synthesis Formats ! 3.2 Filtering (or individual Harmonic Amplitude Adjustments) ! Once tones are created on a synthesizer, timbre must be determined. Timbre is set in a variety of ways, but starts with the addition or omission of higher order harmonics and their relationships with one another, as well as the fundamental frequency. Table 3.2 illustrates the various ways in which harmonic relationships are set amongst the various synthesis formats. Although each format may have a different way in which to set the amplitude relationships of higher order harmonics, the term ltering is used in this study in order to create a universal and consistent nomenclature. ! 40
Table 3.2: Harmonic Amplitude Control Sources of Different Synthesis Formats ! ! 41
3.3 Envelope Generators (or Amplitude and Timbre Adjustments Over Time) ! Once initial timbre is set on a synthesizer, envelope generators, sometimes referred to as contour generators, are used to change timbre and amplitude of the sound over time. Envelope generators controlling lters or higher order harmonics will change the timbre overtime, while envelope generators controlling ampliers will change the amplitude overtime. Although most synthesis formats refer to this type of control as envelope generators, there are a few instances where they are referred to by different names. The term envelope generator will be used throughout this study in order to create a universal and consistent nomenclature. ! Table 3.3: Time Based Timbre and Amplitude Control of Different Synthesis Formats 42
3.4 Modulation (or Periodic Timbre and Amplitude Variation) ! LFOs, or low frequency oscillators, are used to vary the amplitude and timbre of a sound in a cyclical manner. Almost every synthesizer, with very few exceptions, contain traditional LFOs in order to modulate various parameters. ! Table 3.4: Modulation Sources of Different Synthesis Formats ! 43
3.5 Conclusions ! The tables above prove that most synthesizers, despite the synthesis type they utilize, are effectively doing the same thing; creating and manipulating sound. Using the information in the aforementioned tables, it can be shown that teaching each synthesis format separately is in fact counterproductive. This conclusion, although speculative, is determined based from the possible confusion that can result from teaching the same function in different ways. ! For example, when teaching subtractive synthesis, the student will invariably learn what an oscillator is as well as what it does. After subsequent lessons on subtractive synthesis, the denition of an oscillator will hopefully be instilled in a student. After concluding lessons on subtractive synthesis and moving on to FM synthesis, a student will learn about an operator. Although an oscillator and operator are effectively the same thing, it is possible that the student will come to the conclusion that they are quite different because of them being introduced independently of one other. This same theory holds true with the introduction of partials in additive synthesis and grain generators in granular synthesis. ! Therefore, it can be shown that teaching tone generation in one lesson that incorporates oscillators, operators, grain generators, and ROM derived sounds can be benecial to show that each device is effectively doing the same thing; creating a base tone which will then be manipulated by the various parameters of the synthesizer. ! ! ! 44
CHAPTER IV COMPILING THE SYNTHESIZER DATABASE !! In order to create a universal curriculum for teaching synthesis, it was necessary to determine what the average synthesizer was capable of, regardless of synthesis format. I decided to research every commercial synthesizer that I could nd information on, in order to compile a cumulative list of parameters, that could then be examined in order to gauge the capabilities of the average synthesizer. Having a cumulative list is important because it allows one to tailor the curriculum in such a way as to make it relevant to everyone, whether they have access to a low or high end synthesizer. ! 4.1 Breaking Up the Database ! Due to the sheer size of an all encompassing parameter database, it was necessary to split up the database into a number of relevant sections. I found that splitting the database into ve individual sections was sufcient. The ve sections are: General, Oscillator, Modulation, Filter, and Envelopes Each of the ve sections include a number of various relevant parameters. Each synthesizer in the database was meticulously researched and their individual parameters were then entered in the ve sections of the database. ! 4.2 General Section ! The general section of the database includes various overall synthesis and performance control subsections. The subsections are designed to encompass the various synthesizer formats. ! 45
Figure 4.1: Synthesizer Database General Section ! 4.2.1 Synthesis Type ! The synthesis type section is used to determine the synthesis format the synthesizer utilizes. Each synthesizer will fall into one or multiple synthesis formats. 31 4.2.2 Format ! The format section designates whether the synthesizer uses analog, digital, or software control. Certain synthesizers utilize both analog and digital control such as the Sequential Circuits Prophet 5 which utilizes digitally controlled oscillators with analog lter, envelope, and amplier circuits. When a synthesizer utilizes multiple means of control, each format is listed. ! 4.2.3 Overall Tune ! On average, most synthesizers will have multiple tone generation sources whether they are oscillators, operators, or samples. Each of these tone generators will typically be tuned individually. Some synthesizers additionally offer a means of tuning the entire synthesizer while maintaining the tuning proportions of the individual tone generators. The overall tune subsection is used to determine if the synthesizers offer this function. ! 46 Refer to Chapter 2 for the different types of synthesis formats. 31
4.2.4 Glide ! Glide, sometimes referred to as portamento or slew, is a performance control featured on many synthesizers. With glide initiated, notes will rise or fall in pitch to the next note played when the next key is triggered, rather than abruptly jumping to the next pitch. Glide is akin to a trombone sliding from one note to another. ! 4.2.5 Pitch Wheel ! A pitch wheel is a type of performance control found on most synthesizers. A pitch wheel will allow the user to raise or lower the pitch of the instrument to a specied degree. Typically in the form of a physical wheel that is used by the non-playing hand; pitch control can also be in the form of levers, joysticks, or potentiometers. When the synthesizer features a pitch wheel a "YES" is entered. When the synthesizer features another means of pitch control such as a joystick, the type of control will be entered into the corresponding box. ! 4.2.6 Modulation Wheel ! A modulation wheel, most often abbreviated to mod. wheel, is a performance control feature found on most synthesizers. Typically the modulation wheel will control the depth of the LFO. Like the pitch wheel, the modulation wheel can also be in the form of a wheel, joystick, or potentiometer. Likewise, the form of the modulation control will be entered into the corresponding box. ! ! 47
4.2.7 Noise ! Many synthesizers allow users to mix noise into the synthesizer's sound. Noise is found on many of the synthesis formats and is typically a sound source added in the mixer section of the synthesizer. ! 4.2.8 Ring Modulator ! A ring modulator is a signal processing function that in essence, multiplies two signals together in order to create a modulated end sound. Ring modulation was adopted early in electronic musical instruments and has become a staple amongst synthesizers. Many synthesizers in a number of various synthesis formats will feature a ring modulator. Because ring modulation is a staple synth sound, many rompler and sample based synthesizers will feature ring modulation type sounds. Likewise, additive synthesis is capable of generating ring modulation type sounds. Therefore, when a synthesizer features a ring modulator, a "YES" will be entered into the corresponding box and when a synthesizer is capable of producing ring modulation type sounds through a variety of means, "LIKE SOUNDS POSSIBLE" will be entered. ! 4.2.9 Arpeggiator ! An arpeggiator is a performance control function that creates rhythmic melodies. An arpeggiator will take the various notes held down on a synthesizer's keyboard and produce them, one at a time, at a set rate and rhythm. Most analog, digital, and software based synthesizers feature arpeggiators. Although arpeggiation is accomplished through different means, the effect is universally the same. ! 48
4.2.10 Sequencer ! A sequencer is a controller, alternate to the main controller, that produces pitch and timing information in order to play the synthesizer. Sequencers were originally stand-alone units, but began being incorporated onto synthesizers themselves in the late 1960's. As with arpeggiators, analog, digital, and software based synthesizers may feature sequencers. ! 4.2.11 C.V. Connectivity ! Early analog synthesizers were controlled via voltages known as control voltage, or C.V. Many analog subtractive synthesizers both vintage and modern will feature C.V. inputs and outputs for external control. Some digital synthesizers will also offer C.V. inputs and outputs by having an onboard MIDI to C.V. converter. ! 4.2.12 MIDI ! Starting in 1983 MIDI, or Musical Instrument Digital Interface, was introduced as a control standard designed to replace control voltage, in order to have synthesizers communicate with one another and with computers. Most digital synthesizers, as well as modern analog synthesizers, will feature MIDI connectivity. ! 4.2.13 External Signal Processing ! Since the early days of synthesis, some form of external signal processing has been present. External signal processing can be the ability to route external audio into a synthesizer's lter or as a means to control the synthesizer, via external audio, through 49
the use of envelope followers. Sample based synthesizer's will offer external signal 32 processing as a means of recording samples. Many combination synthesizers will utilize these aforementioned means of external signal processing, as well as the ability to record audio for multi-track recording purposes. A "YES" will be entered in the corresponding box if the synthesizer is able to process external audio through its onboard lter or as a means of envelope triggering. When external signal processing in used for recording samples or for multi-track recording, the way in which it is utilized will be entered into the corresponding box. ! 4.3 Oscillator Section ! The oscillator section of the database refers to the synthesizer's raw tone generation capabilities. Each synthesis format will refer to tone generation differently. Subtractive synthesizers feature oscillators. FM and phase distortion synthesizers feature operators. Additive synthesizers feature harmonics, partials, or sine wave generators. Sample based synthesizers feature samples. Granular synthesizers feature granuals. And nally rompler synthesizers feature ROM sounds. Although the various tone producing devices are labeled differently, each refers to the synthesizer's raw tone generator. A synthesizer's tone generator will typically produce waveforms. Subtractive synthesizers will produce the standard synthesis waveforms while an additive synthesizer will typically only produce sine waves. An FM synthesizer can produce sine waves, or in some instances more complex waveforms, while a phase distortion synthesizer will almost always produce complex waves. Rompler synthesizers will also typically have sampled waveforms. Therefore, I have included subsections for each of 50 An envelope follower is a circuit which creates envelope and gate information 32 from audio being fed in at the input.
the different synthesis waveforms. Finally, additional sub sections on basic tone generator functions have been included. ! Figure 4.2: Synthesizer Database Oscillator Section ! 4.3.1 Number of Oscillators ! This sub-section refers to the number of tone generation sources on each synthesizer. Although each synthesis format labels their tone generation sources separately, I have chosen to use the term oscillator to denote tone generators. I have done this not only because it aids in preventing confusion, but also because many synthesizers will use the term oscillator regardless if it truly features a traditional oscillator. ! 4.3.2 Sub Oscillator ! A sub oscillator is a copy of an oscillator that is typically pitch shifted down one or two octaves. Many synthesizers will feature sub oscillators in order to add low end to a sound. When a sub oscillator is present a "YES" is entered into the corresponding box. Certain synthesizers feature more than one sub oscillator, in which case the number of sub oscillators is entered into the box. ! ! 51
4.3.3 Sine ! This sub-section refers to the synthesizer's capability of producing a sine wave. The onboard tone generators capable of producing sine waves will be listed in the corresponding box. When a slightly altered sine wave or non-traditional sine wave is produced, it will be referenced in the box. ! 4.3.4 Triangle ! This sub-section refers to the synthesizer's capability of producing triangle waves. The onboard tone generators capable of producing triangle waves will be listed in the corresponding box. ! 4.3.5 Sawtooth ! This sub-section refers to the synthesizer's capability of producing sawtooth waves. The onboard tone generators capable of producing sawtooth waves will be listed in the corresponding box. ! 4.3.6 Reverse Sawtooth (Ramp) ! This sub-section refers to the synthesizer's capability of producing reverse sawtooth, or ramp waves. The onboard tone generators capable of producing ramp waves will be listed in the corresponding box. ! 4.3.7 Pulse/ Square ! This sub-section refers to the synthesizer's capability of producing pulse and square waves. The onboard tone generators capable of producing pulse and square waves will be listed in the corresponding box. 52
4.3.8 Oscillator Sync ! Oscillator sync is a function that links two oscillators together. When the master oscillator's waveform starts its cycle, it will re-trigger the slave oscillator's waveform despite where it is in its own cycle. Oscillator sync is a common synthesis sound. For this reason, many Rompler synthesizers will include sampled oscillator sync sounds. It is also possible to create oscillator sync type sounds on an additive synthesizer. When traditional oscillator sync is available, a "YES" will be entered in the corresponding box. When oscillator sync type sounds can be produced, "LIKE SOUNDS POSSIBLE" will be entered into the box. ! 4.3.9 Independent Oscillator Tuning ! Many synthesizers feature the ability to tune each tone generator separately. In these instances, a "YES" will be entered into the corresponding box. When a synthesizer features a limited amount of independent tone generator tuning, the word "LIMITED" will be entered into the box. ! 4.4 Modulation Section ! The modulation section of the database refers to the LFOs found on the synthesizer. It is important to note that the word modulation, in this context, is not a reference to any form of modulation synthesis, such as phase modulation, ring modulation, or frequency modulation. This section features a number of sub-sections not only related to LFO functionality, but also to various possible routings of the synthesizer's LFO(s). ! 53
Figure 4.3: Synthesizer Database Modulation Section ! 4.4.1 Number of LFOs ! This sub-section refers to the number of independent LFOs featured on each synthesizer. Some synthesizers had little or no literature available detailing the wave shapes produced by their LFO(s). In these instances, the lack of literature will be referenced in this sub-section. When wave shape could not be determined on an LFO, it was assumed to be triangular due to the triangle wave being the most common LFO waveform. ! 4.4.2 LFO(s) Independent of Oscillators ! Certain synthesizers do not feature a designated LFO. These synthesizers typically allow the user to lower the frequency of one or more oscillators, in order to use said oscillator as an LFO. This sub-section designates if the synthesizer features an independent standalone LFO. ! 4.4.3 Sine ! This sub-section refers to the capability of the LFO(s) to produce a sine wave. If the synthesizer has more than one LFO, each LFO will be listed that is capable of producing a sine wave. When only one LFO is present and it is capable of producing a sine wave, a "YES" will be entered into the corresponding box. ! 54
4.4.4 Triangle ! This sub-section refers to the capability of the LFO(s) to produce a triangle wave. If the synthesizer has more than one LFO, each LFO will be listed that is capable of producing a triangle wave. When only one LFO is present and it is capable of producing a triangle wave, a "YES" will be entered into the corresponding box. ! 4.4.5 Sawtooth ! This sub-section refers to the capability of the LFO(s) to produce a sawtooth wave. If the synthesizer has more than one LFO, each LFO will be listed that is capable of producing a sawtooth wave. When only one LFO is present and it is capable of producing a sawtooth wave, a "YES" will be entered into the corresponding box. ! 4.4.6 Ramp (Reverse Sawtooth) ! This sub-section refers to the capability of the LFO(s) to produce a ramp, or reverse sawtooth, wave. If the synthesizer has more than one LFO, each LFO will be listed that is capable of producing a ramp wave. When only one LFO is present and it is capable of producing a ramp wave, a "YES" will be entered into the corresponding box. ! 4.4.7 Pulse/ Square ! This sub-section refers to the capability of the LFO(s) to produce a pulse or square wave. If the synthesizer has more than one LFO, each LFO will be listed that is capable of producing a pulse or square wave. When only one LFO is present and it is capable of producing a pulse or square wave, a "YES" will be entered into the corresponding box. ! 55
4.4.8 Pitch Routing ! This sub-section refers to the LFO(s) ability to be routed to one or more of the tone generators in order to modulate them. ! 4.4.9 Filter Routing ! This sub-section refers to the LFO(s) ability to be routed to the lter in order to modulate it. ! 4.4.10 Pulse Width Modulation ! This sub-section refers to the ability to vary the width of a tone generator's pulse wave. A "YES" will be entered into the corresponding box if the width of a tone generator's pulse wave can be adjusted either by the LFO or manually by the user. Pulse width modulation is a common synthesizer sound. Due to this, rompler and sample based synthesizers will often times have a sampled pulse width modulated sound. Likewise, pulse width modulation type sounds can be created using additive synthesis. Therefore, if a synthesizer does not feature pulse width modulation but is capable of creating similar sounds, "LIKE SOUNDS POSSIBLE" will be entered into the box. ! 4.4.11 Sample and Hold ! A sample and hold circuit takes gurative snap shots of an incoming signal, then applies these snap shots to a designated parameter. Typically, a sample and hold circuit will have white or pink noise inputted into it, causing it to produce random output 56
voltages onto a given parameter. Because sample and hold is typically used in this 33 way, many synthesizers will feature a random voltage generator although it may or may not be created using a sample and hold circuit. Therefore, if a synthesizer has a designated sample and hold circuit or a random voltage generator, a "YES" will be entered. ! 4.5 Filter Section ! The lter section of the database refers to the lters found on each synthesizer. A number of sub-sections for various lter types and lter functionality are included. Some sub-sections referring to envelope generators are included in this section only when they have direct effect over the synthesizer's lter(s). ! Figure 4.4: Synthesizer Database Filter Section ! 4.5.1 Number of Filters ! This sub-section refers to the exact number of lters featured on each synthesizer. When a synthesizer features a limited lter, no lter, or some type of uncommon lter, it will be referenced in this box. ! 57 The random voltages are caused due to the fact that noise inherently is made 33 up of all frequencies. When the sample and hold circuit takes its snapshot, a different frequency will be apparent at every step.
4.5.2 Low Pass Filter ! This sub-section refers to whether or not the synthesizer's lter is capable of producing a low pass type lter slope. If more than one lter is present, each lter that is capable of a low pass lter slope will be listed. ! 4.5.3 High Pass Filter ! This sub-section refers to whether or not the synthesizer's lter is capable of producing a high pass type lter slope. If more than one lter is present, each lter that is capable of a high pass lter slope will be listed. ! 4.5.4 Notch Filter ! This sub-section refers to whether or not the synthesizer's lter is capable of producing a notch type lter slope. If more than one lter is present, each lter that is capable of a notch lter slope will be listed. ! 4.5.5 Band Pass Filter ! This sub-section refers to whether or not the synthesizer's lter is capable of producing a band pass type lter slope. If more than one lter is present, each lter that is capable of a band pass lter slope will be listed. ! 4.5.6 Resonance ! Resonance is, in essence, a feedback circuit that makes the lter "ring" at the cutoff frequency of the lter. At higher levels of resonance, many lters will begin to self oscillate and produce a sine wave at the frequency of the cutoff. Resonance is a 58
common synthesis staple and will be found on a wide variety of lters of most synthesis types. ! 4.5.7 Independent Low Pass Filter Resonance ! Certain synthesizers, such as the Korg MS-20, will feature more than one lter. In the case of the MS-20, both lters feature resonance with one lter being a low pass and the other being a high pass. When a synthesizer has more than one lter, this subsection will be used to determine if the onboard low pass lter has its own resonance control. If the synthesizer only has one lter, an "N/A" will be entered into the corresponding box. ! 4.5.8 Independent High Pass Filter Response ! This sub-section refers to when a synthesizer's dedicated high pass lter features resonance control. It is a partner sub-section to the previous section and the two are used in conjunction with one another. 34 4.5.9 Designated Envelope ! This sub-section is in reference to the synthesizer's lter(s) having a designated envelope. A "YES" will be entered only when an additional envelope to the main amplitude envelope is permanently routed to the lter(s). If additional envelopes are present and must be routed to the lter(s) by the user, "NO-ROUTABLE" will be entered. In the case of modular synthesizers that must be patched, "NO-PATCHABLE" will be 59 It is important to note that some synthesizers feature more than one lter with 34 each being a multi-mode lter. In these cases, it would technically be possible to set one lter to high pass and one to low pass, each with separate resonances, but because these lters are not solely low or high pass lters, they will not be counted in this subsection.
entered. If a synthesizer has one envelope that is permanently routed to both to the amplier and the lter(s), a "NO" will be entered because this is not a designated lter envelope. ! 4.5.10 Reverse Envelope Capability ! This sub-section refers to the designated lter envelope in the previous subsection. If the envelope is able to have its polarity reversed so that it affects the lter in a negative or positive way, a "YES" will be entered into the corresponding box. ! 4.5.11 Envelope Amount ! This sub-section also refers to the designated lter envelope. If the user is able to adjust the degree to which the envelope affects the lter, a "YES" will be entered. Some synthesizers feature set amounts for the envelope to affect the lter, in these cases, a "YES" will also be entered. A "NO" will be entered only when there is no way to control the depth at which the lter envelope affects the lter. ! 4.5.12 Keyboard Track Amount ! In many voltage controlled synthesizers, the voltages produced from the keyboard are able to control the lter. This means that the higher one plays on the keyboard, the more the lter opens and vise-versa. This type of functionality is also found on digital and software synthesizers that are not controlled via control voltage. If a synthesizer features this function, a "YES" will be entered regardless of it being achieved through control voltage or digital means. ! 60
4.5.13 Pole Select ! The rate at which lters attenuate frequencies is referred to as the lter's "pole". Many synthesizers allow users to select the lter's rate of attenuation and will typically label this feature as pole select or something similar. If a synthesizer features this functionality, a "YES" will be entered. ! 4.6 Envelope Section ! This nal section of the synthesizer database deals solely with the functionality of the synthesizer's envelope generator(s). ! Figure 4.5: Synthesizer Database Envelope Section ! 4.6.1 Number of Envelopes ! This sub-section refers to the total of number of envelopes found on each synthesizer. The number of envelopes featured on each synthesizer will be entered into the corresponding box. ! 4.6.2 Attack ! This sub-section is used to determine if the synthesizer's envelope(s) contain a user adjustable attack. Each envelope generator will be listed that has a user adjustable attack time. If one of the envelope generators contains set attack times rather than a fully adjustable attack time, the word "LIMITED" will be inserted after the corresponding envelope generator. 61
4.6.3 Decay ! This sub-section is used to determine if the synthesizer's envelope(s) contain a user adjustable decay. Each envelope generator will be listed that has a user adjustable decay time. If one of the envelope generators contains set decay times rather than a fully adjustable decay time, the word "LIMITED" will be inserted after the corresponding envelope generator. ! 4.6.4 Sustain ! This sub-section is used to determine if the synthesizer's envelope(s) contain a user adjustable sustain amount. Each envelope generator will be listed that has a user adjustable sustain amount. If one of the envelope generators contains set sustain amounts rather than a fully adjustable sustain amount, the word "LIMITED" will be inserted after the corresponding envelope generator. ! 4.6.5 Release ! This sub-section is used to determine if the synthesizer's envelope(s) contain a user adjustable release. Each envelope generator will be listed that has a user adjustable release time. If one of the envelope generators contains set release times rather than a fully adjustable release time, the word "LIMITED" will be inserted after the corresponding envelope generator. ! ! ! 62
CHAPTER V INTERPRETING THE SYNTHESIZER DATABASE !! The synthesizer database, found in appendices A-E, was created in order to determine the average commercial synthesizer's capabilities. As stated in the previous chapter, each section and subsection of the database was chosen in order to ascertain the most amount of information possible, while keeping it relevant to all synthesizers. Using this database, a pattern emerges that can be used to determine the capabilities of the average synthesizer. ! The database was coded in order to ascertain averages and patterns that emerged amongst individual instruments. Subsections that included numbers were averaged, while subsections that required yes or no answers were converted into a binary format. Converting yes and no entries into ones and zeros allowed the columns to be ltered and counted, in order to determine if more synths featured a particular parameter or not. An example of this coding technique can be found in the gure below. ! Figure 5.1: Synthesizer Database Before and After Coding ! ! 63
5.1 General Section ! The table below will serve to display the rounded averages of parameters found on each synthesizer examined, relating to the general section of the synthesizer database. ! Table 5.1: Synthesizer Database General Section Averages ! 5.2 Oscillator Section ! The table on the following page will serve to display the rounded averages of parameters found on each synthesizer examined, relating to the oscillator section of the synthesizer database. ! ! 64
Table 5.2: Synthesizer Database Oscillator Section Averages ! 5.3 Modulation Section ! The table below will serve to display the rounded averages of parameters found on each synthesizer examined, relating to the modulation section of the synthesizer database. ! Table 5.3: Synthesizer Database Modulation Section Averages 65
5.4 Filter Section ! The table below will serve to display the rounded averages of parameters found on each synthesizer examined, relating to the lter section of the synthesizer database. ! Table 5.4: Synthesizer Database Filter Section Averages ! 5.5. Envelope Section ! The table on the following page will serve to display the rounded averages of parameters found on each synthesizer examined, relating to the envelope section of the synthesizer database. ! ! 66
Table 5.5: Synthesizer Database Envelope Section Averages ! 5.6 The Average Synthesizer ! Using the averages ascertained from the synthesizer database, patterns begin to emerge that show the capabilities of the average synthesizer. This research has determined that on average, synthesizers feature four independent oscillators that can produce triangle, sawtooth, and square waves. Each oscillator is capable of having independent pitches but cannot be synced. Next, the research shows that most synthesizers feature three ADSR envelope generators; one routed to the amplier, one routed to the lter, and a nal one being user assignable. These envelope generators are also capable of having their polarities reversed. Further, it has been found that most synthesizers feature two independent low frequency oscillators that are capable of producing triangle and square waves. These LFOs can be routed to the oscillators for pitch modulation and to the lter cutoff for lter modulation. In addition to this, both pulse width modulation and sample and hold capabilities are present. It has also been found that the average synthesizer contains a single resonant low pass lter with keyboard tracking and envelope amount controls. In addition, the research shows that the average synthesizer features overall pitch control, glide, a pitch shifting control of some sort, a noise generator, and MIDI connectivity. ! 67
Figure 5.2: Average Synthesizer Block Diagram ! 5.7 Rounding and Outliers ! When averaging the information in the synthesizer database, a few precautionary measures were taken in order to ensure not only an accurate result, but a result which could be applied to a synthesizer that can be adopted and used in a classroom setting. Firstly, when averaging subsections in the general section, I converted any type of pitch shifting and modulation amount control such as joysticks, levers, and potentiometers into a 1 in the pitch wheel subsection. I did this to ensure that all forms of pitch and modulation control were included. Next, in the oscillator section, I removed the additive and granular synthesizers from the number of oscillators average. The reasoning behind this decision was that while most subtractive, FM, phase distortion, sampling, rompler, physical modeling, vector, AM, and wavetable synthesizers have single digit sound 68
sources, additive and granular synthesizers can have sound sources numbering in the hundreds. If I had included additive and granular synthesizers when averaging the number of individual oscillators, or sound sources, the number of individual oscillators would be much greater than most synthesizers can offer, because granular and additive synthesizers make up a small percentage of the synthesizers in the database. Finally, I rounded the amount of individual oscillators, lters, envelope generators, and LFOs in order to ascertain a whole number than can be used when creating a synthesis course. 35 5.8 Reections on the Average Synthesizer ! I found the results in this section both predictable and surprising. I had assumed that on average, synthesizers would feature triangle, sawtooth, and pulse waves as well as noise generation, pitch shifting, pitch and lter modulation, glide, and two ADSR envelopes. I was surprised to see that on average synthesizers did not have a modulation control, nor feature oscillator sync capabilities. I was also extremely surprised to see that the average number of individual oscillators and LFOs was higher than I had originally thought at four and two respectively. My assumption had put those numbers at two and one respectively. Finally, I was surprised to see that on average, pulse width modulation and sample and hold capabilities were also present. ! 5.9 Applying the Average Synthesizer Model in the Classroom ! A University course based off of the information in this thesis is presented, in full, in chapter seven. This University course will not only serve to impart a deep understanding of synthesis onto students, but will also provide a wealth of hands on 69 I have included the original decimal numbers in parenthesis after the newly 35 rounded numbers in the corresponding tables above.
demonstrations and exercises where students use synthesis. Building this course around this average synthesizer idea is crucial for a number of reasons. Firstly, knowing what the average synthesizer is capable of allows the instructor to understand the limitations that might be present in the synthesizers some students might have access to. By creating a course around the average synthesizer model, an instructor can accommodate for the differences in each student's individual synthesizers. Secondly, by utilizing the average synthesizer model, a professor can be aware of the capabilities of any software synthesizer they might chose to use in class. The average synthesizer model allows instructors more freedom when teaching because the theories taught will be accessible by most students, rather than only being ideas that cannot be realized on their individual synthesizers. ! ! ! ! ! ! ! 70
CHAPTER VI CURRENT SYNTHESIS EDUCATION !! The current state of synthesis education, in both the classroom and in literature, is somewhat limited when examining the full scope of synthesis technology. Most courses and books segregate each synthesis format from one another and teach them separately. Although this segregation approach may be successful when only teaching one format, it proves repetitive and confusing when teaching each synthesis format in succession. ! The main concern in current synthesis education is that segregating each synthesis format can impart unwanted biases onto students. For example, once a thorough knowledge of subtractive synthesis is gained, all subsequent synthesis formats taught will be examined through the lens of subtractive synthesis. This proves troublesome because rather than appreciating that other synthesis formats are performing similar functions through different means, students will instead be comparing each format to subtractive synthesis. This will cause students to attempt to justify why each separate format exists as well as how they must be superior, or inferior, to subtractive synthesis. ! Another concern in current synthesis education is the unproductive repetitiveness inherent in teaching each synthesis format in succession. For example, when teaching subtractive synthesis, students will learn how lters are used to lower the amplitude of higher or lower order harmonics, in order to affect the overall timbre of the sound. Once subtractive synthesis has been covered in full and the course moves on to additive synthesis, students will learn about manually raising and lowering the amplitudes of higher and lower harmonics, in order to affect the overall timbre of the sound. In 71
essence, these two lessons are near identical. By teaching them as separate entities however, some students may nd it difcult to remember all aspects of both, since they are inputting them into memory as separate things and at separate times. This confusion increases when the same lesson is taught in different ways for each of the synthesis formats. Therefore, this thesis proposes to teach parameters such as harmonic amplitude adjustments together for each synthesis format. When teaching synthesis in this way, the student is free to absorb harmonic amplitude control in each format together, rather than having to recall information later in the course, after being taught a variety of other parameters in the interim between the lessons. ! 6.1 Synthesis Literature ! Because synthesis has been taught in literature for a number of decades, it is important to examine some of the key books currently available on the subject. The following pages will detail select pieces of literature available to prospective students, in order to examine how each approach the teaching of various synthesis formats. 36 6.1.1 The Synthesizer by Mark Vail ! Mark Vail's The Synthesizer acts as a learning guide for both new and experienced synthesists. Although Vail explores each of the various synthesis formats, the book mainly centers around subtractive synthesis, especially in the modern modular format. The book begins with a history of sorts that examines some of the famous synthesizer models, while exploring the various synthesis formats in no apparent order. The book makes no mention of the similarities of the various synthesis formats, save for 72 The books listed in this section are in no particular order and were selected 36 based off of their relevancy to this thesis, their use as course textbooks, their inclusion of the various synthesis formats, and their reviews.
a few small mentions of overlapping technologies. The Synthesizer is a common example of current synthesis literature because it takes a single format and runs with it, while making brief mention to other synthesis formats. ! The Synthesizer provides the reader with very basic explanations of common synthesis parameters, such as oscillators, lters, and envelopes. The denitions given for various synthesis parameters are often short and are typically centered around subtractive synthesis. When other synthesis formats are covered, common synthesis parameter terms are not used, and therefore, do not provide the reader with an information bridge in which to compare the various formats and their similarities. The Synthesizer although extremely well written and researched, is insufcient for teaching synthesis as a whole. ! 6.1.2 Rening Sound by Brian K. Shepard ! Rening Sound is perhaps the closest book to this thesis. Shepard presents a way of teaching synthesis independent of format, in order to teach synthesis as a theory rather than a way of creating sound on a particular instrument. The book starts with a brief history of synthesis and then quickly moves into the core content of the book. Rening Sound is divided into chapters which correspond to certain aspects of synthesis. For example, there is an oscillator chapter, an oscillator combining chapter, a lter chapter, an envelope chapter, a modulation chapter and so on. This proves to be an effective way of teaching synthesis because Shepard is free to examine oscillators found on all types of synthesis formats in a single chapter, while the information is fresh in a reader's mind. ! The chapters on oscillators, as well as combining oscillators, are perhaps the most effective chapters in conveying this unied teaching method. In these two chapters, 73
Shepard shows that an oscillator is still an oscillator, whether it is used in subtractive, additive, granular, modeling, or FM synthesis. Although Shepard does not cover all of the synthesis formats, the reader quickly begins to understand that all synthesis formats utilize some form of an oscillator that acts identical despite the format. However, after the oscillator chapters, Shepard seems to revert to only teaching about subtractive synthesis while mentioning other formats less frequently in each subsequent chapter. This is perhaps due, in part, to the fact that the software synthesizer which accompanies the book, and is used to reinforce ideas in each chapter, acts most like a subtractive synthesizer than any other format. ! The diminishing mention of each synthesis format within each chapter is unfortunate, since the book held so much promise in the beginning. One saving factor that still makes this book relevant in regards to this thesis however, is the inclusion of a week to week lesson plan at the end of the book. This lesson plan was designed by Shepard in order to provide instructors with a format to follow for teaching synthesis as a theory. Since the lesson plan is vague, it can be used quite successfully in order to teach synthesis concepts without conning them to specic synthesis formats. ! 6.1.3 How to Make a Noise Series by Simon Cann ! The How to Make a Noise series by Simon Cann is a three volume set detailing how to create sound using synthesizers of varying formats. The three books are How to Make a Noise: Analog Synthesis, How to Make a Noise: Frequency Modulation Synthesis, and nally, How to Make a Noise: Sample-Based Synthesis. Before opening these books, it becomes apparent that the synthesis formats are segregated entirely from one another. It is also suspect that only subtractive (referred to by Cann as 74
"analog"), FM, and sample-based synthesis are covered while not mentioning any of the other synthesis formats. ! The How to Make a Noise series is however, successful in teaching the three aforementioned synthesis formats and covers the various parameters of each in depth. Each book breaks down its corresponding synthesis format and gives an in-depth explanation of each common parameter found. These books are perfect examples of the current state of synthesis education which teaches each synthesis format separately, rather than as a whole. ! 6.1.4 Power Tools for Synthesizer Programming by Jim Akin ! Similar to Mark Vail's book The Synthesizer Power Tools for Synthesizer Programming by Jim Akin acts as an all purpose synthesis tutorial text. Although the history of synthesis is covered in Power Tools of Synthesizer Programming the book focuses more heavily on the theory and practice of synthesis rather than signicant advancements in the technology. The book offers in-depth explanations of the various parameters found on most synthesizers as well as a look at the various synthesis formats. The book makes small mention of the similarities of the various synthesis formats but does not provide the reader with any signicant ways of learning all synthesis formats as a whole. ! Quite similar to the other books mentioned in this section, Power Tools for Synthesizer Programming is most often viewed in the lens of subtractive synthesis and therefore, only provides the reader with information on subtractive synthesis. Other synthesis formats are certainly discussed, but in the viewpoint of additions to subtractive synthesis rather than similarly powerful and useful synthesis formats in and of themselves. 75
! Perhaps the greatest strength of the book is that it is presented in a multi-media format, giving readers access to various sound and song examples, which in turn reinforce the information provided in the text. The book, although perhaps more useful than the Simon Cann's How to Make a Noise series and Mark Vail's The Synthesizer, is still a limited resource for learning all aspects of synthesis and serves as a traditional educational text for the current state of synthesis education. ! 6.1.5 Becoming A Synthesizer Wizard by Simon Cann ! Becoming a Synthesizer Wizard goes into more depth than Cann's other work, How to Make a Noise. This book not only covers more synthesis formats than the three covered in How to Make a Noise but also covers individual parameters with much more clarity and context. The book has a strong emphasis on software-based synthesizers and provides little information, outside of the synthesizer history section, on hardware synthesizers. The book details not only the parameters one might expect to nd on a synthesizer, but the concepts behind them as well as techniques to utilize them when creating sounds. ! Becoming a Synthesizer Wizard does however utilize the traditional segregation technique of teaching the various synthesis formats. Not only does the book segregate the various synthesis techniques, it solidies the notion that the various synthesis formats are indeed extremely different from one another, which only serves to implant this idea further into the reader's mind. Becoming a Synthesizer Wizard is by no means limited in its content. However, Becoming a Synthesizer Wizard is quite limited in the scope of this thesis and in teaching synthesis as a theory independent of synthesis format. ! 76
6.1.6 Analog Synthesizers by Mark Jenkins ! Analog Synthesizers is perhaps the most limited title in this list, but due to its presence in classrooms, it must be included. As the title suggests, Analog Synthesizers covers synthesizers built with analog circuitry; both vintage and modern. Although other formats such as additive and FM synthesis have been utilized on analog equipment, this book limits its scope to subtractive synthesis. ! Analog Synthesizers provides the reader with well written and thorough explanations of each aspect of a subtractive synthesizer, as well as a fairly detailed history of analog subtractive synthesis. Due to its in-depth analysis, easy to comprehend layout, and relatively low cost, it has been the featured text in a number of synthesis based university courses since its release. Due to its narrow scope however, it neglects to cover the various other synthesis formats, and like most of the other books in this section, imparts a narrow view of synthesis onto its readers. ! 6.1.7 Sound Synthesis and Sampling by Martin Russ ! Perhaps one of the more technical books in this section, Sound Synthesis and Sampling is an in-depth examination of the various forms of synthesis and how they work. Unlike some of the other selected books in this section, this text does not attempt to apply synthesis to a musical setting and instead focuses on how the various synthesis formats work. Despite giving brief mention to the various synthesis formats being of the "Source and Modier" model, the book segregates each of the synthesis formats as a whole. ! Sound Synthesis and Sampling represents the current state of synthesis education by segregating each of the formats. However, it does more than most synthesis books, save for Rening Sound in showing a limited view of the similarities of 77
the various formats. By separating synthesis from music, the book succeeds at teaching synthesis as a concept or theory rather than a musical instrument, but does so in a limited manner, resulting in the book still falling into the common concerns of modern synthesis education. That being said, the explanations provided in the book are extremely educational and are well tailored for the classroom. Another concern, besides the segregation of synthesis formats, is that the book denes certain parameters multiple times in a chapter, each time revealing more information. Although it can be said that this repetitive technique helps solidify information, it often times is more confusing and can even persuade a student to skim a repeated section or skip it entirely resulting in the new information being lost on the student. ! 6.2 Courses in Synthesis ! Current University courses offered on the topic of synthesis follow a similar pattern to the current literature available. Based on the courses examined, each one separated the various synthesis formats into separate lectures. Electronic music and synthesis courses taught at the University of Colorado Denver, Berklee College of Music, New York State University, Brown University, University of Texas at Austin, Indiana University, Duke University, and the University of Illinois were examined. None of the aforementioned schools offer courses which teach the various synthesis formats as one. 37 ! 78 When researching these courses and programs, syllabi were requested from 37 faculty. In the cases of syllabi not being available, online course descriptions were examined.
6.3 The Need for a Unied Curriculum ! Both the literature and courses listed above teach synthesis in a format segregated manner. The research in this thesis has shown that despite the synthesis format, all synthesizers are in essence performing the same function; creating and manipulating sound through electrical means. Although each synthesis format is extremely similar in function and outcome, they are still taught as separate entities. This form of segregated teaching not only imparts bias onto the students, but also disillusions students in certain cases. When learning each synthesis format separately, many students may feel apprehension when learning the next subsequent format. The reasoning behind this feeling is resultant of students feeling like they have overcome a hurdle in learning the rst format and feeling almost worn out when starting to approach the next format from the beginning. This feeling of apprehension can be avoided by simply teaching the various formats together. It has been shown in chapter three that each synthesis format's parameters can be categorized together because of the similar functions they perform. Therefore, it would seem logical to teach each synthesis format's parameters together in various related lessons. By teaching synthesis as an all encompassing theory, students will be more willing to learn each format simply because it is not presented as a completely different entity, but rather as an extension of the entity which is being taught. ! ! ! 79
CHAPTER VII PROPOSED SYNTHESIS COURSE ! The idea that each synthesis format is closely related has been introduced in the previous chapters. Due to the close symmetry the various synthesis formats hold with one another, a curriculum which teaches the various elements needed for sound manipulation, regardless of synthesis format, can prove to be extremely benecial. Therefore, it is the purpose of this chapter to provide suggestions for instructors on how to cover the topic of synthesis, while utilizing this all-inclusive format. This course has been created as a standalone synthesis course, which exists in an undergraduate recording arts degree. Although intended for a recording arts degree, the course could easily be adapted to accommodate a music technology or music performance degree. Students should have lab access to workstations containing computers with music production and synthesis software installed. Individual student workstations in the classroom is highly suggested. ! 7.1 Course Lessons ! A comprehensive lesson plan has been created, in order to introduce a cumulative curriculum for teaching synthesis, which incorporates the research presented in this thesis. The structure of this course, along with the content, assignments, 38 projects, demonstrations, and hands on learning opportunities, has been tailored to utilize this new form of synthesis education. ! 80 This lesson plan is based on a seventeen week, semester long course model 38 which meets twice a week. Sixteen week long lessons have been provided to accommodate a one week break.
7.1.1 Week 1: Sound & Hearing Review ! This course begins with a comprehensive review which covers sound and hearing. The topics covered in this lesson include a review of sound waves and how they are created, the way in which the ear interprets sound, overtones and the harmonic series, a harmonic analysis of the standard wave shapes, and nally, a digital audio refresher. The digital audio refresher will focus on sample rate, bit depth, le format, aliasing, and digital le storage. Beginning the course with a review on sound and hearing is needed to reinforce critical information learned in other courses that will be built upon in the duration of this course. For example, when discussing sound sources in week three, a basic understanding of the harmonic series will be required in order to teach how to build simple waveforms such as sawtooth and square waves. ! Both lessons in week one will be in lecture format. Students will be encouraged to participate in discussions as well as answer questions posed by the instructor. The sound and hearing review lesson is designed to reintroduce students to concepts they have previously learned as they will both be relevant and necessary in this course. An assignment should be assigned where the students must calculate the rst six harmonics of each standard synthesis wave shape. The students should nd both the frequencies and amplitudes of each harmonic present for each wave shape. The instructor will need to provide the amplitude and frequency of the fundamental frequency which will be applied to each wave shape. The goal of this assignment is to instill the fundamentals of standard synthesis wave shapes into the students. ! ! 81
7.1.2 Week 2: What is Synthesis/History of Synthesis ! The rst class of week two will begin with a discussion on what synthesis exactly is. Students will be called upon to argue how a string section or brass section could technically be considered a form of synthesis. This exercise will aid in the realization that synthesis is a concept, independent of synthesizers, which will become useful later in the course when students are introduced to synthesis theory independent of synthesis format. The discussion will then move to the coining of the term synthesizer for electronic musical instruments which will lead into a lecture on the history of synthesis. The second class in week two will continue with the history of synthesis and will focus on key gures, inventions, and instruments in synthesis' rich history. A brief history of synthesis is important to teach because it helps illustrate that each synthesis format is in actuality, an evolution of a previous format, rather than a completely new and different entity. ! Lesson one in week two will be call and response discussion based. Students will be encouraged to both answer questions and participate in lively discussions on the topic of synthesis versus synthesizers. Students will be encouraged to think critically about scenarios in which synthesis is utilized without synthesizers. Lesson two will be presented in a lecture format which covers the various milestones in synthesis history. Pictures as well as audio examples of individual instruments should be provided by the instructor. If the university has access to any historic synthesizers, they should be displayed for the students to examine. ! ! 82
7.1.3 Week 3: Sound Sources and Combinations ! Week three will begin with a lecture on sound sources being the fundamental building block of all synthesizers. Oscillators, both voltage and digitally controlled, operators, grains, partials, samples, and wavetables will all be discussed and demonstrated throughout the week. The students will be introduced to these sound sources in the same lesson in an attempt to demonstrate that despite synthesis format, every synthesizer is built around fundamental sound sources. By grouping each synthesis format's sound source into one lesson, the student will hopefully gravitate away from any preconceived notions that the various synthesis formats are independent from one another. Once each sound source is explained and demonstrated, the various ways in which to combine sound sources will be covered. The combination aspect of this lesson will include combining single harmonics together in order to create simple waveforms, combining simple waveforms together in order to create complex waveforms, and nally, combining complex waveforms together. The students will then get to combine various waveforms and harmonics on a synthesis program of the professors choosing in order to hear the results. 39 The rst lesson of week three will be lecture and demonstration based. Once the professor covers a topic, a relevant demonstration of said topic should be provided. A modern DAW such as Logic Pro, Ableton Live, or Pro Tools would be sufcient for these demonstrations. The demonstrations should focus on the similarities of sound sources in each synthesis format as well as various ways in which to combine these tone sources. The second lesson of week three will continue where the rst lesson left off and lead into a student exercise. Students will be instructed to create the standard synthesis wave 83 Although the synthesis program is left up to the professor, Native Instrument's 39 Reaktor, Camel Audio's Alchemy and Max/MSP are all suggested programs for this exercise.
shapes using additive synthesis. A software synthesizer such as Loom or Alchemy should be used if available. If the instructor or lab tech is uent in Max/MSP, this software would also be benecial for creating an environment where students can combine simple sine waves. ! 7.1.4 Week 4: Amplitude Control ! Once a thorough understanding of sound sources is gained, the students will learn about ways in which to control these sound source's amplitudes. A lecture on traditional oscillator design, in which sound is always present, will be given in order to show the need for amplitude control for playability. Ampliers and gate signals will be introduced as a means of creating sound only when desired. Sound examples of various acoustic instruments will then be provided, which in turn, will be analyzed by the professor and the students. The purpose of these sound examples is to show that instruments typically do not just produce and discontinue sound instantaneously but instead, have a rise, peak, and fall to their amplitudes. Once this is understood, envelope generators will be introduced and each envelope stage will then be covered in depth. Envelope demonstrations will be provided using a variety of the sound sources covered in the previous lesson. A portion of the second class during the week will be set aside for students to experiment with amplitude envelopes in order to better understand their function. ! The rst lesson of week four will continue with the discussion of tone production in the form of traditional analog oscillators. By introducing students to the fact that traditional oscillators produce sound regardless of a key being depressed, the lecture will naturally move on to the need for ampliers and gate signals. The lecture will then move on to the need for amplitude control, namely envelope generators, in order to convey 84
natural progressions of sound. Envelope shapes of various instruments should be examined. The lecture will then move onto a demonstration of an envelope generator that is controlling an amplier. The second lesson of the week will continue with examples of envelope generators and their parameters. Students should then be given ample time on a software synthesizer in order to adjust envelope parameters and discover the way in which they affect the amplitude of the synthesizer. Nearly any software synthesizer which comes stock in any major DAW, will work for this exercise. ! 7.1.5 Week 5: Harmonic Manipulation ! This lesson will focus on the next most important aspect of synthesis after sound sources; the manipulation of higher order harmonics, or ltering. Rather than introducing students to lters in the traditional subtractive synthesis sense, students will rst gain a thorough knowledge of why it is necessary to either add or subtract higher order harmonic content in order to create a desired sound. Sound examples will be provided demonstrating how the timbre of sound changes with the addition, or omission, of higher order harmonics. Once this knowledge is instilled, traditional synthesis lter types such as low pass, high pass, band pass, and notch lters will be discussed. A lecture on resonance will be given at this time as well. While lecturing about and demonstrating the common synthesis lter types, students will learn about the many similarities between these lters and the EQs used in their other classes. Students will then get to experiment with manipulating higher order harmonics in a number of software settings. Once a 40 thorough understanding of higher order harmonic manipulation is attained, the lecture 85 The software used for this exercise will be determined by the professor but 40 AIR's Loom and Max/MSP are suggested.
will move on to using lters musically outside the realm of synthesis, such as their use with external audio and over full musical tracks. ! The rst lesson of week ve will mainly be presented in lecture and demonstration format. A discussion on timbre changes through harmonic manipulation will be provided. The instructor should go over the common synthesis lter shapes and discuss the merits of each. Once the instructor feels that a basic understanding has been conveyed to the students, he or she should begin giving demonstrations of the various lter shapes in order for students to begin recognizing them. A brief lecture on resonance and resonant circuits should be provided once the basic lter shapes are covered. The second lesson in week ve will consist of an exercise in which students build the basic lter shapes using additive synthesizers. Any additive software synthesizer will be suited for this lecture but Loom is recommended. The goal of this exercise is to allow students to fully understand the mechanisms and theories at work behind the common synthesis lter shapes. ! 7.1.6 Week 6: Harmonic Control/Filtering Demos ! Week six's lesson introduces the concept of controlling higher order harmonic manipulation by other than manual means. Similar to amplitude control, the lectures on harmonic control will focus on the use of multi-stage envelope generators. Each stage of an envelope generator will be examined in depth to determine how it affects timbre when controlling harmonics and lters. Students will have the ability to experiment with these principles. The second class of the week will deal with more harmonic manipulation and ltering demos which will not only reinforce the knowledge of the ways in which to manipulate harmonics, but also help students identify the different sounds created by the various methods. 86
! The rst lesson in week six will be in lecture format. A lecture on using envelope generators to control lters will be provided along with a wealth of instructor demonstrations and audio examples. The second lesson of week six will be entirely set aside for student exercises. Students should be encouraged to explore the possibilities of each lter shape as well as the sonic changes that occur from increasing resonance. Students should also be encouraged to experiment with lter envelope shapes. An inclass assignment should be given to encourage attendance. This in-class assignment could be creating complex lter envelope shapes while strictly using additive synthesis, in order to further solidify student understanding of harmonic manipulation and control. ! 7.1.7 Week 7: Modulation & LFOs ! Once a thorough understanding of sound sources, amplitude control, harmonic manipulation, and harmonic control has been attained, the course will move on to modulation and low frequency oscillators. The lesson will begin with a lecture pinpointing the similarities and differences between an oscillator and low frequency oscillator. Traditional vibrato and tremolo effects through the use of LFOs will be examined, as well as LFO routings to the lter, partial, grain density, pulse width and panning parameters of a synthesizer. Using LFOs to cycle through vector planes and wavetables will also be discussed. Students will learn the waveforms they can expect to encounter when using LFOs, as well as the ways in which each waveform will affect the parameter they are routed to. An immense amount of audio examples and demonstrations will be provided, as well as a large amount of time for student experimentation using LFOs. Finally, the lesson will end with an exploration into some of the more complex modulation matrixes available on past and current instruments. 87
! In similar fashion, the rst lesson of week seven will be comprised of a lecture and demonstrations. The lecture will encourage student participation and will actively engage students through call and response as well as discussions between students and the instructor. Once the lecture is nished, the instructor will provide the students with audio examples and demonstrations on the various uses of low frequency oscillators. The second lesson of week seven will begin with a brief lecture on some complex modulation matrices found on some digital synthesizers and then move on into the student exercise portion. The exercise for week seven will involve students experimenting with various LFO routings. The instructor should assign students to create common modulation effects such as tremolo, vibrato, and wah, effects. During the exercise, the instructor should walk around the room and answer questions and assist students as needed. ! 7.1.8 Week 8: Audio Rate Modulation/Midterm Review ! Continuing the topic of modulation, week eight's lesson will involve modulation using audio range modulators. The rst class of the week will introduce students to using audio rate sound sources in order to modulate other sources. The lesson will include extensive explorations into frequency and phase modulation, as well as oscillator sync capabilities and ring modulation. The lesson will evolve from simple modulator-carrier models to high order modulator and carrier models, which include not only frequency modulation, but ring and amplitude modulation as well. The second class of the week will be split into two parts; the rst part being comprised of audio examples and hands-on demonstrations of audio rate modulation, while the second part will act as a review of the material covered thus far in the course. 88
! The rst lesson in week eight will serve to continue the discussion on modulation with a lecture on audio rate modulation. Topics such as ring modulation, amplitude modulation, oscillator sync, and frequency modulation will all be covered. The difference between modulators and carriers will be examined, as well as the differences between synthesizers which utilize single modulator carrier pairs and multiple modulator carrier pairs. The end of the rst lesson as well as the beginning of the second lesson will be set aside for students to experiment with audio rate modulation. The remainder of the second lesson will be comprised of a midterm review. Students will be encouraged to bring questions to the discussion as well as topics that might need more clarication. ! 7.1.9 Week 9: Analog & Digital Design/Physical Control ! Week nine begins the second half of the course wherein the material covered veers from sound manipulation parameters into design, theory, and practice of synthesis. The rst class of the week will focus on analog and digital circuitry. Control voltage and MIDI will be covered extensively as well as the pros and cons inherent with analog and digital instruments. A brief lecture on connecting various synthesizers together will also 41 be given. The second class of the week will wrap up any analog and digital design topics not covered in the rst class and then move into physically controlling synthesizers. The physical control section of the lesson will explore traditional piano style keyboard controllers, both in control voltage and MIDI formats, as well as esoteric controllers such as touch plates, antennas, ribbon controllers and pressure plates. A lecture on step sequencers and arpeggiators, both analog and digital, will wrap up the lesson. 89 It is suggested that the professor bring both an analog and digital synthesizer 41 into the class in order to aid with the lesson.
! The beginning of the rst class in week nine will be dedicated to the midterm exam. Once the exam is completed, a lecture on the differences between analog and digital design in synthesizers will be discussed. The pros and cons of each format will be examined as well as a discussion which seeks to determine if students have a preference to one format over the other and if so, why. The second lesson of week nine will be in lecture format and consists of a discussion which covers various ways in which to control a synthesizer. Topics such as traditional black and white keyboards, controller wheels, arpeggiators, sequencers, ribbons, joysticks, and touch plates will all be explored. If possible, the instructor will grant the students access to an esoteric synthesizer controller. If a physical controller is unattainable, an iPad app such as Moog's Animoog will sufce. ! 7.1.10 Week 10: Synthesis Signal Flow/Putting it all Together ! Week ten's lesson takes all the information covered thus far in the course and applies it to creating sound on a synthesizer from start to nish. The Lesson will begin with a lecture and demonstration on synthesis signal ow. The students will be introduced to traditional synthesis signal paths and the reasoning and theories behind them. The students will then be encouraged to come up with individual inter-connections that will then be tested on a software such as Max/MSP. The second class of the week will revolve around individual synthesizer signal paths. The professor will then deconstruct the signal paths of a number of commercial synthesizers he or she is familiar with on a computer or white board. By examining the signal paths on commercial synthesizers, students will learn why certain aspects of a synthesizer are connected to others and why certain circuits fall in line before, of after, others. 90
! The rst lesson of week ten will be in lecture and demonstration format and will attempt to take all of the individual functions and theories learned thus far in the course and combine them. Students will learn about synthesis signal ow through demonstration and lecture by the professor. Discussions on why certain synthesizer parameters are typically placed before others in a signal chain will be examined. The second lesson of week ten will be example driven. The instructor should deconstruct various famous synthesizers' signal ows on the white board or projector for the students to examine. By deconstructing various synthesizers' signal ows, students will be able to gain a thorough understanding of synthesis signal ow in the real world. ! 7.1.11 Week 11: Synthesis Signal Flow Boot Camp ! Week eleven's lesson revolves around imparting a deep understanding of synthesis signal ow and patching onto the students. It is highly recommended that the professor bring in a small modular or semi-modular synthesizer such as a eurorack system, Synthesizers.com setup, or an ARP 2600. By having students move physical 42 patch cables in order to connect various aspects of a synthesizer, a deeper understanding of synthesis signal ow can be attained as they work through problems and troubleshoot the system. These two class meetings will be solely comprised of hands on patching by the students. It is recommended that the professor let the students work together in order to solve any patching problems, but should be available to assist if needed. Synthesis signal ow is such a crucial aspect of synthesis that it is benecial to spread this exercise over both class meetings, in order to give each and every student a hands on opportunity. 91 If a physical modular system is not available, many iPad apps and software 42 exist which mimic a modular system. Arturia's Moog Modular software and the iPad app Modular are recommended.
! Both lessons in week eleven will be demonstration and hands on exercise based. The lessons will serve to instill a deeper understanding of synthesis signal ow onto students through hands on experience. It is highly recommend that the instructor have access to a small modular synthesizer of some sort. A variety of new modular synthesizers are available on the market which are reasonably priced. Therefore, a Synthesizers.com system, Eurorack system, or even Arp 2600 is highly recommended. If a physical modular synthesizer is unattainable, a software or iPad app will sufce. ! 7.1.12 Week 12: Creating PatchesDemonstration/Exercise ! Week twelve will revolve around the instructor demonstrating approaches for creating patches on a wide variety of synthesizers, as well as a hands-on exercise for the students. It is recommended that the lesson be divided into bass, lead, pad, sequences, and sound effect sections. During each division, the instructor will demonstrate how he/she creates these types of sounds on a large number of different synthesizers. It is desirable that the various synthesizers used will not only be of differing layouts and companies, but also price points. By covering a wide array of different synthesizers, a knowledge of obtaining the most amount of exibility on a budget or high end synthesizer can be gained. Both hardware and software synthesizers are recommended for this demonstration. Once the creating patches demonstration has been completed, the following lesson will revolve around students making their own patches on their personal computers or provided workstations. The exercise should be divided into the same sound categories as the demonstration from the previous lesson. The students will be given a certain amount of time to create a patch in each of these categories. During the lesson, the instructor should be available to answer any questions and should walk around the room frequently in order to provide encouragement and tips 92
to individual students. Once the week is through, the professor should create a section on the class website where the patches can be uploaded, giving students the opportunity to download and play with other students patches. ! Week twelve has divided lessons where the rst is strictly lecture and example based while the second is strictly demonstration and hands on exercise based. The rst lesson will focus on creating patches. The instructor should provide examples and dene patch sounds such as lead, bass, drum, pad, and sound effect sounds. Once each type of patch is explained, a discussion on ways to create these patches should be provided. A wealth of audio examples should be given by the instructor in order to familiarize the students with these types of sounds. The second lesson will focus on the instructor giving examples of how to create these same patch sounds and then leave the students to create patches on their own. The instructor should allow the students to choose two types of patches they want to create and then give them the remainder of class to create said patches. While the students are creating their patches, the professor should walk around the room and answer questions and assist students as needed. Once the students have completed their patches, the patches should be exported and uploaded to the student website such as Canvas. ! 7.1.13 Week 13: Sound Recreation Theory/Demos ! All of the lessons taught thus far will aid in week thirteen's lesson of re-creating sounds. The purpose of this lesson is to teach students how to build a patch on any given synthesizer that sounds like an existing instrument or sound. Having to build a sound from scratch, that must sound similar to an existing sound, is an extremely valuable lesson as it teaches students to think critically about each aspect of the synthesizer and its parameters. It also forces students to think deeply about what they 93
want the synthesizer to do, rather than just having them adjust parameters and stumble upon sounds they like. The rst class of the week will involve a lecture about the things students must pay attention to in the original sound in order to mimic it. FFT and 43 harmonic analyses are recommended in order to demonstrate the rich harmonics of each sound. The instructor should then demonstrate this re-creation technique on a synthesizer of their choosing. The second class of the week will have students recreating the sounds of their choice. 44 Week thirteen will be structured the same as week twelve. The rst lesson will be completely lecture and example based while the second will be demonstration and hands on based. The professor should begin the lesson with an examination of common instruments using some type of spectral imaging software such as an FFT. The instructor will then cover the various ways one would go about creating a particular instrument on a synthesizer. This lesson will cover a lot so the entire class period should be designated for it. The second lesson will begin with the instructor showing how to go about recreating instrument sounds. The second half of the class should be reserved for students to re-create their own chosen instrument. Once again the instructor should be available to answer questions and assist students as necessary. The patches the students create should be exported and uploaded to the student website so that they can be examined by the rest of the students. ! ! 94 An FFT, or Fast Fourier Transform, is a visual representation of the harmonics 43 present in a given sound. The sound or instrument the student attempts to recreate should be chosen by 44 the student but should be approved by the instructor.
7.1.14 Week 14: Getting it to Fit in the Mix ! Now that students have gained a thorough knowledge on synthesis and how to create their own patches, a discussion and demonstrations will be provided in order to teach students tricks and theories behind getting a particular synth sound to t in a mix. First, students will learn about quantization and humanization when using MIDI, as well as audio tempo adjusting tools when not using MIDI. Next, students will learn about spectral and dynamic effects such as equalization, compression, and limiting. Finally, time based effects such as reverb, delay, chorusing, anging, and phasing will be discussed. The students should already have a thorough knowledge of how spectral, dynamic, and time based effects work and what they do from their other music production courses. The scope of this course, in regards to these effects, is to simply provide the students with tips and tricks on using these effects in order to get a synth sound to t better in a mix. Additionally, this information can be used to build upon a synth sound with non-synthesis devices. ! Week fourteen will alternate between lectures, examples, and demonstrations. Discussions on how to get particular synthesis sounds to t nicely into a mix will be provided along with related audio examples. The instructor should demonstrate the variety of techniques covered in the lesson in a wide array of musical genres. Ample discussion on how nuances present in a particular synth patch will change when it is inserted into a mix, should be provided. ! 7.1.15 Week 15: Synthesis Exploration in Commercial Music ! Week fteen will involve an analysis of synthesized sounds present in commercial music. The lecture should begin with a history of synthesis in commercial music ranging from musique concrete pieces to early experimentations by Morton 95
Subotnick, Vladimir Ussachevsky, and Walter/Wendy Carlos. Next, an analysis on the changing role and sounds of synthesizers in commercial music from the sixties to present should be given. Select pieces of music chosen by either the instructor or the students should be examined and discussed. This exercise is helpful in order to provide students with a baseline of how synthesizers are used in commercial music, as well as open up their minds to potential synthesis uses that are not usually thought of in regards to commercial music. ! Resembling, week fourteen, week fteen will consist of lectures and examples. A brief history of the synthesizer's role in commercial music as well as lm scores should be provided in lecture format with relevant audio examples. Once the history section is covered, an exploration of varying synthesis sounds amongst different genres of music should be provided as well. A generous amount of audio examples should be provided by the instructor to help solidify the information being discussed. ! 7.1.16 Week 16: Final/Final Project ! The nal week of the course should consist of a nal cumulative exam as well as a presentation and discussion of each student's nal project. Each student's project should be introduced by the student and then a discussion involving constructive input by the students should be led. The instructor will provide their in-depth feedback privately. ! 7.2 Course Assignments/Projects ! This proposed synthesis course will include a number of assignments and projects that coincide with the lessons being taught. In addition to lesson related assignments, a cumulative midterm and nal exam will be given along with two quizzes. 96
Lastly, a nal project will be assigned at the beginning of the semester that should be worked on throughout the course by the student. ! 7.2.1 Final Project ! The nal project for this course will involve each student creating a piece of music in which each instrument is created solely through the means of synthesis. Students will be expected to create each sound from the ground up without relying on presets. The minimum sounds expected to be created are kick drum, snare, bass, rhythm, and lead. Students however, are urged to create more sounds than just these. The students may select a synthesizer or program of their choice but they must be able to demonstrate that they used a variety of techniques learned throughout the course while completing their projects. If students do not feel comfortable creating a piece of music, they may elect to create these sounds and prepare them in a sampler format such as Logic's EXS24. ! 7.2.2 Individual Lesson Assignments ! Depending on the access students have to workstations in the class room, some of the lesson exercises can be adapted as out of class assignments. As this course is not intended for a particular school, the resources each university has available will be unique. If this course is taught at a university with limited or no student in-class access to synthesis programs, a single program which is capable of most, if not all, of the synthesis techniques covered throughout the course should be selected by the 97
instructor. Individual synthesis software or Apps might also serve as a substitute if full 45 or student versions of digital audio workstations are out of reach for the university or the students. ! 7.3 Complimentary Text ! At this time there is no text that exists which fully compliments this course. That being said, the book Rening Sound: A Practical Guide to Synthesis and Synthesizers by Brian K. Shepard is perhaps the most closely related book. Reading should be 46 assigned before each lesson so the students can come to class with some knowledge on the subject covered each day. In addition to Rening Sound: A Practical Guide to Synthesis and Synthesizers a book, chosen by the instructor, which details the history of synthesis is recommended. 47 7.4 Course Outcomes ! This course is designed to provide students with a comprehensive, allencompassing knowledge of synthesis. This course breaks new ground by doing away with traditional synthesis education in which synthesis formats are treated and taught as separate technologies. Instead, this course brings the various synthesis formats together to teach synthesis as a whole. This course aims to demonstrate that despite year 98 The state of currently available software is constantly changing, as is their 45 price points and academic licensing availabilities. A suitable software should be selected based on the resources available to a particular university at the time of teaching the course. At this point in time, Logic, Ableton Live, or Reason are suggested as desirable programs. Rening Sound: A Practical Guide to Synthesis and Synthesizers is covered in 46 depth in section 6.1.2 Although the text should be left up to the instructor, the book The Synthesizer 47 by Mark Vail (section 6.1.1), Analog Synthesizers by Mark Jenkins (section 6.1.6), and Electric Sound by Joel Chadabe are recommended.
manufactured, format, layout, and design, each synthesizer is, in essence, performing the same function of sound manipulation. This course will provide students with the tools necessary to condently create sound on any synthesizer given to them. This course will engage students to think critically through hands-on exercises and assignments, as well as in-class discussions and collaborations. It is my belief that a course which follows this structure will provide students with a deeper comprehension of synthesis than current synthesis courses are able to offer. ! ! ! ! ! ! ! ! ! 99
CHAPTER VIII CONCLUSION !! Sound synthesis is an old technology which continues to stand the test of time. Throughout the decades since its inception, synthesis has gone through many evolutions and re-inventions. Despite its many technological advancements, synthesis has always been, and always will be, a means of creating and manipulating sound through electrical means. Since its inception at the turn of the century, to its mass commercial adoption in the 1960's and 1970's, to its evolution from analog to digital circuitry in the 1980's, to its software implementation in the 1990's, and nally to its analog resurgence in the 2000's, synthesis has continued to be a staple in popular music, lm, and games. ! Throughout the years, synthesis technology has been presented in a wide variety of forms such as subtractive, additive, FM, phase distortion, wavetable, vector, sample based, granular, rompler, modeling, and multiple combinations of each. Despite this wide array of synthesis formats however, synthesis as a whole is still just a means of creating and manipulating sound. Because of this, synthesis has not changed at its core but merely in its technology and implementations. ! This thesis has served to present a new methodology in the teaching of synthesis which focuses on the act of synthesizing sound itself rather than focusing on individual implementations. By focussing on synthesis as a concept, this methodology is able to be taught decades into the future while still maintaining relevancy. What is more, this new methodology will impart a deeper understanding of synthesis onto a student than is possible by simply teaching one or two synthesis formats in a single course. Students 100
will be able to take the information gained and apply it to virtually any synthesizer from the oldest models, to ones not yet invented. ! In order to create this new methodology of synthesis education, each synthesis format has been explored in depth and compared to each subsequent format in an attempt to show their similarities. Likewise, a database of commercially available synthesizers, as well as their functions and parameters, has been created in order to tailor the curriculum to accommodate the average synthesizer. Current synthesis education and literature has been examined in order to create what they have not yet accomplished; which is an all encompassing synthesis education. ! This new methodology for teaching synthesis has been crafted into an undergraduate course structure and has been presented as a course outline and lesson plan. Lecture topics, as well as demonstrations, exercises and assignments have been proposed that can be adopted by a music technology or production instructor. It is my belief that a course such as the one presented in this thesis, is necessary in order to advance the knowledge of sound synthesis in music technology and production programs. ! Sound synthesis will remain a crucial part of the music and entertainment industries for the foreseeable future. By adopting the curriculum presented in this thesis, music production and technology programs, which teach synthesis, will be able to maintain their relevancy as technology continues to change. 101
BIBLIOGRAPHY ! Akin, Jim. Power Tools for Synthesizer Programming. San Francisco, California: Backbeat Books, 2004. Beauchamp, J.W., "Additive Synthesis of Harmonic Musical Tones," Journal of the Audio Engineering Society 14 no. 4 (October 1966): 332-342. Cann, Simon. Becoming a Synthesizer Wizard: From Presets to Power User. Boston, Massachusetts: Cengage Learning, 2010. Cann, Simon. How to Make a Noise: Analog Synthesis. UK: Coombe Hill Publishing, 2011. Cann, Simon. How to Make a Noise: Frequency Modulation Synthesis. UK: Coombe Hill Publishing, 2011. Cann, Simon. How to Make a Noise: Sample-Based Synthesis. UK: Coombe Hill Publishing, 2011. Chadabe, Joel. Electric Sound: The Past and Promise of Electronic Music. New Jersey, US: Prentice-Hall, Inc., 1997. Ernst, David. The Evolution of Electronic Music. New York, New York: Schirmer Books, 1977. Jenkins, Mark. Analog Synthesizers Oxford, UK: Focal Press, 2007. Manning, Peter. Electronic and Computer Music: Revised and Expanded Edition. New York, New York: Oxford University Press, Inc., 2004. Murphy, Michael and Kupp, Eric, "An Examination of Early Analog and Digital SamplingThe Robb Wave Organ Circa 1927", (paper presented at the annual Audio Engineering Society Convention, Rome, Italy, May 4-7, 2013). Russ, Martin. Sound Synthesis and Sampling. Burlington, Massachusetts: Focal Press, 2008. Shepard, Brian K. Rening Sound: A Practical Guide to Synthesis and Synthesizers. New York, New York: Oxford University Press, Inc., 2013. "The Oxford Handbook of Computer Music", New York, New York: Oxford University Press, Inc., 2009. Vail, Mark. The Synthesizer: A Comprehensive Guide to Understanding, Programing, Playing, and Recording the Ultimate Electronic Music Instrument. New York, New York: Oxford University Press, Inc., 2014. Yurii, Vasilyev, "History and Design of Russian Electro-musical Instrument Theremin' ", (paper presented at the annual Audio Engineering Society Convention, Paris, France, May 20-23, 2006). 102
! ! ! ! ! APPENDIX A SYNTHESIZER DATABASE GENERAL SECTION ! ! ! ! ! ! ! ! ! ! ! ! 103
! ! ! ! ! APPENDIX B SYNTHESIZER DATABASE OSCILLATOR SECTION ! ! ! ! ! ! ! ! ! ! ! ! 120
! ! ! ! ! APPENDIX C SYNTHESIZER DATABASE MODULATION SECTION ! ! ! ! ! ! ! ! ! ! ! ! 150
! ! ! ! ! APPENDIX D SYNTHESIZER DATABASE FILTER SECTION ! ! ! ! ! ! ! ! ! ! ! ! 173
! ! ! ! ! APPENDIX E SYNTHESIZER DATABASE ENVELOPE SECTION ! ! ! ! ! ! ! ! ! ! ! ! 190