Content ProSyn Print [PDF]

Zitiervorschau

Peter Gorges Programming Synthesizers

Peter Gorges

Programming Synthesizers

w

Author

Peter Gorges

Translation Herb Quick Editor Reinhard Schmitz Cover art M0type, www.motype.de Interior design & layout Uwe Senkler © 2005 Wizoo Publishing GmbH, www.wizoo.de ISBN

3-934903-48-7

All rights reserved. No part of this book may be reproduced in any form or by any electronic or mechanical means, including information storage and retrieval systems, without permission in writing from the publisher. All product names and company names mentioned in this book are either trademarks or registered trademarks of their respective owners.

Welcome To say that the synthesizer plays a major role in my life would be a monumental understatement. Since carrying home my first hard-earned Korg MS-20 at 13, synthesizers have accompanied me on stage and in the studio as a musician and arranger. In addition, I have written several books and countless articles about them, programmed sound banks for manufacturers and at some point even became involved in the development of several synthesizers. If I had to name one of these activities as my true passion, it would have to be sound programming. It’s one of the very few things that can cause me to forget everything else and awaken hours later out of a trance to find that day has turned to night, my coffee has turned cold and I have missed yet another appointment. I get an enormous amount of pleasure out of tweaking knobs and dials to tease exactly the sound I’m hearing in my mind’s ear out of a synthesizer. Some people learn how to do this right away, or just have a natural talent for it—kind of like riding a bicycle or playing soccer. But I have also met professional sound designers that struggle painfully with each new synthesizer. Over the years I have held several seminars on the subject, shown fellow musicians a trick or two now and then, and written sound programming columns for various trade magazines. As a result, it has become clear to me that sound programming can be taught—yet there has never been a book that does this successfully. From 2000 to 2003 I published a comprehensive sound-programming column in the German magazine “Keyboards,” with the prospect of a future book in the back of my mind. I had never received so many enthusiastic letters from readers as I did for this column. A delayed run of the col-

umn in the US magazine “Keyboard” brought a similar positive reaction with many of the same comments, most of which read something like this: “I have been fooling around for years and have read countless articles and books. Your SoundForum suddenly made everything clear.” Reason enough to publish the entire series as a book. To that end, the columns have been edited to serve the book form and condensed and/or augmented as necessary. The result is a comprehensive course designed to teach you how to program sounds on your synthesizer(s) from square one. Although the course references a particular software synthesizer, which was developed especially for this purpose, the techniques and methods described are applicable to any synthesizer and any type of synthesis. Think of it like a driver’s license: you may learn and take your test in a particular car, but the license is not restricted to that model. I wish you lots of fun and success with the book. Sincerely,

Peter Gorges

Contents 1 Hello Again . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Free Synthesizer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

15 15 16 17

2 The SoundForum Synthesizer . . . . . . . . . . . . . . . . . . . . . . . . . System Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . How Do I Play and Hear the SoundForum Synthesizer? . . . . . . Configuration—Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Audio Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Latency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MIDI I/O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . De-Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Configuration—Mac OS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Audio Output Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Latency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MIDI Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Using ASIO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Buffer Size and Audio Routing . . . . . . . . . . . . . . . . . . . . . . . Playing the SoundForum Synthesizer . . . . . . . . . . . . . . . . . . . . . MIDI Note Playback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Menus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Toolbar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The User Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

19 20 20 21 21 22 22 23 24 24 24 24 25 25 26 26 26 27 31 33

7

Contents

3 Just Do It . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ladies and Gentlemen, Start Your Synthesizers! . . . . . . . . . . . . . It Just Doesn’t Want to Work! . . . . . . . . . . . . . . . . . . . . . . . . . Go! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . How Does Simple Synthesis Work? . . . . . . . . . . . . . . . . . . . . . . . . From Alarm Signal to Pad . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Color Your Sound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Transistor Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Filter-free Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

39 40 40 40 41 43 44 45 46

4 Synthesis Tricks—For Experts Only . . . . . . . . . . . . . . . . . . . . . 49 Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Beating between Oscillators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Tonal Shades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 PWM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 The Right Combination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Frequency Modulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 The Nature of Frequency Modulation . . . . . . . . . . . . . . . . . . 57 What Does Simple FM Sound Like? . . . . . . . . . . . . . . . . . . . 58 Vibrato Is FM Is Vibrato … . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 … Or Trills . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Tonal Color . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Interlude . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Visual FM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Science Fair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Sync and Ring Modulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 What Does Sync Sound Like? . . . . . . . . . . . . . . . . . . . . . . . . . 66 How Is Sync Generated? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Sync Lead . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Metallic Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Heavy Beating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 Ring Modulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Basic Ring Modulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Bells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

8

Contents

Make It Dirty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

72

5 Conclusion of the Basic Course . . . . . . . . . . . . . . . . . . . . . . . . Our Curriculum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Laser Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synth Basses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synth Horns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synth Strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Weather and Echo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

75 75 75 77 78 80 82

6 SoundForum Synthesizer Tutorial . . . . . . . . . . . . . . . . . . . . . 83 Basic Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Loading and Storing Sounds and Banks . . . . . . . . . . . . . . . . 84 Storing Single Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Interactive Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Remote Control of the SoundForum Synthesizer . . . . . . . . 87 Maximum Number of Voices . . . . . . . . . . . . . . . . . . . . . . . . . 89 Unison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Monophonic Sounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Another Tip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Sound Generation Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Audio and Control Modules . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Audio Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Sound Generation and Sound Shaping . . . . . . . . . . . . . . . . 94 Control Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 A Brief Introduction to Control Voltage . . . . . . . . . . . . . . . . 96 Automating Sound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Activating Control Modules . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Brief Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Sound Parameters in Depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Example Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Oscillator 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Oscillator 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Mixer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Amp Env . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

9

Contents

Filter Env . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 Filt-Env→Osc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 LFO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 7 Advanced Sound Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . From the Beginning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensual Perception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Power of Precedents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Taste Is No Secret . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Brief History of Sound Synthesis . . . . . . . . . . . . . . . . . . . . Just One More Thing … . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Levels of Perception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Five Levels of Sound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Minimum Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sound Preferences Result from Conditioning and Habit . . . Music and Sound Preferences Are Intertwined . . . . . . . . . . Culture, Personality and Situation . . . . . . . . . . . . . . . . . . . . . Perception and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Levels of Sound Perception in Overview . . . . . . . . . . . . An Analysis of GM Instruments . . . . . . . . . . . . . . . . . . . . . . . Homework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Effects and Mix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Solution to the Homework Assignment . . . . . . . . . . . . . . . . Effects and the Sound Perception Levels . . . . . . . . . . . . . . . . EQ and the Sound Perception Levels . . . . . . . . . . . . . . . . . . . End of the Lecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

121 122 122 123 124 124 125 126 127 128 129 129 130 131 132 132 133 134 135 135 139 139 141 143 145

8 Advanced Hands-on Course . . . . . . . . . . . . . . . . . . . . . . . . . . . Sound Perception Levels and Synthesizer Modules . . . . . . . . . . . A Brief Tour of the Synthesizer Modules . . . . . . . . . . . . . . . . . . . . Oscillators and Mixer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Filter Envelope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

147 148 148 149 151 152

10

Contents

Amplitude (Volume) Envelope . . . . . . . . . . . . . . . . . . . . . . . . LFO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mission: Possible—Exercises in Goal-oriented Sound Design . . From Levels to Adjectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . Missions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mission 1: Soft Melody out of Basic Square . . . . . . . . . . . . . Mission 2: Juicy Bass out of Basic Saw . . . . . . . . . . . . . . . . . Mission 3: Opera Voice out of Basic Sine . . . . . . . . . . . . . . . Mission 4: Electronic Delay out of Basic Tri . . . . . . . . . . . . . Mission 5: High Winds out of Basic Noise . . . . . . . . . . . . . . Mission 6: Extra-wide Brass Pad out of Basic PWM . . . . . . Mission 7: Pitched Reverb out of Basic String . . . . . . . . . . . Insight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

153 154 155 156 157 157 159 160 161 162 163 166 168 169

9 Puzzle Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Completing Puzzles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tonal Color and Tonal Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Basic Tonal Colors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . What’s Missing in the Tonal Colors? . . . . . . . . . . . . . . . . . . Which Controls Do What? . . . . . . . . . . . . . . . . . . . . . . . . . . . Sub-bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Octave Square Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TB Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Black Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sync Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fat PWM Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E-Pianos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Plucked Strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bells/Mallets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . String Ensemble . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Brass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Voices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Choir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Breath . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

171 171 172 173 173 174 174 175 176 176 177 178 178 179 179 180 181 181 182 183

11

Contents

Envelopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Four-phase Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ADSR—with a Handicap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Piano . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electric Piano . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Harpsichord . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vibraphone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xylophone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Perc(ussive) Organ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pipe Organ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acoustic Guitar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electric Guitar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acoustic Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Slap Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synth Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strings/Ensemble/Choir . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Solo Brass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Brass Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synth Brass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . LFO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Who Is LFO? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Snapshots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PITCH—Vibrato . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PITCH—Laser Gun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PITCH—Trill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PITCH—Chorus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PITCH—Flanging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PITCH—Random . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PWM—Chorus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PWM—Ensemble . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PWM—Trance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FILT—AutoSweep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FILT—Repeat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

12

184 185 186 186 186 187 187 187 188 188 188 188 188 188 189 189 189 189 189 189 190 190 190 191 191 191 192 192 192 193 193 193 193 194 194

Contents

FILT—AutoReverse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194 AMP—Tremolo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194 AMP—AutoTrig . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 AMP—Echo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 Observing LFO Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . 195 10 Factory Bank Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Example Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Factory Bank Sounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1: Butterfinger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2: Baked Beans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3: Syncomator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4: Metallurgy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5: Ciaoup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6: PPG 2002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7: Logical Sound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8: Fjord Lord . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9: Machine Head . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10: OBY 16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11: FlexPad . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12: Chime On . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13: Dream Theam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14: AmbientraK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15: Chick Magnet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16: ShakAtak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17: Revell Yell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 to 20: Sound FX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Factory Bank Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

197 198 198 198 199 200 201 202 203 204 205 206 206 207 208 209 210 211 212 213 213 215

11 Box of Tricks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Beating between Oscillators Done Right . . . . . . . . . . . . . . . . . . . Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Special Case: Bass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pulse Width Modulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Which Type of Fat for Which Recipe? . . . . . . . . . . . . . . . . . . . . . .

219 219 219 220 221 221 222

13

Contents

Which Waveform Should I Start with? . . . . . . . . . . . . . . . . . . . . . Where is the Tonal Color Variety Hiding? . . . . . . . . . . . . . . . . . . . Filter Slope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Key Tracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Setting the Filter Range Correctly . . . . . . . . . . . . . . . . . . . . . . . . . Resonance as a Third Oscillator . . . . . . . . . . . . . . . . . . . . . . . . . . . What Are Band-pass and High-pass Good for? . . . . . . . . . . . . . . . Juicy or Creamy? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

224 224 225 226 226 227 228 229

Contents of the CD-ROM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Mac OS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233

14

11 Hello Again While developing the SoundForum concept, my goal was to create a complete solution for myself and for all participants. The only way to realize this was to have a synthesizer that all participants had access to. I intentionally use the word “participant” rather than “reader” for the simple reason that the SoundForum was conceived not as “dry reading,” but rather as a pro-active instructional guide for learning sound programming or expanding and deepening current knowledge.

Free Synthesizer During my research for the SoundForum, I tested numerous software synthesizers, hoping to find one that fulfilled my list of criteria. My “wish synth” had to: ❖ offer the necessary feature set yet be expandable, ❖ sound good, ❖ run on Windows and Mac OS, ❖ be adaptable to our needs and ❖ allow free distribution to SoundForum participants. Several dozen commercial, shareware and freeware synthesizers were omitted from the list just as quickly as they were installed. I finally chose Native Instruments. Reaktor developer Stephan Schmitt was immediately willing to develop a special software synthesizer for the SoundForum—a simple basic model without any real bells and whistles, but with all the necessary and sensible features. Not surprisingly, it was dubbed “SoundForum Synthesizer.”

15

1 Hello Again

For a short time now, the unofficial “successor” to the SoundForum Synthesizer, the “SoundSchool Analog,” has been available from Native Instruments. This synthesizer is based on the Reaktor 4 technology and runs on Windows XP as well as Mac OS X. In spite of numerous improvements and augmentations, the SoundSchool Analog can import snapshot (.ssf) files from the SoundForum Synthesizer. Thus the new synth can easily be used with this book. You can download the SoundSchool Analog free of charge from the Native Instruments website. Type the following URL into your browser’s address field: ✧ www.native-instruments.com/wizoosoundschool.info

Go to “Download,” select the appropriate version (Mac or PC), and log in using the following data: ✧ Username: [email protected] ✧ Password: soundschool

Register by filling out the required fields on the download page and start the download.

The Concept This book utilizes a hands-on approach to help you learn the basics of sound design or to broaden your knowledge on the subject. If you’re looking for academic acrobatics like “Filters: A Blessing or a Curse?” or “A Critical Discussion on the Harmonic Series of Basic Waveforms,” you’re in the wrong place.

16

The Experiments

The Experiments I’m sure you’re familiar with the principle of the electronic construction set: rather than the tedious study of how resistors, condensers or transistors work, you build a simple radio, sound generator or light barrier, and gather the necessary knowledge in the process. That’s exactly how this book is conceived, except that our “transistors” are oscillators, our “resistors” are filters and our “condensers” are envelopes. I have often said that a thorough knowledge of synthesizer theory is secondary for a good sound designer. I am a firm believer in the “experiment first, ask questions later” principle: tweak and find the sound, then figure out what you did. It’s a fast, effective way to learn, and you get results along the way. If you’re serious about learning sound programming or improving your skills, this book will help you get there quickly. This I promise you.

17

2 The SoundForum Synthesizer The SoundForum Synthesizer is on the enclosed CD. It is optimized to correlate perfectly with the book, and is programmed so that it will run even on older computers. The sound quality and functionality are comparable to that of a professional hardware synthesizer (analog or virtual-analog), making it the perfect tool for a comprehensive sound design course. SF-Synth

The SoundForum Synthesizer.

19

2 The SoundForum Synthesizer

System Requirements The SoundForum Synthesizer is based on Native Instruments’ “Reaktor” technology, and runs on Windows or Mac OS with the following requirements: ❖ Windows 95 or better, Pentium 233 MHz, 32 MB RAM, 16-Bit sound card, MIDI keyboard (recommended); ❖ Mac OS 8.5 or better, PPC 604e, 250 MHz, 32 MB RAM, MIDI keyboard (recommended).

Features The SoundForum Synthesizer is a high-quality, well-equipped polyphonic analog synthesizer in software form. The feature set is quite impressive: ❖ Two oscillators, each with four partially continuously adjustable wave forms with pulse width modulation. ❖ Ring modulator, FM, sync. ❖ Multimode filters with switchable slope and resonance. ❖ Filter envelope with modulation matrix. ❖ ADSR amplitude envelope. ❖ LFO with four wave forms including random. ❖ Oscilloscope. ❖ Up to 64 voices. ❖ Unison mode. As you can see, the SoundForum Synthesizer packs plenty of power for creating dazzling sounds—like the ones you’ll be programming after reading this book.

20

How Do I Play and Hear the SoundForum Synthesizer?

How Do I Play and Hear the SoundForum Synthesizer? You can play and hear the SoundForum Synthesizer using your computer’s normal MIDI and audio interfaces. If you don’t have a MIDI keyboard handy, the computer keyboard will also work.

Configuration—Windows The SoundForum Synthesizer requires a sound card to output its sounds. No special driver is necessary when using a standard sound card. The SoundForum Synthesizer utilizes standard MME and DirectX drivers, which are installed with the sound card. You will have to make several settings in the software to achieve optimal performance. Please select Sound Card under System > Audio Port, and open the settings dialog under System > Audio Settings. Note: The In Port cannot be used with the SoundForum Synthesizer and is not available for activation under Soundcard Properties. AudioSettings

Audio Settings dialog for standard sound cards.

21

2 The SoundForum Synthesizer

Audio Drivers This dialog allows selection of a particular sound card driver for the SoundForum Synthesizer when more than one sound card (or driver) is installed. The Out Port list shows options listed as MME: or DirectSound. Newer DirectSound drivers generally deliver audio output with far less delay (latency) than earlier MME drivers and are usually preferable. I recommend trying all available drivers and using the one with the least latency. Do not use emulated DirectSound drivers (usually listed as such). These are actually MME drivers that look like DirectSound drivers, and usually deliver the worst performance. In order to use DirectSound drivers, the Windows extension DirectX 5.0 (or better) must be installed. Note: The latency-reducing technology employed by the SoundForum Synthesizer places high demands on the sound card driver. Many drivers, particularly older ones, have trouble dealing with this, causing errors and system crashes. Please ensure that you have the latest driver version for your sound card. Latency The delay in audio output (latency or reaction time) depends on several factors, including the size of the audio buffer assigned to the sound card by the software. The minimum buffer size necessary for glitch-free operation will depend on what type of sound card and driver you use. When first installing the SoundForum Synthesizer, you may want to skip the settings described in the following paragraphs, so that you can familiarize yourself with the software. Once you have learned the basics of the synthesizer, you can return to these instructions to optimize the latency and performance.

22

Configuration—Windows

To determine the optimal buffer size for your system, move the Play ahead fader to the left while playing the SoundForum Synthesizer with your MIDI keyboard until glitches in the audio output appear. Then move the fader slowly to the right, increasing the buffer size, until the glitches disappear. You now have the optimal buffer size setting for your system. If Play ahead is set too low, an MME driver may cause incessant stuttering, whereas a DirectSound driver will usually “stumble” once, after which the effective latency is very long (approximately 1 second). Important: To ensure optimal performance, the Play Ahead setting should be manually optimized every time you change the sound card or install a new driver. By the way: The polyphony (number of voices) and sample rate used by the software have no influence on latency or timing. They do, however, influence the demands on the processor and thus the behavior of the entire system. The output level of the sound card is adjusted via the sound card’s mixer, which can be controlled by the Windows accessory Volume Control, the control panel’s Multimedia Properties, and/or by the mixer software included with the sound card. MIDI I/O The settings for the MIDI interface, with which the SoundForum Synthesizer communicates with the outside world, are adjusted in the MIDI Port dialog, accessible via MIDI Settings in the System menu. All MIDI ports installed under Windows are displayed here. If an in port is being used by another program, it will not be available to the SoundForum Synthesizer and will not be displayed. In this case, either de-assign the port in the other program or start the SoundForum Synthesizer first. Similarly, an in port intended for use by another program must be deleted from the Installed Inports list within the SoundForum Synthesizer before it can be used by another program.

23

2 The SoundForum Synthesizer MidiPort

The MIDI Port Dialog.

De-Installation The following procedure is recommended for complete removal of the SoundForum Synthesizer from your PC: 1 Select Start > Settings > Control Panel > Add/Remove Programs. 2 Select the SoundForum Synthesizer in the list of installed programs in the Install/Uninstall dialog. 3 Select Add/Remove and confirm removal with Yes.

Configuration—Mac OS Audio Output Settings The Sound Manager settings are accessible via the Sound control panel (or, in older Mac OS versions, the Sound tab in the Monitors and Sound control panel). Latency The delay between a MIDI event and the resulting audio signal (latency or reaction time) should be as short as possible (less than 20 ms).

24

Using ASIO

Important note: OMS (Open MIDI System) slows the MIDI data flow noticeably when virtual memory is activated. For best results, select the Memory control panel and set virtual memory to Off (you will need to restart the computer for this change to take effect). MIDI Input The SoundForum Synthesizer utilizes OMS (Open MIDI System) for receiving MIDI data. To select a MIDI input, select OMS Settings in the System > MIDI Settings menu.

Using ASIO ASIO (Audio Streaming Input Output) is a driver architecture for audio cards, which was developed by Steinberg. Available for Mac OS and Windows, ASIO features low latency and supports cards with multiple tracks. Most manufacturers supply an ASIO driver with the audio card, allowing it to be used with any ASIO-compatible software. Although harddisk recording has typically been the primary application, ASIO’s low latency has made it increasingly important for native software synthesizers. The SoundForum Synthesizer supports ASIO on both Mac OS and Windows platforms. Achieving minimal latency is primarily dependent upon two things: 1) the quality of the driver and 2) correct configuration. Mac OS: ASIO drivers must be located in a folder of the same name (“ASIO Drivers”), within the same folder as the SoundForum Synthesizer itself is located. Otherwise, they will not be found by the SoundForum Synthesizer and cannot be utilized by the program. To activate ASIO in the SoundForum Synthesizer, select ASIO in the System > Audio Port menu. If more than one ASIO driver is installed, the desired driver can be selected via System > Audio Settings.

25

2 The SoundForum Synthesizer

Which options are available in the Audio Settings window depends primarily on the card you are using; some cards can only be configured via their own control panel (which can be opened via the Open Control Panel button in the Audio Settings window). Buffer size and clock rate, however, can usually be adjusted in the SoundForum Synthesizer’s Audio Settings window. If your card can be synchronized to an external source but does not automatically adapt to its clock rate, you can usually choose a sync source in this window. Buffer Size and Audio Routing There is one notable peculiarity with ASIO regarding buffer size: ASIO card manufacturers generally specify a range, in which they are guaranteed to work properly. Most ASIO drivers are conceived with harddisk recording in mind, and not for real-time synthesis. Thus, this range is usually unnecessarily high in order to avoid pops even in complex systems. The SoundForum Synthesizer allows a far lower setting than the manufacturers generally specify. With poorly written drivers, this can lead to program crashes. For this reason, the default SoundForum Synthesizer’s buffer size will always correspond with the card’s specifications. Experiment to determine whether lower buffer settings are possible. The SoundForum Synthesizer’s two audio outputs can be routed to the audio card’s ASIO channels via the System > Audio Routing menu.

Playing the SoundForum Synthesizer Once installed and configured, the SoundForum Synthesizer is ready to play. Start the application by selecting the appropriate entry in the Start menu (Windows) or by double-clicking the program’s icon (Mac OS). MIDI Note Playback Assuming you have a MIDI keyboard connected to the computer and the appropriate MIDI Inport activated (MIDI Settings), you should be able to play the synthesizer via the MIDI keyboard.

26

Playing the SoundForum Synthesizer

The virtual controls are easily operated using the mouse. To adjust a value assigned to a knob, click on the knob and hold the button down while you move the mouse up (forward) to increase and down (backward) to decrease the value. Switches are simply activated and deactivated via a single click. Menus The program controls and settings are located in various context menus and in the menu bar of the main window. The main functions and parameters found in the menus are described below. The File Menu ❖ Open: Allows you to load a previously saved SoundForum Synthesizer instance (suffix “.ens”). Native Instruments’ Reaktor series instruments cannot be loaded. ❖ Save: Saves the currently active SoundForum Synthesizer with all settings (snapshots, instrument info, MIDI settings etc.) as an instance (.ens) file. ❖ Save As: Also saves the currently active SoundForum Synthesizer with all settings as an instance file, giving you the option to rename the file. ❖ Load Snapshots: Snapshots can be saved and loaded independently of instances using the Snapshot File Format (suffix “.ssf”). To load a snapshot, select File > Load Snapshots. A dialog will appear, allowing you to define how the snapshot will be imported. Three options are available: Append: The new snapshot is loaded into the position after the last current snapshot. Insert: Allows you to determine the exact position into which the new snapshot will be loaded. All other current snapshots are moved back one step. Overwrite: Allows you to determine the exact position into which the new snapshot will be loaded. The current snapshot in that position is replaced by the loaded snapshot.

27

2 The SoundForum Synthesizer LoadSnapshots

The Load Snapshots Dialog.

❖ Save Snapshots: Allows you to save snapshots as “.ssf” files. A dialog appears, allowing you to define which snapshots will be saved via First and Last parameters. To save a single snapshot, enter the same number for First and Last. Use the Select All button to save all current snapshots. SaveSnapshots

The Save Snapshots Dialog.

❖ List of recently loaded files: Allows you to load one of the eight most recently loaded SoundForum Synthesizer instances via a single mouse click. ❖ Quit: Closes the program. If any changes have been made to the current instance since the last save, you will be asked if you want to save these before quitting.

28

Playing the SoundForum Synthesizer

The System Menu The items in the System menu serve primarily to control the MIDI and audio inputs/outputs, audio parameters like sample rate, and to monitor the CPU load of the SoundForum Synthesizer. ❖ Run/Stop Audio: Allows you to start (Run Audio) or stop (Stop Audio) the audio processing function of the SoundForum Synthesizer—sort of an on/off or mute switch. The same function is accessible via the “On/ Off switch” button in the toolbar. ❖ Sample Rate: Allows you to set the internal sample rate for audio signals. Higher sample rates result in higher audio quality; but will also noticeably increase the CPU load. If a different internal sample rate is selected than that used by the sound card, the rate will be converted at the audio inputs and outputs. ❖ Audio Port: Allows you to select an audio output for the SoundForum Synthesizer, as well as an audio input to the synthesizer for processing. Installed (available) cards are listed in black; non-installed (unavailable) options are in gray. ❖ Audio Settings: Opens a dialog window in which the audio performance of your sound card can be optimized. For more details, please see the appropriate “Configuration” text on page 21 (Windows) or page 24 (Mac OS). ❖ Audio Routing: Please refer to “Using ASIO” on page 25. ❖ MIDI Settings: Allows you to assign MIDI inputs and outputs to the SoundForum Synthesizer. To assign one or more MIDI inputs, select the desired input in the Available Inports list and copy it to the Installed Inports list using the Insert button. The Delete button removes a selected input from the Installed Inports list. Click OK to confirm your changes and exit the dialog. If you select a MIDI interface driver as an Inport, you should be able to control the SoundForum Synthesizer from an external MIDI device (e.g. master keyboard) using the corresponding MIDI input.

29

2 The SoundForum Synthesizer

The Instrument Menu ❖ This menu contains the same functions as the toolbar. For details, please refer to the Toolbar section below. The “?” Menu (Windows only) ❖ Help: Opens the HTML help document in your default Internet browser. ❖ About: Opens the info window for the SoundForum Synthesizer. The software version is displayed in the lower section. The two buttons Soundforum on the Web and Visit the NI* Website open your default Internet browser and navigate directly to the corresponding Website. * Native Instruments The View Menu (Mac OS only) ❖ Show/Hide Toolbar: Turns the toolbar display on and off. ❖ Show/Hide Hints: Turns the “Help Balloons” on and off. When activated, pointing the mouse cursor at one of the SoundForum Synthesizer’s controls or panels will open a balloon with a brief explanation of that control or panel. ❖ Show Synth: Brings the synth window to the front (in case it has been “buried” by windows of other programs). The Help Menu (Mac OS only) ❖ SoundForum Synth Help …: Opens the HTML help document in your default Internet browser. Any other items in the Help menu are Mac OS items, dependent upon which OS version you are running.

30

Playing the SoundForum Synthesizer

The Toolbar Toolbar

The SoundForum Synthesizer Toolbar.

The Toolbar gives you easy one-click button access to many functions of the SoundForum Synthesizer. These are described below, from left to right as they appear in the toolbar: The “On/Off” switch on the left activates/deactivates audio processing, just like the Run/Stop Audio item in the System menu. This can be useful to reduce CPU load when no sounds are being generated by the synth. All audio processing routines are reset each time audio processing is stopped and started again. The CPU load display to the right of the switch shows the CPU time used by audio processing in percent; CPU “overload” is displayed as Over. The maximum range for glitch-free operation is generally between 60% and 80%—in any case far below 100%. Why? Well, other processes (i.e. audio data transfer to sound card, MIDI processing, graphic display), other applications that may be running simultaneously with the SoundForum Synthesizer, and of course the operating system all need CPU power, too. To determine exactly where the limit is for your computer, simply increase the number of voices until an overload warning appears (Processor Overload), and then reduce the number by one or two voices. Note: The maximum number of voices depends on your CPU and increases proportionally with the CPU’s clock rate. The Out display shows the level at the audio output. If no driver is assigned to the audio output, the display remains grey. The Show Hints button (arrow/question mark) activates/deactivates the synth’s proprietary balloon help. The Info button opens a window containing a brief description of the SoundForum Synthesizer, including the version number.

31

2 The SoundForum Synthesizer

The Properties button opens a window in which various parameters of the current synth are displayed and can be edited. For details, please see the “Properties” section on page 34. The MIDI Learn button allows you to easily assign a MIDI controller to a parameter of the SoundForum Synthesizer. To do so, simply select the desired parameter by clicking on its control element (the parameter name turns red), activate MIDI Learn, and operate the desired control element (e.g. modulation wheel) on your external MIDI controller. To cancel the assignment, double-click on the control element and deactivate Remote in the Properties window. The MIDI Activity “LED” illuminates whenever a MIDI event is received. The field to its right allows selection of the MIDI receive channel, which can also be defined in the Properties window (the MIDI send channel can only be changed in the Properties window). The next field allows you to select a snapshot from the list of loaded snapshots. The Store Snapshot button (camera symbol) to its right accesses the snapshot file window (for details, see “The File Menu” on page 27). The Compare button allows you to compare two versions of sound edits or parameter settings: the current edit and the edit stored in the Compare buffer. When you click on the Compare button, the current edit is stored in the Compare buffer, while the previous status is restored—audibly and graphically. You can now use either edit as a basis for further tweaking. As soon as you begin to edit a parameter again, the current status is first stored in the Compare buffer. Thus you are always comparing your current status with your last edit, rather than with the original (stored) snapshot, so that you can sound-design your way to your desired result step-by-step. Once you have clicked the Compare button, you must always choose between one of the two edits, which is then stored in the Compare buffer, as basis for further editing. When you select a new snapshot, the current edit is first stored in the Compare buffer. This allows you to compare your edit with any snapshot. If you accidentally select a new snapshot during editing, there’s no need to panic: Simply click the Compare button to recall the previous status.

32

Playing the SoundForum Synthesizer

The VCS field allows you to define the number of voices (also adjustable in the Properties window). The UNISON field allows you to define the maximum number of unison voices per note. The unison effect is enabled by entering a value greater than 1. Detuning between unison voices is set in the Properties window (Unison-Spread), as is the minimum number of unison voices per note (Min Unison Voices). The User Interface The SoundForum Synthesizer’s user interface adheres by and large to operating system conventions, so that the learning curve is minimal assuming you have some experience with Mac OS or Windows. Nonetheless, I would like to point out a few particularities and explain a few features that may be new to you. Mouse Techniques In general, when I refer to the mouse button, I mean the left mouse button (Windows). Virtually every function of the application can be carried out via the mouse. Double-clicking with the mouse button on an object (including fields in windows) has various effects, depending upon the object.

33

2 The SoundForum Synthesizer

Properties Properties

The Properties Window.

The SoundForum Synthesizer features polyphonic voice allocation. The Number of Voices field allows you to determine the maximum number of voices that can be generated simultaneously. When activated (box checked), the Automatic Voice Reduction option automatically reduces the polyphony when the CPU load exceeds a pre-defined limit. This is an effective way for the program to automatically adapt to the available processing power. Unison mode is activated when a value greater than 1 is entered in the Max Unison Voices field. Several voices are then assigned to each note, and each voice can be slightly detuned, resulting in a fat, modulating sound. The value entered in the Max Unison Voices field determines the number of voices assigned to a newly played note (assuming this number of voices is available). Conversely, the Min Unison Voices field determines the minimum number of voices assigned to a newly played note regardless of availability. This value can not exceed Max Unison Voices.

34

Playing the SoundForum Synthesizer

Unison Spread determines the amount of detuning between unison voices in hundredths of semitones or cents. A typical setting is “0.05” (5 cents), which means that each voice is detuned from the next by 5 cents (“1.00” = 1 semitone). The Snapshot Recall by MIDI option allows you to load snapshots via MIDI program change commands. When activated (box checked), a program change command sent by a connected MIDI controller (e.g. master keyboard) will cause the SoundForum Synthesizer to load the corresponding snapshot (assuming a snapshot with the corresponding number is available in the current instrument). The Receive Channel and Send Channel fields determine the MIDI channels for the SoundForum Synthesizer’s MIDI input and output, respectively. In other words, the synthesizer receives MIDI data sent over the channel entered in the Receive Channel field and sends MIDI data over the channel entered in the Send Channel field. Essentially any external MIDI device or software sequencer can be at the other end. When the All Incremental option is activated, all control elements in the current instrument are set to incremental mode for remote control. This is necessary in order to control the SoundForum Synthesizer with Native Instruments’ 4Control MIDI Controller. If MIDI Learn is activated, incremental mode is automatically set as needed. The Soft Takeover function, which can be activated/deactivated in the Properties window, helps to prevent glitches when using external controllers. The Panel to MIDI Out option, when activated, causes all movements of the synthesizer’s control elements, as well as snapshot changes, to be transmitted as MIDI control data via the MIDI out. This allows you, for example, to “record” these changes via a sequencer for “automated” playback. Similarly, the Remote to MIDI Out option causes all movements of the synthesizer’s control elements performed via remote control to be transmitted as MIDI control data via the MIDI out.

35

2 The SoundForum Synthesizer

The Upper Note Limit and Lower Note Limit fields enable you to enter MIDI note numbers to define a range of notes that will be processed by the instrument. Any MIDI note information received that falls outside of this range will be ignored. Note Shift allows you to transpose the instrument in half-steps against received MIDI note data. If, for example, you want the SoundForum Synthesizer to play a part one octave lower, enter “−12” in this field. Sustain Control enables you to define the MIDI controller number to be interpreted as sustain pedal data. (The MIDI standard defines controller number 64 as sustain.) When sustain is on, e.g. the value of the assigned controller is greater than 64, a played note will continue to sustain after the key is released. Hold Control allows selection of a MIDI controller number for hold pedal (“sostenuto”) data. (The MIDI standard defines controller number 66 as hold.) Any notes sounding while hold is switched on (controller value greater than 64) continue to sound even after the key is released. Hold does not effect notes played after it is switched on. Snapshots Snapshots are the SoundForum Synthesizer’s sound programs, just like the “programs” or “patches” of other programmable synthesizers. A snapshot stores the current positions of the instrument’s control elements. Loading a snapshot sets all control elements to the stored status. The SoundForum Synthesizer allows you to store up to 128 snapshots. Loading snapshots is easily done via the pull-down menu in the toolbar next to the Store Snapshot button (camera symbol). Simply open the list by clicking on the arrows, move the cursor to the desired snapshot (it highlights) and click. Snapshots can also be loaded via MIDI program change commands. To enable this, activate the Snapshot Recall by MIDI option in the Properties window. The snapshot number displayed before the name is identical to the corresponding MIDI program change number.

36

Playing the SoundForum Synthesizer

To store a snapshot, click on the Store Snapshot button (camera symbol). This opens the snapshot dialog: The No field allows you to enter a snapshot (and MIDI program change) number. This field will default at the next available number, e.g. 21 if 20 snapshots are currently stored. Label allows you to enter a name for your snapshot. Clicking on the Store button saves the snapshot and adds it to the instrument’s pull-down list. The top half of the Snapshot window gives you the option to overwrite or delete individual snapshots. To this end, a Select pull-down list is provided, as well as an Overwrite button and a Delete button. The overwrite function stores your current settings in the place of the snapshot selected in the Select list. The new snapshot retains the number and name of the overwritten snapshot. SnapshotsRename

The Store Snapshot Window: Store, Delete, Overwrite, Copy and Rename Snapshots.

To delete a snapshot, simply select it in the Select list and click Delete. To copy or rename a snapshot, first load it via the pull-down menu in the toolbar. Then click on the Store Snapshot button (camera symbol) to open the snapshot dialog. To rename the snapshot, enter its current number in the No field, then enter a new name in the Label field and click Store.

37

2 The SoundForum Synthesizer

To copy the snapshot, enter a new (unoccupied) number in the No field and, if desired, a new name in the Label field, then click Store.

38

3 Just Do It In this section you will complete an initial “fitness course” for sound designers. On this, I have good news and bad news. First, some good news: You will do things that you never believed you could. Now some bad news: It’s entirely possible that you won’t understand a single thing about what you’re doing. And, finally, some more good news: You will learn all of this and more in the course of this book. I’d like to try to kill two birds with one stone: ❖ The more experienced among us will get a chance to get to know the SoundForum Synthesizer, while ❖ the beginners among us will receive some basic explanations. Seeing as how the strong should help the weak, not only in fairy tales, we experienced folks will gladly accept even the simplest explanations. We will avoid theory at all costs. To use the analogy of driving school again, theory does not help us learn to combine the pure enjoyment of driving with technical finesse. It’s no different here.

39

3 Just Do It

Ladies and Gentlemen, Start Your Synthesizers! Before we can begin this tour de force of finesse and enjoyment, you need the SoundForum Synthesizer. If you are working through this book in a traditional fashion, e.g. from front to back, you have probably already installed and configured the synth. If not, you will find it on the supplied CD-ROM in the SoundForum Synthesizer folder. The installation program leads you through the simple installation process. Regarding configuration, please refer to the appropriate “Configuration” section (Windows on page 21 or Mac OS on page 24). It Just Doesn’t Want to Work! Should you—for whatever reason—experience problems with the SoundForum Synthesizer, feel free to contact Wizoo support directly at [email protected].

Go! You will find the snapshot menu (Save Snapshot; camera symbol) in the toolbar. As the name implies, snapshots are, well, snapshots of the synthesizer, e.g. its current parameter settings. Using snapshots could hardly be easier: ❖ Store a snapshot: make all your desired edits, click on the camera symbol, enter a position and a name, click Store, voilà! ❖ Load a snapshot: select a snapshot from the pull-down list (to the left of the Save Snapshot button). As you can see, I have already programmed 20 simple snapshots, which serve as “base stations” on our little fitness course.

40

How Does Simple Synthesis Work? snapshots.bmp]

The Load Snapshots Menu.

How Does Simple Synthesis Work? I sense a hint of higher education, of physics and history, in the room. Countless dissertations have been written on this subject, most of them in barely understandable scientific jargon that never seems to end. All the while, simple synthesis—also known as “subtractive synthesis”—is just a wee bit more complicated than tying shoes. In its simplest form, it works like this: The oscillator—nothing more than something that swings or varies back and forth—generates a raw, unrefined waveform at a pitch corresponding to the key that is pressed. Press the key, the tone appears; release the key, the tone disappears. ❖ Load snapshot 3, “Basic Saw,” and try it yourself! In this case, the oscilloscope displays a sawtooth waveform. This technicalsounding term really only refers to the shape of the waveform and has no deeper meaning. Further down the signal path, the vibration from the oscillator is routed to a filter; after all, not all sounds in this world should remind us of a saw. ❖ Turn the filter’s cutoff knob fully counter-clockwise. You have just experienced and thus now understand the basic principle of the filter. We not only want to vary the color of the sound, but also the volume. We may want it to fade in, like a soft pad, or fade out slowly like a piano.

41

3 Just Do It

And here comes the only real hurdle to understanding: contrary to the oscillator and the filter, the third component in the signal path—the amplifier—is not directly visible. So it is with the SoundForum Synthesizer. The amplifier’s controller, however, is very visible: the envelope, another terribly technical-sounding term. No cause for panic, though: the envelope is nothing more than a time-based process with adjustable length and shape; a curve defining the development of the component or sound over time. ◆

Adjust the “Amp Env” section to the following values: A: 32, D: 12, S: 0.4, R: 52.

The envelope display shows exactly how the sound’s amplitude or volume develops over time (from left to right). In a simple synthesizer, the signal flows through three components: oscillator, filter and amplifier. The other components simply serve to adjust, control and automate these three basic components, allowing them to create more variations and more complex sounds. One of these control components is the LFO, which stands for Low Frequency Oscillator. The LFO generates very slow vibrations that are not directly introduced into the signal, but rather serve to automatically modulate or vary the pitch, amplitude (volume) or color (e.g. Cutoff; you have already experienced this control) of the sound. ◆

Adjust the “Amount” knob in the LFO section to a higher setting and play around with the “Rate” knob.

Self-explanatory, right? OK, one could add that the yellow button “Osc P 1” assigns the LFO to the pitch of Oscillator 1. I don’t think it’s necessary to explain how that sounds. If you understand the process so far, you have essentially learned what simple synthesis is all about. The rest is details and subtleties.

42

From Alarm Signal to Pad

From Alarm Signal to Pad I know, I know: it’s easy for me to say how simple this all is; but I want to make it just as easy for you. 1 Load Snapshot 2, “Basic Square.” Sounds like a tilted pinball machine, huh? A pad, on the other hand, is supposed to be warm, fat, complex and soft. We’re only a few knob turns and button pushes away. 2 Adjust “Symm” in the Oscillator 1 section to 0.45. The knob is roughly at 12:00. Now we’re going to automate it with the LFO. 3 Activate Sym for both oscillators in the LFO section (“Sym 1” and “Sym 2” buttons should be yellow) and deactivate Pitch (“P 1” and “P 2” buttons should be gray). Don’t hear any difference yet … 4 Adjust the LFO “Amount” to 0.6. Voilà—the sound begins to fly! 5 Go to the mixer section and set “Osc 2” to 0.7, then head to the Oscillator 2 section and set “Puls-Sym” to 0.55 and “Detune” to 0.09. We just increased the “phat factor” by about 50%. 6 Now hit the filter section and set “Cutoff” to 85 and “Resonance” to 0.2. Now we’re talking warm. 7 Adjust “A” (Attack) and “R” (Release) in the “Amp Env” section to 50. That’s it: from pinball machine to first-class pad in six easy steps. Now you see that it doesn’t really take much to get to where you want to go. In this respect, simple synthesis is very different from more complex types of synthesis like frequency modulation (FM) or additive synthesis. What all types of synthesis do have in common, though, is that you first have to understand which steps will lead you to your goal.

43

3 Just Do It

By the way: you won’t find the pad we just created anywhere “ready-to-go.” What kind of crash course would this be if all you had to do was load a preset ? You could quickly store your new pad as a snapshot, though: 8 Click on the camera symbol, type “21” in the “No” field and “My First Pad” (or whatever) in the “Name” field, click on “Store”—done.

Color Your Sound Snapshots 2 to 8 are very simple basic sounds, unfiltered and sporting “organ envelopes:” on when you press a key and off when you let go. Try them out briefly, one after another. As you can see, they are all generated by Oscillator 1, with the exception of “Basic Detune,” which demonstrates the modulation created when you combine two oscillators. Creating polished sounds out of these basic patches is as easy as following the steps listed in the method table below, which gives you everything you need to turn a pig’s ear into a silk purse, so to speak. #

What

Where

1

Vary the basic sound

Oscillator 1 Mixer

2

Add oscillator modulation

Oscillator 2 Mixer LFO Filt-Env→Osc

3

Program the filter characteristics

Filter

4

Program the filter curve

Filter Env

5

Program the amplitude curve

Amp Env

The method table.

44

Transistor Bass

In practice, of course, things don’t always follow this method. Nonetheless, it does give you a very good basic framework, which you will adapt to your own methodic as time goes on. In the following experiment, as in several subsequent ones, we will stick to this method table while creating a usable sound out of a raw snapshot.

Transistor Bass The vision behind the sound we will now program is the legendary bass/ synthesizer/sequencer combination TB-303—a more basic synthesizer can now only be found in musical greeting cards. Bild TB03;fehlt nochrs

The legendary Roland TB-303.

1 Step 1: Load snapshot 2, “Basic Square” or 3, “Basic Saw.” Set “Interval” in Oscillator 1 to −24—after all, it is a bass sound we’re going after. 2 We can skip step 2, since we only need one oscillator. 3 Step 3: Set the filter to “LP2” (a less steep slope), then set “Cutoff” to 80 and “Resonance” to 0.94.

45

3 Just Do It

4 Step 4: Now we’re going to make the filter really crack. Set the filter envelope (Filter Env) as follows: A: 17, D: 38, S: 0, R: 39. 5 Last but not least, set the “R” (Release) in the amplitude envelope section (Amp Env) to 40. Pops pretty good already, doesn’t it? A little distortion—maybe not quite what the doctor ordered, but very popular with techno, electro, acid and co.—is also no problem for the SoundForum Synthesizer: 6 Set the “Level” knob in the Master section to 0. 7 Like to see the cute little animation in the waveform caused by the filter ? Then crank the Decay (“D”) knob in the Filter Env(elope) section all the way up, hold a note and watch the oscilloscope. Here’s a little pearl of wisdom you can take with you: the less overtones the filter lets through, the rounder the resulting waveform. While the philosophers among you brood about what this statement has to do with sawtooth and sine, the rest of you can follow me:

Filter-free Filtering One of the special features of the SoundForum Synthesizer, which you will be hard pressed to find in your hardware park, is the possibility to create a virtually limitless universe of waveforms by modifying the wave symmetry. To illustrate this, we will now take a first look at—and listen to—this capability: 1 Load snapshot 4 “Basic Sine.” A sine wave is initially a waveform without overtones; the aural experience should be self-explanatory. 2 Hold a note and slowly turn up the “Symm” knob while observing the oscilloscope and listening to the way the sound changes.

46

Filter-free Filtering

The more angular the waveform, the more overtones the sound contains, until a rounded sawtooth appears. As the corners get sharper, so does the sound. The “Symm” knob can be automated via the filter envelope; this is the purpose of the somewhat cryptically labeled “Filter-Env→Osc” section. 3 Set the Symm knob to 0.4. 4 The Filter Envelope also gets a pre-tweak: A 30, D 50, S 0.5—the release value (R) is not important here. Now for the automation: 5 Adjust the Amount value in the “Filt-Env→Osc” section to 1 and switch the upper “Sym” button on. Sounds a little like the trumpet in Sesame Street, right? If you’d like a chance to visually experience the automation, set D (decay) in the Filter Envelope to 90, adjust the oscilloscope and hold a note.

47

4 Synthesis Tricks—For Experts Only After warming up with the last chapter, we now advance to synthesis tricks that only experts can comprehend. Experts like us … The cryptic, fear-instilling terms that we will encounter in the course of the next pages turn out to be entirely harmless. I promise. When we’ve finished the first section of this chapter, terms like LFO, PWM and filter envelope will be flowing out of your mouth with the greatest of ease. You will be canonized at the latest after the following two sections, which contain even nastier terms …

Snapshots In this chapter we’re going to undertake a couple of experiments—just like in the good old school lab. There is, however, one difference to the good old school days: Here, we’re not trying to explode our chemistry teacher, but rather to learn (and have some fun doing it). Ready-made experiment sets are available in the form of snapshots. We will also make some sensible use of the oscilloscope, something the poor thing rarely experiences, especially in synthesizer courses.

49

4 Synthesis Tricks—For Experts Only

Beating between Oscillators Among other things, the term “beating” refers to the effect of mixing two oscillators that are slightly out of tune with one another. The sound gets fatter and livelier. Finding the right amount of beating is just as important in sound design as swallowing is in eating. So here we go: 1 Load snapshot 7, “Basic Detune,” from the pull-down menu in the toolbar. 2 Select the “Puls” waveform for both oscillators and adjust both “Sym(m)” knobs to 0. 3 Set the “Mixer” for Oscillator 2 to 0. If you now play a note in the medium range, you hear a square wave, which does a wonderful job as a visual study object. If you adjust the oscilloscope properly, you will see the wave: ◆

Adjust the “Time” knob until you can only see one or two waves in the display. Now set “Level” so that the height and width of the waves are roughly the same.

Now let’s see what happens when we add a second vibration: 4 Slowly turn up the “Mixer” level for Oscillator 2 and watch the oscilloscope. You see that the waveform becomes animated, as if a second one was moving on top of the first. Simultaneously you can hear that the sound is no longer lifeless, but has also become animated. 5 Turn the “Detune” knob in oscillator 2 slowly clockwise to around 0.36. The animation accelerates—both visually and aurally. This is called “beating,” and the oscilloscope shows very clearly what you hear. Since oscillator 2 is always tuned a bit higher (= faster) than oscillator 1, both oscillators meet occasionally at maximum level; then oscillator 2 “passes” oscillator 1 again, then they meet at minimum level and so on. The more out of tune, the faster oscillator 2 “passes” oscillator 1.

50

Beating between Oscillators

Tonal Shades We’ll let that be for the moment and take a look at another effect caused by this addition of vibrations: 1 Set Detune in oscillator 2 to exactly 0. 2 Hold a note and turn up the Mixer for Oscillator 2 to 1. Now, proceed with feeling: 3 Turn the Detune knob slightly clockwise or counter-clockwise, until the two vibrations are added to each other—you will see a very big waveform. 4 Now set oscillator 2’s Interval first to 12, then to 24 and to 36. Observe the oscilloscope: The waveform grows “teeth;” the higher the interval, the more teeth. This is easy to explain: An interval of 12 half-steps (24, 36) means that oscillator 2 vibrates twice (or four times, or eight times) as fast as oscillator 1. That’s exactly what you see. 24square

The “teeth” in the square wave come from adding an oscillator two octaves higher.

❖ Turn oscillator 2 off completely (“Osc 2” knob in the Mixer) and then back to 0.5, so that you can see the difference between the two waveforms. ❖ Experiment with the “Detune” knob in the Oscillator 2 section a bit; turn it up to different levels and finally back to 0.

51

4 Synthesis Tricks—For Experts Only

Now you can understand why beating can exist between oscillators that are actually an octave apart—the couple hundredths of half-tones (cents) shouldn’t really make a difference … Just for fun, you might try setting oscillator 2’s waveform to “Saw” or “Sine.” Play around with the mixer level, interval setting and waveform. This will probably help you understand a lot that you didn’t understand before and that can’t really be explained with words alone. Detuning is, however, only one way to create beating. Not all synthesizers offer two or more oscillators. One possible alternative is PWM. PWM Load snapshot 8, “Basic PWM.” Play a note in a medium range and observe the oscilloscope. You’ll see that the width of the pulse (the upper edge of the waveform) varies compared to the pause of the pulse (lower edge). Pwmscheme

The oscilloscope shows the variation in pulse width with PWM.

This is called Pulse Width Modulation, or PWM for short. The effect is similar to beating between oscillators but sounds more dramatic. How is it created?

52

Beating between Oscillators Pwmlfo

The PWM-relevant parameters in the LFO section.

1 De-activate the yellow “Sym 1” button in the LFO section: The effect disappears. 2 Grab the “Symm” knob in oscillator 1 and turn it up and down. Hear it? Have the LFO do this for you automatically: Re-activate the “Sym 1” button in the LFO section. 3 Now vary the LFO rate. You’ll discover that there is a point where the modulation even sounds detuned, roughly at a “Rate” of 12 and an “Amount” of 0.8. If you lower the Amount value, the detuning disappears. You’ll also find out that the detuning seems stronger when you play lower notes, without changing the settings. This is why savvy sound designers will set the LFO rate to follow the keyboard whenever possible. We’ll make a mental note of this and continue. 4 Set the LFO rate to 0 and select various LFO waveforms, one after another. Note that the LFO waveform has no direct effect on the waveform of the audio signal. The LFO waveform merely determines the shape of its “movement” (think of the manual knob movement you did in the last experiment). Once you’ve experienced this in real-time it should be easy to understand. The LFO is not the only possible control element for PWM. (See? We’re already tossing these terms around as if we grew up with them.)

53

4 Synthesis Tricks—For Experts Only

Load snapshot 8 again. You will now generate PWM via the filter envelope instead of the LFO: 1 First things first: Set “Symm” in oscillator 1 to 0. 2 Adjust the filter envelope (“Filter Env”) as follows: A: 43, D: 54, S: 0. We don’t need to worry about R. 3 Switch the yellow “Sym 1” button in the LFO section off. 4 Turn the “Amount” knob in the Filt-Env→Osc section all the way up and activate the “Sym 1” button. Now the pulse width is controlled by the filter envelope. When the envelope level is at 0 (at the beginning and at the end), you hear a square wave. When it is at maximum level, you hear a narrow pulse—sort of a nasal sound. To hear this more slowly and more clearly, increase the attack (“A”) and decay (“D”) times in the filter envelope (“Filt Env”) and observe the oscilloscope. Play around with the envelope settings to get a feeling for the envelope itself and the effect of the resulting pulse-width envelope. We’ve learned enough today. For good measure, though, let’s combine the two techniques to create a truly universal sound—usable for almost everything from classical to techno—the good old Wurlitzer electric piano. Gaining fame through hits like “Logical Song” by Supertramp, “I’m Not In Love” by 10 CC and “A Girl Like You” by Foreigner, this sound is most often heard today in the instrumental framework for divas such as Shania Twain or Anastacia. We’re going to program a production-ripe hard-rock variation complete with a super-phat chorus. The Right Combination OK, grab those knobs! Step 1: Basic Sound The tonal character of the Wurlitzer piano is closely related to the square wave, so we’ll start with the square snapshot. 1 Load snapshot 2, “Basic Square.”

54

Beating between Oscillators

2 Set the “Symm” knob in Oscillator 1 to 0.02; this adds a little dirt to the square wave character. 3 We might as well go ahead and do the same with the “Puls-Sym” knob in Oscillator 2, although this oscillator isn’t in the game yet. Step 2: Beating, Oscillator Modulation First the beating: 1 Set Osc 2 in the Mixer section to 0.9 and lower Osc 1 to 0.9. 2 Increase “Detune” in Oscillator 2 to 0.04. Now we’re going to emulate the metallic tone that is generated by the metal tines being hit; this tone fades as the note rings out. 3 Activate the “Sym 1” and Sym 2” buttons in the Filt-Env→Osc section (both should be yellow) and set the “Amount” knob to 0.24. You can hear the way the sound develops. The change is still too fast, but we won’t let that bother us right now. Step 3: Filter Characteristic Since the sound is way to full of overtones for a harmless little electric piano, we’re going to filter it. The envelope will open the filter again, so we have to go for the darkest sound possible. 1 Set “Cutoff” to 83 and “Reson” (resonance) to 0.15 in the Filter section. We’re using resonance here—as always, actually—to leverage the filter a bit, making its effect more prominent. Step 4: Filter Progression Now we’re going to program the filter’s progression, or movement, which will—due to the settings we made in step 2—effect the basic sound. 1 First of all, we have to allow the filter envelope to affect the filter. This is done via the “Env” knob in the Filter section, which we will now set to 20. As you have probably noticed, this parameter is also controlled by the dynamics of your playing.

55

4 Synthesis Tricks—For Experts Only pianenv

Filter and volume envelopes for an electric piano sound.

2 Setting “A” to 0 in the “Filter Env” section ensures that the piano sounds like it is hit, rather than blown or bowed. 3 Next, set “S” to 0. “S” stands for sustain and represents the level at which the envelope remains as long as a note is held. Since nothing stays put with a piano, we set this parameter to 0. 4 A crucial parameter here is the decay time (“D”); set this to 58, and just like magic the sound darkens with a smooth curve. Step 5: Volume Progression It already sounds pretty much like electric piano, but the volume progression more resembles an organ. Time to adjust the amplitude (volume) envelope: 1 Set this envelope exactly like the filter envelope. To prevent the sound from ending quite so abruptly, set the release time (“R”) in both envelopes a little higher, to 23. That’s it. Sure, you could do a lot of fine-tuning, a tweak here, a polish there. But that would be defeating our purpose—we’re not out to win a beauty contest. The pros among us may be thinking that a seasoned sound designer would have to go a different route; much more mathematical. To that I can only say, “Congratulations! Have fun in the bragger’s zone! (And don’t forget to write.)”

56

Frequency Modulation

Frequency Modulation In this and the following section, we will gaze beyond the horizon of simple analog synthesis to marvel at some of the more exotic qualities of many analog synthesizers. The subject is frequency modulation, ring modulation and sync (or oscillator synchronization, to be more precise). Terms, once again, that sound so boring you would never suspect that they represent marvelous things. Things that enormously expand the sound spectrum of the SoundForum Synthesizer. This section is dedicated to frequency modulation, while ring modulation and sync are discussed together in the next section. It may be worth noting that at that point we will have dealt with each and every parameter of the SoundForum Synthesizer. The Nature of Frequency Modulation Frequency Modulation: This mega-sexy youth term represented a brand new form of sound synthesis in the mid-eighties. Lifted to cult status with the help of Yamaha’s DX synthesizers, FM virtually exiled good old analog synthesis until it was finally brought back to life in the early nineties thanks to the techno revolution. In the eighties, just about every voice from California to Berlin was almost exclusively framed in clangorous electric pianos, metallic basses and staccato marimbas that swirled through the stereo image. The classic chiming DX7 electric piano is still a standard instrument, especially in tear-jerkers by artists like Celine Dion. Today, the sounds of the eighties are at best still popular among retro fans. Besides, the FM capabilities of the SoundForum Synthesizer are no match for the complexity of the infamous DX series and its successors. They are included as an addition and are definitely valuable as such. Contrary to simple analog synthesis, a minimal understanding of the basic technology is absolutely essential if you want to program sounds in FM with any efficiency. A beginning course in FM with lots of practical experimentation follows for all of you for whom FM has remained a closed book thus far.

57

4 Synthesis Tricks—For Experts Only

What Does Simple FM Sound Like? Oddly enough, the entire FM capability of the SoundForum Synthesizer is actually limited to one single knob in Oscillator 1. And guess what it’s called? Right … ◆

Load snapshot 16, “Basic FM,” and play a few notes.

This patch is programmed so that the SoundForum Synthesizer generates very basic FM a la DX7. The result is a metallic, bell-like sound. How is it generated? ◆

Turn the “FM” knob in the Oscillator 1 section all the way down, to 0. Then, while you play, turn it up slowly until you reach 4500.

When the knob is at 0, a simple sine wave is audible and visible in the oscilloscope. The knob lets you continuously add “bell.” It’s got to have something to do with this knob, but what? Vibrato Is FM Is Vibrato … ◆

Set the FM knob to 50 and decrease the Interval value to −60. Play some low notes, below C3.

Now you hear the sine tone, but with vibrato, right? So do I, and there’s a reason. Take note: In frequency modulation the amplitude of one oscillator controls the frequency of another. In the case of the SoundForum Synthesizer: The output waveform from oscillator 1 affects the pitch of oscillator 2 like a pitch wheel being constantly moved up and down. The oscillator signal is not used as a sound source, but rather as a control source.

58

Frequency Modulation

… Or Trills We’ll soon know whether I’m right or not: ◆

Set the waveform in Oscillator 1 to “Puls” and increase the FM value to 100 to better hear the effect. (After this little experiment, set the waveform back to “Sin.”)

The vibrato turns into a trill due to the edgier shape of the pulse wave. You can’t demonstrate much more clearly what FM is all about. But how do we get from there to a bell sound? Just do everything I do— don’t worry, professional sound designers do the same thing often enough—and pay attention at every step whether you hear exactly what I write: 1 Turn FM all the way up. (You did set the waveform back to “Sin,” didn’t you?) The vibrato turns into a cheap “laser” effect. 2 Increase the Interval value in Oscillator 1 in half-tone steps while playing the g key on the computer keyboard (so that we are all playing the same note). Starting at an Interval value of around 16, you no longer hear the vibrato as a modulation of the sine tone—that is, you don’t hear tone and modulation separately. Instead, you hear a completely new sound. The reason is simple: Just as we, above a certain frame rate (number of images per second), no longer see the individual images but rather a continuous movement, so do we above a certain “speed” no longer hear the modulation as vibrato—we hear a new sound. This is the case when the modulated vibration—here oscillator 1—is in audible range. To test this: 3 Set the “Osc 1” knob in the Mixer to 1 and “Osc 2” to 0. Voilà: The point at which we hear a new sound instead of vibrato is the point where the oscillator enters the audible frequency range.

59

4 Synthesis Tricks—For Experts Only

Tonal Color As the savvy have probably already noticed, we have two sine waves at work here; two “tuning fork” tones with absolutely no overtones. And FM’s strength is the ability to create extremely complex spectrums out of something so utterly simple. The tonal color can be controlled via two parameters: The interval of the two oscillators and the level of the modulator. Let’s give it a try: 1 Set the Interval of both oscillators to 0 and the FM knob to around 3500. The result is a sawtooth-like sound with very few overtones. 2 Try the following Interval settings for Oscillator 1 and see if you can identify the changes in tonal color: 1

Basic electric piano sound

12

Square wave

17

Grandfather clock

19

Harp

24

Sharper square wave

30—40

Mallet (e.g. vibraphone)

41—50

Electric piano attack

3 Now try various combinations of Interval values for Oscillator 1 and 2; utilize the Detune function as well. The overtone or harmonic content is determined by the frequency relationship, referred to as the “ratio” in FM theory. Put simply: A musical interval between oscillator 1 and 2 (e.g. a whole-number ratio) results in a harmonious sound.

60

Frequency Modulation

The second factor—the number of harmonics and with it generally the “edginess” or overtone content of the sound—is controlled by the amount of modulation. This is what the SoundForum Synthesizer’s FM knob does: It acts as a “send” control to an invisible pitch-control input for oscillator 2. By way of automated control of the modulation amount, the tonal color of an FM sound can be modulated in time progression, similar to the popular cutoff-frequency effects in classic analog synthesis (although the effect sounds quite different). The SoundForum Synthesizer allows such automation via an envelope, the LFO or even a MIDI controller. 1 Load snapshot 16, “Basic FM,” again. 2 Set the Interval values at 19 for Oscillator 1 and 0 for Oscillator 2. The result is a simple harp-like sound. 3 While playing slowly, decrease the decay value (“D”) in the filter envelope (“Filter Env”). What happens? The FM knob (or rather the modulation) is being automated by the filter envelope; the modulation progression gets shorter as the decay time is decreased. This process is pre-programmed in the “FilterEnv→Osc” section: As you can see, the “A(mp) 1” button is activated, meaning that Oscillator 1’s amplitude is controlled by the filter envelope. Waveforms The basic study of FM synthesis, such as that developed by Dr. John Chowning at Stanford University in the early eighties, is based on pure sine waves. This makes perfect sense; working with a waveform that is completely free of overtones, you have total control over the resulting sound. That doesn’t mean, however, that other waveforms can’t form the basis of extraordinary FM sounds. Besides, we don’t have six oscillators to play with here, as is the case in the DX7, but only two. So using more complex waveforms as ingredients can not only be helpful but can also lead to very interesting results, as the following experiment demonstrates:

61

4 Synthesis Tricks—For Experts Only

1 Load snapshot 16, “Basic FM,” again. 2 Set the interval value for Oscillator 1 to −24 and for Oscillator 2 to 0. Turn FM up all the way. Already sounds pretty cool. Now for the icing: 3 Set Oscillator 1 to “Puls” (pulse wave). And now pull all the stops: 4 Slowly increase the “Symm” value; it sounds most interesting between 0.3 and 0.6. Anyone who dares to claim he knew all along that the SoundForum Synthesizer was capable of sounds like this will please stand in the corner for the remainder of the class while we point and laugh at him. Now the next experiment: 1 Leave the sound as is and switch Oscillator 1 to noise (“Nois”). 2 Set the attack time (“A”) in the filter envelope to around 66. Instant vacation on the beach, anyone? No neighbors, no “recliner wars.” And speaking of war: 3 Set the attack time back (to 0) and turn release (“R”) all the way up. Turn the FM knob down to around 1100 and set the interval for Oscillator 2 at 24. Greeting from “Das Boot.” Perfectionists can exacerbate the sonar effect by setting “Cutoff” and “Reson(ance)” in the Filter section to 12:00.

62

Frequency Modulation

Interlude I’ve given you a lot to digest in this section so far. I’m sure, though, that you have understood most of it. For this and all other experiments in this book, I heartily recommend the following: Think outside of the box. Play around with all the controls, not just the ones defined in the steps, and try to determine which ones do what in each situation. This will help your “aural memory” learn on its own what the synthesizer does. That—and only that—is the one important thing you should keep in mind during the following sound design exercises. By the way: Why the Synthesizer does what it does (did I already mention this?) is—academic ambitions notwithstanding—secondary, even if it doesn’t hurt to understand it. In any case, it should never distract you from the primary goal: to internalize the often-complex relationships between knob-twisting and the resulting sound. Visual FM We do have a waveform viewer in the SoundForum Synthesizer, and FM is very visual, e.g. it’s easy to see what’s happening. That said, I don’t want to deprive you of this last comprehension-building step: To see FM with your own eyes (on the oscilloscope). First the experiment preparations: 1 Load snapshot 4, “Basic Sine.” 2 In the Mixer section, set “Osc 1” to 0 and “Osc 2” to 1. 3 Press the t key on your computer keyboard and adjust the oscilloscope so that you can just see one and a half waveforms, as shown in the diagram:

63

4 Synthesis Tricks—For Experts Only abb 1

The basic FM sine waveform in the oscilloscope.

Now it gets interesting: 4 Slowly turn the “FM” knob in Oscillator 1 up to around 300 while observing the oscilloscope. You can see and hear how the original sine wave is modulated. You also see a flowing animation, caused by the modulation between the two oscillators. 5 Set the “Detune” value in Oscillator 2 to 0 and slowly turn the “FM” knob up again. The animation is gone, but you can see exactly what FM is doing to the waveform. You might want to set the oscilloscope to “Freeze” now and then when it gets too jumpy. The modulation amount—controlled by the FM knob—has an audible effect (more overtones) as well as a visual effect (more bending) on the basic waveform. Now let’s look at the visual effect of the frequency ratio: 6 Set the “FM” knob to 1000. 7 Hold the t key on your computer keyboard and activate “Freeze” in the oscilloscope. 8 Set the interval value for Oscillator 1 to the following values, one after another; at each value, play t and freeze the oscilloscope: 0, 12, 24, 48, −12, −36, −60.

64

Sync and Ring Modulation

It’s easy to see: When the modulator (Osc 1) vibrates at a higher frequency than the carrier (Osc 2), the waveform grows little bumps. How many depends on the frequency ratio between the two oscillators. We’ll let this subject lie for now, though. Science Fair If you have an insatiably inquisitive mind and would like to learn more about FM, the SoundForum Synthesizer is a wonderful tool. Here are a few experiments that you can easily carry out on your own: ❖ Create a few simple FM sounds based on the “Basic FM” snapshot. Electric piano, harp or square wave lead should be no problem for you. ❖ Load snapshot 11, “Basic Bass,” turn up Oscillator 2 in the Mixer and the FM knob, and try to add a metallic attack to the sound via FM. ❖ This one is only for true geeks: Determine what effect frequency modulation has in snapshot 19, “Basic Thunder,” and how the effect is generated. Hint: Try varying the parameters “FM,” Oscillator 2 “Interval,” the waveforms for Oscillator 1 and Filter “Cutoff.” OK, now you’ve earned a short break. That was a ton of information, which you should try to expand on and further internalize by experimenting on your own. It’ll get a little easier in the next section—I promise!

Sync and Ring Modulation “Now that we’ve warmed up on the FM machine, we’ll exercise the sync muscles and do some presses with the ring modulation barbells.” That’s how this section would start if the author were a fan of quasi-humorous introductions. But he’s not. Rather, I hope you’ve all survived the last section—it was admittedly a bit heavy. Look at it this way, though: It can’t get much worse.

65

4 Synthesis Tricks—For Experts Only

The two techniques we will be learning now, sync and ring modulation, are much easier and—in the eye of the sound designer—merely modified alternatives to frequency modulation. As we have seen, FM is widely used to create metallic sounds or space effects, or in general to add some edge to analog sounds. Sync and ring modulation also have their special applications. Sync lends itself well to the following effects: ❖ “Wiry” sounds (string-like), ❖ pleasantly metallic, cutting or “sawing” sounds and ❖ extreme modulations in which the tonal color also modulates. Sync is therefore very good at aggressive synth basses, biting poly sounds and cutting leads. Ring modulation, on the other hand, is more appropriate for people whose music represents their rebellion against their parents or the establishment. Somehow ring modulation always sounds threatening or depressive. Not always a bad thing, though, since it does present a number of possibilities, such as: ❖ Metallic or bell-like sounds, ❖ adding edge or dirt to innocuous sounds, ❖ noise effects to laser sounds and ❖ horror effects. Have I whetted your aural appetite? OK, let’s get to it. What Does Sync Sound Like? That one’s easy: Load snapshot 17, “Basic Sync.” Reminder: The snapshots for the first chapters are pre-programmed in the SoundForum Synthesizer. You’ll find them in the pull-down menu in the toolbar above the Filter section.

66

Sync and Ring Modulation

This sync sound is pretty raw and could be refined, whether just as a basis or as a lead sound. We’ll get to that a bit later. For now, let’s see how sync feels and how it is generated. 1 Load snapshot 2, “Basic Square.” 2 Turn “Osc 1” in the Mixer all the way down and “Osc 2” all the way up. Then activate the “Snc” button to the left of the “Ring Mod” knob. Now it’s time to play a bit: 3 Turn up the “Interval” knob in Oscillator 2 in steps and play a few notes at each setting. As you can see, sync opens the door to waveforms and tonal colors that lie beyond the normal “analog” spectrum. Make a mental note of the fact that the basic pitch doesn’t change even though you are transposing oscillator 2. Interestingly, though, the sound does shift to the transposed pitch at the octave and fifth settings (12, 19, 24 etc.). 4 Set Oscillator 2’s interval back to 0 and experiment with the interval in Oscillator 2. Now the pitch does change, but really only at positive values. At negative values—at least below around −12—a kind of low-frequency noise is added to the sound. How Is Sync Generated? This section is optional; you can skip it if you prefer quick and easy pleasures to academic exercises … Here’s how sync works: Every time one oscillator reaches the start point of its waveform, sync sets the other oscillator at its start point. In this case, Oscillator 1 sets Oscillator 2 at its start point; it “synchronizes” Oscillator 2. This explains why transposing Oscillator 1 higher changes the pitch, since it then vibrates faster than Oscillator 2, which in turn forces it to vibrate faster.

67

4 Synthesis Tricks—For Experts Only

When you transpose Oscillator 2 higher, Oscillator 1 is still defining the basic pitch; the faster vibrations from Oscillator 2 act more like an additional tonal color. You can observe this phenomenon in the oscilloscope: 1 Set both oscillators’ “Interval” back to 0. 2 Set Oscillator 2 to “Saw.” 3 Play and hold a note and adjust the oscilloscope so that you can see one full waveform. 4 Now begin turning up the “Interval” knob in Oscillator 1. You see how the sawtooth waveform is interrupted—in such a way that it always returns to its null point. This is because Oscillator 1 “syncs Oscillator 2 to its knees” every time it reaches its null or start point. Enough theory; time to see and hear the three above-mentioned sync effects in the flesh. Sync Lead Now we’re going to build a sync lead sound—the chain saw of every good keyboard soloist. 1 Load snapshot 3, “Basic Saw.” 2 Turn Oscillator 1 (“Osc 1”) all the way down in the mixer and Oscillator 2 (“Osc 2”) all the way up. Then activate sync with the “Snc” button. The basic parameters are set. To actually hear the sync effect, we need some kind of pitch modulation. 3 Activate the “P 2” button in the “Filt-Env→Osc” section. This routes the filter envelope to the pitch input of Oscillator 2. Still not really a killer sound, though. 4 Set the “Amount” knob in the same section to 1 and increase the decay time (“D”) in the filter envelope (“Filt Env”) to 50. Now you’ve increased both the effect of the filter envelope (Amount) and the time that it’s audible (Decay).

68

Sync and Ring Modulation

Sounds pretty “syncy” already; now we’re going to use the filter to make it really mean. 5 In the Filter section, set “Cutoff” to 90 and “Env” to 40. Now the filter envelope affects not only the sync sound, but also the filter frequency. Since we have high expectations, we’d also like the lead sound to develop a sort of feedback effect when a note is held: 6 Set “Interval” in Oscillator 2 to 12. Now the sound is edgier and the filter movement ends in a fifth. Now if that doesn’t arm you perfectly for a heavy duel with your guitarist … The sound gets even nastier if you turn up the (filter) resonance; you might even want to set the attack time in the filter envelope to 10 so the sound “snaps.” Metallic Bass Here we don’t use sync to create a cutting tonal movement, but rather as a stationary color to add some metallic edge to an analog bass sound. 1 Load snapshot 11, “Basic Bass,” and set “Env” in the Filter section to 86 to brighten the sound a bit. This results in a neutral pulse-wave bass with a nice attack. We want to make its analog character a little more metallic, though, so we need to add some ore. To do this, we’re going to bring Oscillator 2 into play and use sync to give it a metallic color. Here we go: 2 Activate sync (“Snc”) and turn Osc 2 in the Mixer up to 0.7. Now we’ve got an octave (remember the feedback effect at harmonic intervals ?). 3 Set Oscillator 2’s interval to somewhere between −4 and −1. Voilà: nutritional iron and trace elements; the bass sounds like a good old DX.

69

4 Synthesis Tricks—For Experts Only

The most effective way to vary the tonal character is with the waveform buttons for Oscillators 1 and 2, and the “Interval” knob. A universe of electronic bass sounds awaits your discovery! Heavy Beating Sync also lends itself to a flanger-like amplified modulation effect; very useful for basses, pads or fat lead sounds. 1 Load snapshot 13, “Basic Mini.” A cute little lead sound, which helps many nursing home-residing Keith Emerson fans while away the hours between their cardio funk sessions (No joke—there really are such sessions!). We’d like it a little more modern, or, to be more precise, mean and nasty. A simple task: 2 Activate “Snc” and the “P 2” button in the Filt-Env→Osc section. Our intention here is to create a slow modulation with the LFO that modulates Oscillator 2’s pitch and with it the color of the sync effect. 3 In the LFO section, set “Rate” to −3.5, “Amount” to 0.5 and the waveform to “Tri.” It already sounds a little like oscillator beating; the sound swims a bit. Now for the sawteeth. 4 Adjust “Interval” in Oscillator 2 to a value between 10 and 20, depending on how much bite you want the sound to have. Almost done. Using the following modifications, the sound can be made progressively meaner: 5 Turn Osc 1 down in the mixer. 6 Increase the “Amount” value in the LFO section. 7 Set the filter to a flatter mode, i.e. “LP1” or “LP2,” increase the cutoff value and, if you like, add some resonance. What you just did with the lead sound can of course be applied to any type of synthesizer sound, from basses to effects.

70

Sync and Ring Modulation

Ring Modulation While not quite as copious as sync, ring modulation is still quite interesting. The term was not chosen very wisely—the non-initiated, logical-thinking person couldn’t really begin to imagine what it might mean. It’s actually very simple: In ring modulation, two oscillators modulate each other’s volumes; mathematically speaking a multiplication of their amplitudes. Let’s take a quick look/listen: 1 Load snapshot 3, “Basic Saw.” 2 In the Mixer section, turn Osc 1 all the way down and RingMod all the way up. You can already hear ring modulation, but somehow it doesn’t sound very exciting. 3 Set Osc 2 to “Sin.” 4 Turn “Interval” in Oscillator 2 down to −60 and play a low note. Sounds like tremolo, doesn’t it? Sure does, and there’s a simple explanation: Oscillator 2 is vibrating so slowly that it’s below the audible range and acts more like an LFO. Since it modulates Oscillator 1, the result is a tremolo with a sine waveform. 5 Switch Oscillator 2’s waveform to pulse (“Puls”), and you get abrupt, jumpy modulation. OK, now we’re ready to see and hear ring modulation in action: 6 Hold the z key on your computer keyboard and gradually increase the interval in Oscillator 2. At around −30 the oscillator enters the audible range. Above that we begin to get new tonal colors, most of them non-harmonic, just like with frequency modulation. It sounds a little smoother, by the way, if you set Osc 1 to “Sin” and decrease the “Symm” value. This way, two sine waves are modulating each other. At this point we can begin to make sense of the story that ring modulation yields addition and subtraction of both frequencies. While the SoundForum Synthesizer doesn’t completely stick to the theory, it does harbor some of this. Let’s look at a couple of examples.

71

4 Synthesis Tricks—For Experts Only

Basic Ring Modulation We’ll start out with an instant recipe: Load snapshot 18, “Basic Ringmod.” Here you have the classic laser effect that ring modulation is famous for. To understand the patch, all you need to do is to turn down the “Amount” knob in the “Filt-Env→Osc” section and then turn it up again slowly while playing. You could also play around with the filter envelope. Got it? The filter envelope modulates Oscillator 2’s pitch, sending it through a universe of non-harmonic spectra. Bells One of ring modulation’s specialties is non-harmonics, an important attribute of bell-like sounds. Here’s a basic patch for hundreds of bell sounds: 1 Re-load snapshot 18, “Basic Ringmod.” 2 Deactivate the “Filt Env→Osc” section by switching off “P 2” and set both oscillators to “Sin.” 3 Adjust release (“R”) in the amplitude envelope (“Amp Env”) to 57 so the bells can ring out. Now you can create a virtually infinite number of bell sounds by simply varying the interval values for Oscillators 1 and 2. Two modification tips: ❖ Edgier sound: Set one of the oscillators to “Puls.” ❖ Gong-like sounds: Use low “Interval” settings, lower cutoff and resonance. Make It Dirty The SoundForum Synthesizer gives you separate access to the RingMod output (in addition to the two oscillators). This allows you to mix it into other programmed sounds to add a little dirt. It works best when the two oscillators are set to different intervals. Here are a couple of examples of easy snapshot modifications:

72

Sync and Ring Modulation

❖ Snapshot 11, “Basic Bass:” Set the interval for Osc 2 to 19 and add ring modulation via the RingMod knob. ❖ Snapshot 13, “Basic Mini:” Set the interval for Osc 2 to −5 and add ring modulation. ❖ Snapshot 20, “Modular Echo:” Set RingMod to 1, turn down Osc 1 and 2 in the mixer and experiment with Oscillator 2’s interval. I know you’ve heard it before, but I’m going to say it (again) anyway: Experimentation outside of the examples is not forbidden and can’t hurt … After the next chapter you will have already acquired the basic knowledge necessary to modify existing patches.

73

5 Conclusion of the Basic Course When you have read all the chapters, including this one, and done your share of sound programming experimentation, you will be able to rightfully claim basic sound designer skills.

Our Curriculum In the previous chapters we have continuously utilized the SoundForum Synthesizer’s factory snapshots: Very simple little models that we morphed into usable sounds while getting to know the functions of the SoundForum Synthesizer through flesh-and-blood examples. With this chapter I’d like to wrap up the basic course by examining the rest of the patches, which are a bit more “finished.” We’ll look at how they work and do some modifications. With each example we’ll be studying a different category of synthesizer sounds.

Laser Effects Laser effects, synth drums and many other sounds are based on filter selfoscillation. Filter self-oscillation?!? Sounds complicated, I know, but it really only means that the filter resonance is turned up so high that the filter begins to vibrate, creating its own sound. Thus the filter can be employed as a third oscillator or even as an independent sound source. The filter vibration is always a sine wave; you won’t find a waveform selection option here. 1 Load snapshot 10, “Basic Blip.”

75

5 Conclusion of the Basic Course

You hear a sort of “blip” sound that could otherwise only possibly be created by rubber ducks sporting x-rated modifications … The filter envelope controls the length and time progression of the sound by altering the filter frequency and with it the “pitch” of the filter oscillation. 2 Play around with the attack (“A”), decay (“D”) and release (“R”) times in the filter envelope while playing some notes. 3 Set the sustain (“S”) value in the amplitude envelope (“Amp Env”) to 1 so that the tone doesn’t die when you hold a key. Everything clear so far? What else can we influence? Well, the effect of the curve, for example: 4 Tweak the “Env” knob in the Filter section a bit. This control determines the amount of influence the filter envelope has on the filter frequency (cutoff). Now we’re going to create a proper Jean-Michael-Jarre laser sound by bringing the LFO into the equation. 1 Re-load the same snapshot (10, “Basic Blip”). 2 Set both the “Filter Env” and “Amp Env” parameters to the following values: A: 20, D: 75, S: 0.5, R: 75. You’ve already got a perfect sound for a dive-bombing duck in a cartoon … Moving right along: 3 Activate the “Filt” button in the LFO section and set “Amount” to 0.35. The LFO is now routed to the filter and its modulation depth set, and we’ve got the spiral dive for our rubber duck. Now we’ll turn it into a laser: 4 Adjust the LFO “Rate” to 20, increasing the speed of the modulation. 5 Set “Symm” to −0.99—this turns the soft sine tone into something more sawtooth-like. Feeling brave? Here are three exercises requiring you to apply what you’ve learned thus far: ❖ Turn the LFO modulation into a trill in fifths. ❖ Deactivate the LFO and try to turn “Basic Blip” into an electronic bass drum. It’s really not all that difficult but does require some thought.

76

Synth Basses

❖ Create a wind sound out of “Basic Blip” (you’ll need Osc 1 set to “Nois” (noise)).

Synth Basses Load snapshot 11, “Basic Bass;” a very simple, no-frills pulse-wave bass sound. You could easily use it in place of an electric bass without ending up with a synth-heavy mix. We will now program a variety of synth basses by way of small changes while discovering which parameters control which tonal characteristics. First exercise is a classic, bone-dry synth bass, instantly recognizable as such but not overbearing: 1 Set the waveform for Oscillator 1 to triangle (“Tri”) and turn “Symm” all the way up. This sets the SoundForum Synthesizer to a sawtooth waveform, which immediately outs the sound as synthetic. We also want it short and dry, so we’ll give it more resonance and a very audible, cracking filter crash at attack: 2 Set the filter resonance at 0.43. 3 Adjust the filter envelope’s decay time (“D”) to 28 and sustain time (“S”) to 0.4. 4 Turn the “Env” knob in the Filter section up to 70. And there it is, our arid sawtooth bass. From here we’ll move on to something really impressive: the kind of bass that turns popcorn to dust in the cinema. This sound is extremely fat with that typical dark, threatening decay. 1 Turn Osc 2 all the way up in the mixer. Set its Interval to −24 and Detune to 0.1. Now for the real phat: The SoundForum Synthesizer allows you to assign multiple voices to one note, virtually multiplying the number of oscillators at work.

77

5 Conclusion of the Basic Course

2 Set the number of voices and the unison option (“VCS,” “UNISON,” both in the toolbar) to 3. Talk about power. Now for the decay: 3 Increase the decay time (“D”) in the filter envelope (“Filt Env”) to 55 and the release time (“R”) in both the filter and the amplitude envelopes to 65. Done! Now it’s your turn: ❖ More resonance and less “Env” gives you the typical “Das Boot” bass. ❖ While you’re at it, give all the other filter types from BL4 to HP1 a test drive to see what effect they have on the sound. Tip: The two letters stand for the type of filter (e.g., LP: low pass), while the number indicates the steepness of the filter. You’ll learn exactly what that means a little later ❖ Program the following bass sounds, one after another, and fine-tune them to usable patches: A pulse bass with resonance and an octaver, a sawtooth bass without resonance but with a sub-oscillator (two octaves lower), and a “film score” bass with eight oscillators and slow attack.

Synth Horns Snapshot 14, “Basic Brass,” demonstrates very nicely how synth horns work. There are three crucial factors: 1 The waveform(s) of the oscillator(s); you should always start with a sawtooth. 2 A slight pitch envelope on the oscillator (refer to the “Filt-Env→Osc” section) simulates the pitch-stabilization phase at the attack. When using multiple oscillators, this effect adds to the fatness of the sound if only one oscillator is “bent.” 3 The filter simulates the blowing effect via a fast—if not abrupt—attack time and relatively slowly decaying envelope. Beyond that, our analog possibilities allow us to vary:

78

Synth Horns

❖ The basic sound, ❖ the simulated number of horns and ❖ the “bite” of the sound. First off, we’ll make the sound softer and thicker: 1 Set Oscillator 2’s interval to 1 and detune to −0.18. Adjust “Amount” in the Filt-Env→Osc section to 0.04. You’ll probably ask why increasing the pitch difference softens the audible detuning. Good question—it’s easy to explain, though: The filter envelope deflects the pitch from Oscillator 1. 2 Just to see what happens, turn “Amount” in the “Filt-Env→Osc” section all the way up and listen. Afterwards, set it back to 0.04. Since sustain is not set to zero, the pitch doesn’t end at its intended value. We compensated for this with the interval setting. In real life you would now tune the entire instrument down to adapt it to concert pitch. Softer still: 3 Increase the attack (“A”) and decay (“D”) times in the filter envelope (“Filter Env”) to 45. The sound development is now nice and slow and ballad-like. Now let’s fatten it up: 4 Set the filter to “LP2.” The sound gets softer and, at the same time, more sparkle gets through. 5 Lower “Cutoff” to 55 and increase “Env” to 80. Now it almost sounds like Toto’s “Africa,” known in certain circles as the national hymn of aging Top-40 keyboardists … And here’s your homework on synth horns: ❖ Turn down Osc 2 all the way in the mixer to give the sound a solo trombone character. ❖ How could you make a clarinet sound out of this?

79

5 Conclusion of the Basic Course

❖ Last but not least, program a brass section in octaves. For this, it makes sense to re-load the original snapshot.

Synth Strings Snapshot 15, “Basic String,” simulates something like a quartet of cardboard celli. The typical fat, floating modulation is crucial to synth string sounds— everything else is just a matter of personal taste. Not counting external effects like chorus or ensemble, this effect can be created within the SoundForum Synthesizer in two ways: ❖ Through vibrations caused by detuning the oscillators against each other. One gets a vibrato; the other doesn’t. This is how it’s done in the snapshot. ❖ Through pulse-width modulation, as seen in section “PWM” on page 52. This method doesn’t work well with the SoundForum Synthesizer, since the oscillators can’t be modulated individually—there is, after all, only one LFO. Let’s stick with method a. Right now, the sound isn’t very fat, since oscillators 1 and 2 are set to different waveforms; therefore, the modulation isn’t as effective as it could be. We can change that: 1 Set Osc 2 to “Saw.” 2 Balance the mix by setting both Osc 1 and Osc 2 to 0.9 in the mixer. It already sounds a lot phatter. The speed and intensity of the modulation can be controlled via the “Rate” and “Amount” knobs in the LFO module. You’ll soon discover how important these adjustments are. In order to transform the preset into a nice string pad, we’ll lengthen the attack and decay times and octave the two oscillators: 3 Set “Attack” in the Amp Env to 60 and “Release” in both the Amp Env and Filt Env to 65. 4 Set “Interval” in Oscillator 2 to 12.

80

Synth Strings

Now it’s starting to sound like strings. Whenever you octave the oscillators, be sure that any vibrato is on the higher of the two. Try it the other way around and you’ll see why. Now for the final touch: 5 Set the filter to “LP1” and “Cutoff” to 65. This causes the filter to set in at a lower frequency, while the quartered slope lets some higher frequencies through, making the sound less edgy— a bit smoother. And now a nice space pad to demonstrate the fact that the string model is also capable of producing fully synthetic sounds: 1 Set both oscillators to “Puls” and turn “Symm” (“Puls-Sym” in Oscillator 2) all the way to the left. 2 Set “Interval” to 0 and “Detune” to 0.11. Nice and hollow—artificial and cold. Now we’ll underline that with a cavelike filter effect: 3 Set the filter to “BP4” and “Cutoff” to 60. Now the filter acts as a band-pass, letting only the frequencies right around the cutoff frequency through and suppressing not only the higher but also the lower frequencies. The passing frequency band is accented by the resonance. In addition, we’re going to deactivate the envelope: 4 Set “Env” in the Filter module to 0 and “Reson(ance)” to 0.6. Done. Now it’s your turn again: ❖ Create a warm synthesizer pad. ❖ How can we make the sound fatter? (Hint: we did it already—see above.) ❖ Try transforming the snapshot into a single violin (you’ll have to deactivate one oscillator).

81

5 Conclusion of the Basic Course

Weather and Echo Last but not least you should utilize the knowledge gained in these first chapters to discover how the two effect sounds at the bottom of the snapshot list work. I’m going to leave you on your own here with just a couple of important tips to help you reach the goal: 1 Load snapshot 19, “Basic Thunder.” 2 Vary the following controls in the order listed, always setting them back to their original positions: ❖ FM, ❖ Interval, ❖ Cutoff, ❖ Resonance, ❖ Amount (Filt-Env→Osc). You might want to make a note with each variation regarding which aspect of the sound each parameter effects. Analyze the sound Sherlock-Holmes style and then try to vary it in a useful way. Now for the last snapshot: 1 Load snapshot 20, “Basic Modular Echo.” 2 Vary the following controls in the LFO module: ❖ Rate, ❖ Symm, ❖ Amount. That should have been much easier. Congratulations! If you’ve gotten this far, you now have a basic understanding of how a synthesizer works and how its parameters affect each other. The fact that a good deal of experience is missing and there are still lots of holes to fill shouldn’t diminish your well-earned sense of pride.

82

6 SoundForum Synthesizer Tutorial In this chapter, we’ll explore the individual modules of the SoundForum Synthesizer in depth. Most of the basic knowledge gained will be applicable to virtually every synthesizer, giving you something for the rest of your synthesizing days. In my opening words, I vehemently defended the premise that one does not have to devour endless amounts of dreary synthesizer theory to attain the capability to program sounds. And, in the foregoing chapters, we jointly proved this admittedly somewhat questionable premise to be true: You have already tackled several fairly complex sound-design challenges without even beginning to feel laden with theoretical ballast. If you’ve made it this far, you obviously enjoy programming sounds, and I assume you want to dig a little deeper. Before we can really go much further, though, I have to be able to assume that you’re completely familiar with terms like “resonance” and “LFO rate.” The goal of this chapter is to examine the functions and modules of the SoundForum Synthesizer more closely and, with the help of several experiments, to give you a clear idea of what a synthesizer is capable of. This knowledge can of course be applied to other synthesizers of the same basic type, whether it be a thirty-year old analog dinosaur or a state-of-the-art software synth. This tutorial should also motivate those of you who already felt capable of sound programming to stop and re-evaluate. Or, as John Wayne might have said: “Let the horses drink and get a good night’s sleep. Tomorrow we’ve all got to give our best, and then we’ll show you who’s the boss around here!”

83

6 SoundForum Synthesizer Tutorial

Basic Functions Before we embark on the long trek through sound generation, I’d like to examine a couple of functions that are part of the daily routing of every dedicated sound designer. Among others, these are the functions for loading and storing sounds, assigning controllers and programming especially fat sounds. Detailed explanations for all functions—including the less interesting and more important ones—can be found in the chapter “The SoundForum Synthesizer” on page 19. Loading and Storing Sounds and Banks Filemenu

The File Menu of the SoundForum Synthesizer.

As we learned in the first chapters, a SoundForum Synthesizer sound is referred to as a “snapshot.” The entire status of the synthesizer, including an almost unlimited number of snapshots, can be stored as an “ensemble.” This nomenclature was derived from Reaktor, Native Instruments’ synthesizer modeling system, upon which the SoundForum Synthesizer is based.

84

Basic Functions

Here, “storing” refers to writing data in a file on a storage medium, e.g. a harddisk, whereas “saving” refers to placing a set of data in RAM. When storing an ensemble, not only the snapshots, but also virtually all device settings—including for example sample rate, MIDI controller assignments and maximum number of voices—are stored with the file. We will explore these possibilities in more depth in a following section. Ensembles can be loaded and stored via the first three entries in the file menu. And snapshots? You guessed it: “Load/Save Snapshots.” I suggest using snapshots to share sound banks with other SoundForum Synthesizer users. Using ensembles could wreak havoc on your buddy’s sound card and controller settings, unnecessarily stressing a good friendship. Storing Single Snapshots storesnapshot

Click on the camera symbol, enter a number and name and click “Store”: Voilà—you’ve stored a snapshot.

If, in the course of your experiments, you come up with something useable that you’d like to save for posterity, you can do this by selecting the “Store Snapshot” entry in the instrument menu.

85

6 SoundForum Synthesizer Tutorial

A dialog box opens, enabling you to name and store your current settings as a snapshot, so that you can load it later from the snapshot list. (The same dialog is more easily accessible via the camera symbol.) The upper part of the dialog box also allows you to overwrite or delete individual snapshots. Interactive Help mousehelp

If interactive help is activated via the “Show Hints” button at the left of the toolbar, a text will appear at each control element describing its function.

Each function and parameter of the SoundForum Synthesizer features an integrated help text when the mouse points to the respective control element or area on the panel. ❖ To activate or deactivate interactive help, click on the “Show Hints” button (arrow/question mark symbol) in the toolbar. You’re invited to take a brief tour with the mouse …

86

Basic Functions

I recommend deactivating interactive help once you’ve become accustomed to the SoundForum Synthesizer; as nice as help texts are, they can get on your nerves after a while. Remote Control of the SoundForum Synthesizer As with most comparable synthesizers, the SoundForum Synthesizer’s control elements can be remote controlled. Each parameter can be assigned to a MIDI controller number, allowing MIDI remote control of the parameter. This allows you to control and program the SoundForum Synthesizer using an external controller like a USB keyboard, or to use a MIDI software sequencer to enter and edit controller data, automating the synthesizer with a song. You can also create an editor with a sequencer program for editing and archiving sounds. The reverse applies as well: you can also control external synthesizers with the SoundForum Synthesizer. MIDI controller assignments are “device data,” meaning that they only need to be made once (not per snapshot); they are stored with the ensemble file. The easiest way to assign MIDI controllers is to use the “MIDI Learn” function: 1 Activate the MIDI Learn function via the corresponding button in the toolbar (MIDI jack symbol with an “L”). 2 Click a parameter on the SoundForum Synthesizer panel and turn or push the desired control element on the external controller. Done. Controller assignments can be inspected, edited and/or manually entered in the Parameter Properties dialog box (see illustration below), opened by double-clicking the control element.

87

6 SoundForum Synthesizer Tutorial context_properties

In this example, Oscillator 2’s mix control is assigned to MIDI controller 23.

The left half of the dialog relates to the individual control element, while the right half affects the entire instrument. Here’s the procedure for manual setting: 1 Activate the “Remote” option for the control element. 2 Enter the desired controller number for remote control of the parameter. Done. As an extra option for the more adventurous, parameters can also be controlled via polyphonic aftertouch. We won’t discuss that here; anyone wishing to take advantage of this feature should know how to use it. Generally speaking, you will be using standard MIDI controllers. It makes sense to use the same assignments for the SoundForum Synthesizer that you use for your favorite hardware synth; this saves time and thought, and allows you to run parallel tracks for both. I recommend reserving a dedicated ensemble for this purpose. You’ll have to experiment a bit with the assignments; not all parameters are as clear-cut as “Filter Cutoff” or “Attack Time.” For this reason, I recommend assigning only common parameters for remote control; otherwise you’ll end up in the jungle at some point.

88

Basic Functions

To remote control an external synth with the SoundForum Synthesizer (e.g. to realize parallel filter sweeps), active the “Panel to MIDI Out” option in the Properties dialog. If “Remote to MIDI Out” is activated, controller events received by the SoundForum Synthesizer are forwarded to external MIDI equipment. ❖ The Instrument Properties dialog is opened via a click on the Properties button in the toolbar (paper in hand symbol). Maximum Number of Voices The maximum number of voices, as well as the number of voices for unison mode, can be programmed for each snapshot. The former can be set in the “VCS” field at the right of the toolbar. The maximum number of available voices depends largely upon the processing power and speed of your computer and the sample rate. More processing power and/or a lower sample rate will allow you more voices. I recommend the following procedure to determine the maximum number of voices for your system: 1 Set “VCS” to a low value (2 to 5). 2 Select a sample rate in the System menu (44.1 kHz is recommended for optimal sound quality, whereas 32 kHz is fine for sound programming). 3 Slowly increase the “VCS” value. Wait until the processor load display stabilizes before each increase. Be sure to keep at least a 40-percent reserve in standby mode; the processor load increases as soon as you play. With this kind of reserve, even fast figures shouldn’t overload the processor. Speaking of overloads: They can lead to crashes, especially with systems that are not configured properly, and thus should be avoided (like a keyboardist should avoid the drummer’s girlfriend …). How nice that the SoundForum Synthesizer features an automatic function to prevent overloads:

89

6 SoundForum Synthesizer Tutorial

❖ Activate the “Automatic Voice Reduction” option in the Instrument Properties dialog (paper in hand symbol). To give you an idea of what should be possible: My timid little writing laptop (an IBM ThinkPad 1200 with a 550 MHz Celeron processor) easily generated 32 voices at 44.1 kHz and 40% reserve—far more than many hardware synthesizers. A good—or more appropriately nasty—sound for testing is snapshot 20, “Basic Modular Echo,” with quick figures. Unison The SoundForum Synthesizer allows you to play single notes with several identical voices that are detuned against each other, giving you 4, 6 or even 32 oscillators per note instead of two. This results in extremely fat, lush sounds that work especially well for leads, pads and effects. You can activate the unison mode directly by increasing the Unison value in the toolbar (far right). However, I recommend opening the Instrument Properties dialog and entering “2” in the “Min Unison Voices” field. Even with extreme Unison settings above 4, the SoundForum Synthesizer will attempt to play all notes by automatically reducing the number of unison voices in the case of processor overload. Although we already made acquaintance with the unison mode in the Synth Basses section on page 77, another small experiment is in order: 1 Load snapshot 15, “Basic String.” A friendly little pad sound that could use some fat. 2 Assuming you have activated Automatic Voice Reduction and correctly set Max Unison Voices, set the Unison value to 5. Quite an improvement already, isn’t it? We’re going to take it a step further, though: 3 In the Instrument Properties dialog you’ll find the parameter “Unison Spread,” which determines the amount of detuning between the unison voices. Set this parameter to 0.1.

90

Basic Functions

Even more phat! Careful, though: At values above 0.15 or so it starts sounding more like a hurdy-gurdy. Caution: The increase in the total number of voices also increases the volume, which can lead to distortion. If distortion occurs, try lowering the level in the Master module above the oscilloscope. Monophonic Sounds There are certain lead sounds that are preferably played monophonically. To achieve this, simply set VCS to 1. To play a monophonic sound in unison mode, set VCS and Unison to identical values. In this application there is the risk that the “Voice Allocation” settings (Instrument Properties) may work against you. 1 In the Instrument Properties dialog, set “Min Unison Voices” and “Max Unison Voices” to the same value and deactivate “Automatic Voice Reduction.” This way, the sound always plays monophonically, regardless of how many notes you play. 2 Try this with snapshot 13, “Basic Mini,” using the above-described settings. 3 For a thin but cutting unison sound, set “Unison” to 3 and “Unison Spread” to 0.02. Another Tip Unison sounds are especially hard on your processor. And it’s not much fun playing polyphonic unison pads if you have to stop every 30 seconds to reactivate audio processing due to processor overload. If you have a sampler, it’s not a bad idea to program a really fat sound with the SoundForum Synthesizer and sample it as a multisample. Then you can play it with the sampler’s full polyphony (and use the SoundForum Synthesizer for something else). With that, I’ll close the tips-and-tricks section of the tutorial. In the next section we’ll explore sound generation.

91

6 SoundForum Synthesizer Tutorial

Sound Generation Modules The second part of the SoundForum Synthesizer tutorial will familiarize you with the sound generation, with the focus on the interaction between the modules.

Audio and Control Modules The SoundForum Synthesizer’s sound generation architecture is identical to that of all synthesizers using analog synthesis. This includes old analog models by ARP, Moog or Yamaha, as well as the so-called “virtual analog” synths by Access or Clavia, and software synths like the Pro-53. In the “How Does Simple Synthesis Work?” section on page 41, I briefly explained the concept. Here we’ll be exploring the interaction between the individual sections or “modules” of the synthesizer. These modules can basically be split into two groups: Audio modules and control modules. Audio modules are circuits that generate or directly influence the audio signal. The chain of audio modules forms the signal path. Every synthesizer sound requires a time-based progression, e.g. the development of the sound between pressing and releasing the key, and a concurrent automatically generated modulation. Enter the control modules, which automatically vary or “modulate” such things as waveform, tonal color or volume.

92

Sound Generation Modules

Audio Modules A “normal” analog synthesizer’s audio modules include one or more oscillators, a mixer, a filter und an amplifier. Different synthesizers will vary in the number and exact configuration of modules; the SoundForum Synthesizer represents a good average, offering two well-equipped oscillators and a multi-mode filter above and beyond “standard appointment.” Controllable Sound Parameters

Module

Function

Role

Oscillator

Generates the basic waveform

Pitch Waveform

Sound generation

Mixer

Controls the balance between the oscillators

Output level of each oscillator; amount of ring modulation

Sound mix

Filter

Programs the filter characteristic

Tonal color

Sound shaping

Amplifier

Programs the volume

Volume

Volume control

This table shows the function and role of the various audio modules.

Basic programming of the audio modules looks something like this: ❖ Waveform and pitch are selected for each oscillator. Using different waveforms for each oscillator results in a waveform mix; different pitch settings result in modulation or an interval. ❖ The level balance between the oscillators is set in the mixer. The mixer outputs a waveform mix, which can already include intervals and modulation. This concludes the actual sound generation. ❖ The mixer signal is routed to the filter, where a tonal color is formed from the raw waveform mix. ❖ The amplifier controls the volume of the signal. As in most synthesizers, the SoundForum Synthesizer’s amplifier is invisible.

93

6 SoundForum Synthesizer Tutorial Audio_control

The SoundForum Synthesizer’s audio modules (dark gray) and control modules (light gray).

The illustration above is a stylized depiction of the SoundForum Synthesizer. The dark gray modules are the audio modules. The signal path flows from the upper left to the lower right. For the most part, the user interface follows this layout scheme, which was introduced with the good old Minimoog. The amplifier is not visible in the illustration, but is hidden behind the “Amp Env” module, which actually controls the amplifier (that’s why this module is partially colored dark gray). Sound Generation and Sound Shaping Let’s practice the process illustrated in the functions table on a “live subject,” precisely following the signal path. Important: Each step in the following sections builds upon the previous steps. Please do not overwrite or delete the sound between steps! It’s probably safest to store the sound after each exercise as a snapshot. 1 Load snapshot 3, “Basic Saw.” Oscillator 1 generates a raw waveform; all other modules are currently dormant.

94

Sound Generation Modules

2 You can influence the pitch, either by way of the note played or with the “Interval” control. 3 Change Oscillator 1’s waveform by clicking the four buttons “Puls,” “Tri,” “Sin” and “Nois.” Then select “Tri” and set “Symm” to 1. Now we’ll bring the second oscillator and the mixer into play: 4 Set “Osc 2” in the Mixer to 0.8 and set Oscillator 2’s “Interval” to 12. The simple waveform has been transformed to a mixed waveform. The “Detune” value of 0.02 in the Oscillator 2 module results in a light modulation. So much for sound generation; now we’ll move on to sound shaping: 1 Set “Cutoff” in the Filter module to 70. The sound becomes very dark. 2 Set “Reson(ance)” to 0.85. We’ll leave it at that for now. You can see what a powerful influence the filter has on the tonal color. Now we have a waveform mix whose tonal color has been processed by the filter. The resulting sound is pretty static, though, not to say boring. But wait: We have the control modules …

95

6 SoundForum Synthesizer Tutorial

Control Modules Control modules are synthesizer circuits that the audio signal does not directly pass. Their sole function is to generate control signals that automate specific parameters of the audio modules. These control signals can be cyclic in nature, i.e. vibrato or tremolo, or single progressions which are triggered by pressing a key, such as an envelope. A Brief Introduction to Control Voltage True analog synthesizers utilize control voltage as a control signal. The modules are interconnected via control voltage inputs. Thanks to a uniform control voltage, a principle credited to Bob Moog, virtually any parameter of any signal in the synthesizer can be modulated. Control voltages can be used to directly influence the pitch of the oscillators, the center frequency of the filter or the volume. Like similar virtual analog synthesizers, the SoundForum Synthesizer exactly reproduces this principle. Some modern virtual modular synthesizers like the Clavia Nord Modular even feature “patchable” graphic displays of the modules with inputs and outputs. In the SoundForum Synthesizer, these connections are invisible—at least almost (we’ll come back to the “almost” part later). Automating Sound Take another look at the “Audio and control modules” illustration (page 94) and pay special attention to the light-gray colored control modules. They are not yet active in our current example sound and won’t be until they are expressly activated. That’s exactly what we’re going to do now. First we’ll add some time-based modulation to the sound. We want the sound to develop quickly when a key is pressed and then fade slowly; the fade should affect not only the volume, but also the tonal color.

96

Sound Generation Modules

Time-based modulations are generated with envelopes. The SoundForum Synthesizer’s envelopes are called “Filter Env(elope)” and “Amp Env (elope).” 1 Maximize the amount of influence the filter envelope has by setting the “Env” control in the Filter module to 100. 2 Set the “Filter Env” module’s controls to the following values: A = 20, D = 45, S = 0.55, R = 60. 3 Set the “R” (Release) control in the “Amp Env” module to 60. 4 Play a note. The modulation of the sound is clearly audible: The filter frequency “moves” as if you first quickly opened the cutoff control and then slowly turned it down. This movement is automated via the envelope, which reacts to the key being depressed and released. By increasing the “R” (release) value in the “Amp Env” module, you programmed the slow fade of the sound after releasing the key. Now we’ll add a cyclic modulation to the sound; namely vibrato: 1 Set the “Amount” control in the LFO module to 0.13. 2 Click the “P 2” button in the LFO module to deactivate it. Now, only Oscillator 1 is affected by the vibrato. You’ll find out why in the next section. Contrary to an envelope, the LFO is not dependent upon depression or release of a key, but rather runs its course independently of any other action. The SoundForum Synthesizer’s LFO is reset with each new note. Some synthesizers allow you to choose whether it is reset or not. The effect of the LFO is similar to a continuous back-and-forth movement of the corresponding control.

97

6 SoundForum Synthesizer Tutorial Module

Function

Parameters

Purpose

Filter Env(elope)

Generates a time-based modulation of the filter cutoff frequency. Can be routed to oscillators via “FiltEnv→Osc”

Attack Decay Sustain Release

Tonal color modulation

Filt-Env→ Osc(illator)

Routes the filter envelope to various oscillator parameters

Modulation destination Waveform or pitch (switchable), modulation modulation amount

Amp Env(elope)

Generates time-based amplitude (volume) modulation

Attack Decay Sustain Release

LFO

Generates modulation vibrations

Waveform, symmetry, Effects (vibrato, wahrate, modulation wah, tremolo, trills, destination (switchable), random) modulation amount

Volume development over time

This table shows the function and purpose of the various control modules.

Activating Control Modules While the “Amp Env” module is essentially always active, the control signals of the “LFO” and “Filter Env” modules must be routed to a parameter in order to influence the sound. The “Env” parameter in the Filter module controls the influence of the filter envelope on the cutoff frequency. In addition, the filter envelope can be routed to various oscillator parameters; this is the purpose of the “Filt-Env→Osc” module. The LFO can also be routed to oscillators, filter frequency and amplitude. In the “Basic Saw” snapshot, for example, it is routed to the oscillators. The modulation source/destination routing is programmed via a combination of buttons, which determines the internal patching, as well as the Amount control, which determines the depth of the modulation.

98

Sound Generation Modules

To illustrate: In the “LFO” module, you’ll find a button named “Filt.” This button routes the LFO to the cutoff frequency. When you activate this button, the LFO signal is connected to the filter’s cutoff frequency. The extent to which the LFO influences the cutoff frequency is determined by the “Amount” control. Let’s give this a test run: 1 Using the same example sound, click the “P 2” button in the “FiltEnv→Osc” module and set “Amount” to 1. The result may not sound very pleasant, but it demonstrates the principle well: Oscillator 2’s pitch now follows the progression of the filter envelope. 2 For a somewhat less demonstrative but more pleasant sound, deactivate “P 2” and activate “Amp 2.” Now Oscillator 2 decays more quickly and the filter effect seems to be amplified. 3 For a more dramatic effect, set the oscillator levels in the Mixer module to 0.8 for Osc 1 and 1 for Osc 2. By simply modifying the routing, you have dramatically transformed the effect of the modulation. Brief Summary As mentioned earlier, the purpose of this section is to help you better understand the interaction of the various modules that make up the SoundForum Synthesizer. If you’ve done all the exercises correctly, the example sound that we’ve programmed in this section should look like this:

99

6 SoundForum Synthesizer Tutorial

This is what your example sound should look like.

In conclusion I’d like to present a brief analysis of this sound, which also forms the basis for your next bit of “homework.” Please compare the analysis step by step with the synthesizer panel. If the synthesizer is not available, the foregoing illustration will serve as a substitute of sorts. Correlate the descriptions with the parameter settings. Audio Modules ❖ Oscillator 1 und 2 each generate a sawtooth waveform. They are tuned in an octave and slightly detuned. ❖ The filter is set to a 24-dB low-pass with a high resonance value and is modulated by the filter envelope. ❖ The amplitude (volume) progresses neutrally until the key is released, after which the sound decays slowly. Control Modules ❖ The filter envelope generates a peak when the key is pressed, then remains at half-mast and decays slowly after the key is released. ❖ The LFO generates a light pitch modulation via a sine wave (vibrato), which affects only Oscillator 1, thereby intensifying the modulation.

100

Sound Parameters in Depth

❖ Via the “Filt Env→Osc” module, the filter envelope also controls the volume of Oscillator 2 (independently of the total volume), causing it to decay more quickly. Wasn’t too difficult, was it? Now for the “homework”: 1 Go through all 20 snapshots in the SoundForum Synthesizer and analyze the modules—first the audio, then the control modules. I suggest you take written notes of your results. 2 Select as many of your favorite sounds in your favorite synthesizer—it doesn’t have to be analog—and analyze these in the same fashion. You’ll be surprised at how easy this is, and you’ll almost be able to hear the mysteries of sound synthesis unravel. In the next section we’ll be working with a new set of snapshots while we take a complete tour of the individual modules, during which we will be putting all—and I mean all—parameters to the test.

Sound Parameters in Depth In this chapter we’ll be tweaking each parameter and experiencing what it has to offer.

Example Files For this chapter we’ll expand our snapshots: ❖ Load Tutorial.ens via the Open entry in the File menu.

101

6 SoundForum Synthesizer Tutorial

Oscillator 1 Oscillator1

Oscillator 1

The two oscillators of the SoundForum Synthesizer differ in just a few meaningful ways. 1 Load the snapshot “Oscillator 1.” It all begins with the buttons for waveform selection. The choices are Puls(e), Tri(angle), Sin(e) und Nois(e). Puls The width of the pulse wave can be continuously adjusted via the “Symm” control. 1 Play a few notes. In the default setting, the oscillator generates a square wave—we are reminded of the charm of early video games. This waveform is called “square wave,” meaning that the pulse and the pulse pause are exactly the same width. You can observe this in the oscilloscope. The trademark hollow sound is characteristic for older video game sound effects. 2 While playing, turn up the “Symm” knob. The pulse becomes narrower and the sound transforms from “hollow” to “nasal.” As we experienced in foregoing sections, pulse waves are universally usable.

102

Sound Parameters in Depth

Next to the sawtooth wave, pulse waves are the richest in overtones and therefore especially suited for “cutting” sounds of all types. Tri The triangle wave contains very few overtones, making it less “aggressive” and suitable for adding bass portions or for softer sounds. The SoundForum Synthesizer features a special trick: The “Symm” control “bends” the waveform clockwise until the triangle waveform becomes a sawtooth. 1 Reload the snapshot “Oscillator 1.” 2 Switch the waveform to “Tri.” 3 Hold a note, turn up the “Symm” knob and observe the oscilloscope. In this way, you can continuously vary the overtone content and with it the “sharpness” of the waveform. The resulting sawtooth is edgy and cutting. The sawtooth waveform contains all overtones in the harmonic series and is best suited for pads, brass, strings and other “whole grain” sounds. Sin The sine is a pure tone with no overtones, similar to the tone created by a tuning fork. As most of you know, any waveform can be dissected into individual vibrations with different frequencies and amplitudes; in essence, the sine wave is the atom of sound synthesis. The SoundForum Synthesizer also allows you to continuously transform a sine wave into a sawtooth; there is an important difference, however: the sine-based sawtooth is rounder and fuller than its triangle counterpart, and thus more suited to basses and pads. ◆

Observe the sine-sawtooth transformation on the oscilloscope.

103

6 SoundForum Synthesizer Tutorial

Nois Last but not least, Oscillator 1 also offers white noise. Common uses include the breath noise of a pan flute, snare sounds, wind, waves and steam. Technically speaking, white noise is a signal which contains all frequencies, statically and in equal portions—comparable to white light as the sum of all colors. Symm We’ve already used this control: It bends a triangle or sine wave clockwise, distorting the curve and generating waveforms with additional overtones. Interval This control transposes Oscillator 1 in half-tone steps. FM We explored frequency modulation in section “Frequency Modulation” on page 57. As a brief refresher on the effect of this control: 1 Load the snapshot “Oscillator FM.” Here, Oscillator 2 is turned up in the Mixer module. Why? Because it acts as the carrier in the FM oscillator pair, while Oscillator 2 acts as the modulator. The “FM” parameter determines the “volume” at which the output of Oscillator 1 is routed to the input of Oscillator 2; basically a sort of modulation depth control. In other words, the “FM” parameter determines the overtone content of the resulting sound spectrum, while the “Interval” parameter determines the overtone allocation or tonal color. 2 Experiment with both parameters and listen to the results. The waveforms of both oscillators also influence the resulting tonal color. Tip: Try white noise with a low FM value!

104

Sound Parameters in Depth

Oscillator 2 Here we only need to look at the differences from Oscillator 1: ❖ The “Puls-Sym” control determines only the pulse width. ❖ A fixed sawtooth wave is included. ❖ The symmetry of the “Tri” and “Sin” waveforms is not variable. ❖ There is an additional “Detune” control. Detune This control allows you to detune Oscillator 2 continuously. Well, almost continuously: the values are in hundredths of whole notes, or cents. Detuning Oscillator 2 of course only makes sense in combination with Oscillator 1—then it results in a modulation. ◆

Load the snapshot “Oscillator Detune” and listen to the modulation effect.

The effect is most noticeable when both oscillators are set to the same waveform, volume and interval values, especially with overtone-rich waveforms. Mixer Except for the “RingMod” parameter I would assume that the Mixer module is self-explanatory. Next to the cautionary tip not to set both oscillators to full level to avoid distortion, only one aspect of the mixer seems to merit mention: RingMod We paid our respects to ring modulation in section “Ring Modulation” on page 71. At this point I’d simply like to briefly reiterate on what the control does: ❖ There’s a small, invisible module inside the SoundForum Synthesizer that multiplies the output signals from Oscillator 1 and Oscillator 2.

Mixer

105

6 SoundForum Synthesizer Tutorial

❖ The “RingMod” control mixes the output of this signal with the two oscillator signals. Check out what ring modulation sounds like: 1 Load the snapshot “Mixer RingMod.” 2 Tweak the “Interval” and “Symm” controls on both oscillators. 3 Be sure to also experiment with different waveforms. Filter Filter

The Filter module of the SoundForum Synthesizer.

If one is to believe the comments found in numerous analog-freak forums, the filter is the most important aspect of any synthesizer. Some think this is absolute nonsense. We prefer not to become involved in subjective arguments, but rather to stick to the facts: The SoundForum Synthesizer’s filter is a rather luxurious multi-mode model. In other words, the filter characteristic is not fixed, as with simple filters, but rather selectable. To this end there are eight buttons on the left of the Filter module, each labeled with two letters and one number. What exactly do these labels mean? ❖ The first letter represents the actual filter mode: Low-pass (L), bandpass (B) or high-pass (H). Except for the combined band-/low-pass modes, the second letter is always P for pass.

106

Sound Parameters in Depth

❖ The number in the third position stands for the number of serial switched individual filters or poles. Each of these poles generates a slope of 6 dB per octave. Thus, a 4-pole filter is the steepest, with 24 dB/octave. A Brief Introduction to Slope The frequency at which the filter begins to operate is known as the cutoff frequency. A slope of 12 dB/octave means that frequencies which are one octave from the cutoff frequency will be attenuated (cut) by 12 dB. But it’s always better to hear than to read: 1 Load the snapshot “Filter Slope Test.” Like the name implies, this snapshot serves to test the slope and is programmed so that the signal from Oscillator 1—a sine wave—lays almost exactly one octave above the cutoff frequency when you play G2 (note number 55) on a MIDI keyboard or the t key on your computer keyboard. 2 Switch the filter between LP1, LP2 and LP4 while playing G2 or the t key. In LP (low-pass) mode, the filter only allows frequencies below the cutoff frequency to pass unchanged. The filter takes effect above the cutoff frequency. This explains why a low-pass filter is often compared to a high-frequency or treble control. You’ll notice that LP1 has the least noticeable effect (6 dB attenuation), while LP2 is a bit stronger (12 dB) and LP4 very noticeably attenuates (24 dB). 3 Set “Interval” in Oscillator 1 to 24. The attenuation now increases to 12 dB with LP1, 24 dB with LP2 and 48 dB with LP4, since there are now two octaves between the audio signal and the cutoff frequency. One more small experiment: 1 Set “Interval” to −12. The cutoff and oscillator frequencies are now identical. 2 Play G2 (or the “T” key) and slowly turn up the Interval control.

107

6 SoundForum Synthesizer Tutorial

The higher you turn up the Interval control, the quieter the tone becomes as the oscillator frequency is tuned further away from the cutoff frequency. Filter mode Following what we’ve learned so far, the HP1 filter is a high-pass filter with a 6 dB/octave slope, while BP4 represents a band-pass filter with a 24 dB/ octave slope. What exactly do low-pass, high-pass and band-pass mean? ❖ We just explained low-pass: It filters the frequencies above the cutoff frequency (highs). ❖ A high-pass does exactly the opposite: it attenuates the frequencies below the cutoff frequency (lows). ❖ Band-pass is a combination of the other two; it lets only a small frequency band around the cutoff frequency pass through unaltered. But we’re getting too theoretical again—time for another experiment: 1 Load the snapshot “Filter—Mode Test.” The basic signal in this patch is white noise, which allows us to perfectly hear the effect of the filter since all frequencies are present. 2 Turn the “Cutoff” knob back and forth while playing and listen carefully to the effect. That fat, smacking tone you hear is due to the half-opened filter resonance, which we’ll get to in just a moment. Now the actual experiment: 3 Switch through the various filter modes one at a time while playing, and pay attention to their effect. 4 Compare your impressions to the following table:

108

Sound Parameters in Depth Cutoff value

0

70

140

Mode

Filter lets through …

HP

Everything

Mids/highs

Only highs

LP

Only lows

Lows and mids

Everything

BP

Only lows

Only mids

Only highs

Which filter mode has which effect?

There are two more things you should know: ❖ A band-pass is made up of a high-pass and low-pass combined. ❖ The special combination BL4 consists of a 6-dB high-pass and an 18-dB low-pass. This often sounds more musical than the BP4 mode, since it attenuates highs nicely but is a little more gentle with the lows. That’s about the least amount of filter theory we can get away with. Now we can get on with the rest of the actual parameters. Resonance The filter resonance may be just one small, humble knob among many, yet it’s the thing that gives a sound that unmistakable synthesizer character. Technically speaking, resonance emphasizes the signal in the range of the cutoff frequency. Once again, listening is the key to understanding: 1 Load the snapshot “Filter Resonance Test.” Again, white noise is used. 2 Hold a note while gradually increasing the resonance value, and turn the cutoff knob back and forth with each new resonance setting. 3 Try the same with different filter modes. If you like, you can also switch to sawtooth or pulse waveforms occasionally. Although the filter seems to get “edgier” with increased resonance, what actually happens is that a very narrow frequency band in the cutofffrequency range is amplified—the filter effect remains the same. 4 Before you finish, set Resonance to 1.

109

6 SoundForum Synthesizer Tutorial

Such a high resonance value causes the filter to self-vibrate. You don’t only hear this vibration, but you can also play it—sort of like a third oscillator. K-Track This parameter allows you to play the filter like an oscillator. K-Track stands for keyboard tracking, and means exactly that: The filter tracks the keyboard. The higher this value, the more the cutoff frequency is transposed according to the notes played. This is especially effective with instrument sounds; without it, lower notes sound to bright and higher notes too dull. Again, a little experiment: 1 Load the snapshot “Filter K-Track.” This organ sound is created using the two oscillators tuned in octaves, with the filter acting as a third sine wave playing the part of a 2-2/3' stop. 2 Hold the i key on your computer keyboard. 3 Turn the “K-Track” knob down; the cutoff frequency moves. Compensate by turning “Cutoff” up to 92. 4 If you now play various notes with K-Track set to 0, the filter will not be transposed and the tonal color will remain the same regardless of the note played. When K-Track is deactivated, low notes sound brighter than high ones since they are further removed from the cutoff frequency. Tip: You should lower the master level when using high resonance values to avoid distortion. We forgot one control, “Env.” I have to admit, though, that this was at least partially intentional, to increase the anticipation of the following sections, in which we’ll discuss envelopes, LFO and the modulation module.

110

Sound Parameters in Depth

Amp Env The “Amp Env” module allows us to program a time-based amplitude curve for the sound by determining attack (A), decay (D), sustain (S) and release (R) times. Hmm … we’ve actually just explained the parameter/control labels … AmpEnv

A simple ADSR envelope (“Amp Env”) generates the amplitude (volume) curve.

1 Load snapshot 28 “Amp Envelope Test.” I did my best to program the envelope so that it’s not only audible, but also visible. 2 Play a note in the middle range and follow the envelope on three levels: acoustically, visually using the graphic display and optically using the oscilloscope. Now follow the curve through four phases: 1 First the volume increases (attack; “A”). 2 Next it abruptly falls (decay; “D” at 0). 3 Then it remains at a low level (sustain; “S”). 4 After the key is released, the sound fades (release; “R”). This type of envelope is referred to as an ADSR envelope. The name is derived from the four fixed phases: 1 Attack: The time after a key is pressed before the sound reaches its maximum level;

111

6 SoundForum Synthesizer Tutorial

2 Decay: The time until the sound reaches its “hold” level; 3 Sustain: The “hold” level of the sound; 4 Release: The time until the sound completely fades after a key is released. The following factors are important to note: ❖ All ADSR parameters control time, except for the Sustain parameter, which determines the hold level until the key is released. ❖ The A, D and S phases are run through while a note is held; the R (release) phase is entered once the key is released (hence the term “release”). Why a pre-programmed time-based curve? Well, an ADSR envelope is sufficient for a vast majority of all curves needed in practice, and is easy to program. Beginners love it, sound designers hate it, but it has survived since the early analog days. Complex digital synthesizers often use freely programmable phases with completely adjustable transition times and final levels. So much for basic knowledge. The following table shows ADSR settings for a variety of instruments and serves as a reference point; the settings represent approximately the instruments’ natural amplitude curves. Program these settings with snapshot 28 and vary them to get a feeling for the envelope’s effect.

112

Sound Parameters in Depth Instrument

A

D

S|s

R

Piano

0

65

0.0

25

Harp

0

40

0.0

40

Xylophone

0

25

0.0

25

Organ

0

0

1.0

0

Pipe organ

20

50

0.8

20

Strings

40

50

0.85

20

Horns

25

30

0.7

15

Reverse

50

0

0.0

0

Long Pad

65

0

1.0

65

Table of typical amplitude curves (Amp Envelope).

A couple of comments on these curves: ❖ When the decay and release times are identical, as with the harp, it makes no difference how long you hold the key. This of course is not the case with the piano. ❖ Minimal steps, like those in the decay times for pipe organ, strings or horns, are barely audible in the amplitude curve—but just wait till we get to the filter envelope! ❖ These settings are only meant to be simplified reference points and will vary from instrument to instrument, as from range to range.

113

6 SoundForum Synthesizer Tutorial

Filter Env Although the structure of the filter envelope is identical to that of the amplitude envelope, there are three crucial differences: ❖ The filter envelope controls the cutoff frequency and with it the development of the tonal color. ❖ While the amplitude envelope always fully influences the amplitude (similar to an inserted effect), the effect of the filter envelope can be varied or even completely deactivated via the “Env” control. ❖ The filter envelope can be inverted by setting the “Env” parameter to a negative value. 1 Load snapshot 29, “Filter Envelope Test,” in which the amplitude envelope and the filter are accordingly pre-programmed. The same curve is programmed here that we had in the amplitude envelope previously. By switching between snapshots 28 and 29, you can easily hear how much more audible the same curve is when it affects the filter instead of the volume. The following table shows several typical filter settings. You can see that the ADSR parameters alone are not enough—the other filter parameters are also very important. 2 Program the settings in the table, one after another. I intentionally did not pre-program snapshots, since hearing the finished settings would never have the same learning effect as step-by-step programming. Be sure to play a bit occasionally while programming to hear the effects of the changes being made.

114

Sound Parameters in Depth Instrument

Cut

Reso

Env

A

D

S|s

R

Piano/percussion

60

0.0

60

0

65

0.0

25

Disco bass

45

0.45

100

0

40

0.5

35

Techno bass

65

0.8

50

0

20

0.0

20

Long drone

45

0.5

85

0

75

0.0

75

Brass

70

0.0

100

25

50

0.5

10

Duck

70

0.7

70

30

50

0.25

50

Lead

80

0.25

55

10

45

0.6

0

Sweep

60

0.7

100

55

65

0.0

55

Reso blip

50

1.0

100

20

35

0.0

20

120

0.0

−100

40

0

0.0

80

85

0.0

−100

50

0

1.0

0

Double attack Echo

Table of several typical filter curves (Filter Env).

3 Also switch to band-pass and/or high-pass with each setting and pay attention to the changes in sound! Filt-Env→Osc This module of the SoundForum Synthesizer allows you to utilize the filter envelope to modulate the pitch, waveform and volume of the oscillators. FiltEnv_Osc

This module allows you to modulate various parameters via the filter envelope.

1 Load snapshot “Filt Env→Osc Test.”

115

6 SoundForum Synthesizer Tutorial

Again, the same curve is pre-programmed that we experienced with filter and amplitude. It’s clearly audible: The effect of an envelope on pitch is much more apparent than on tonal color or amplitude. The parameters in the Filt-Env→Osc module operate as follows: ❖ The “Amount” control acts sort of like an input control, determining how “much” of the filter envelope is routed to a specific parameter and thus the depth of the modulation. ❖ There are three buttons for each oscillator. “P” stands for “Pitch,” “Sym” for “Symmetry” and “Amp” for “Amplitude.” The number after each letter refers to the corresponding oscillator. Activating one of these buttons routes the modulation to the corresponding parameter of the respective oscillator. Activating “Amp 2,” for example, causes the filter envelope to modulate the amplitude of Oscillator 2, while activating “Sym 1” causes the filter envelope to modulate the symmetry of Oscillator 1—in the case of a pulse waveform this would modulate the pulse width. 2 With snapshot “Filt Env→Osc Test“ loaded, alternately click the “Sym” and “Amp” buttons for Oscillator 1 while playing. Got it? We’ll explore the use of these modulation possibilities in more depth later on. I did take the liberty, however, of programming a few snapshots for you: ❖ Snapshot 31 demonstrates how the interaction between the filter envelope and pitch with activated sync function can generate some heavy overtone action. ❖ Snapshot 32 demonstrates the brass effect that results when one of the two oscillators runs through a mild pitch curve in the attack/decay phase. ❖ Snapshot 33 is a synth bass sound using a single oscillator. The pulse width modulation caused by the filter envelope makes it sound as if two oscillators were active.

116

Sound Parameters in Depth

❖ In Snapshot 34, the filter envelope shapes the tonal color by controlling the volume of Oscillator 1 and thus the FM portion of the sound. The result is a classic FM piano sound—courtesy of a simple analog synthesizer. LFO The SoundForum Synthesizer’s main modulation source is the LFO. Contrary to an envelope, which generates a time-based curve, the LFO generates a repeating cyclic curve. The structure of the SoundForum Synthesizer’s LFO is fairly simple. The parameters serve to determine rate (speed, frequency), selection and variation of the waveform and routing to a modulation destination. LFO

The modulation center of any synthesizer: the LFO.

1 Load snapshot 35, “LFO Test.” This snapshot makes it easy to test all LFO parameters and hear their effect on the sound. 2 While reading the following parameter descriptions, turn and click the corresponding knobs and buttons and listen to the results. Reload the snapshot after each parameter.

117

6 SoundForum Synthesizer Tutorial

Rate This parameter determines the speed of the modulation. The possibility to set negative values is not a bug, by the way. Negative values are best suited for celestial, floating modulation effects; positive values for rotating effects (vibrato, fat pulse width modulation, rotor effects). Amount This one should really be self-explanatory. I should note, however, that the amount should not be set to over 0.16 for pitch modulation effects such as vibrato if the result is going to be halfway useable. LFO Wave Form Now we’re getting to the heart of the matter. Contrary to a regular oscillator, the waveform of the LFO has no influence on the tonal color, but rather on the character of the modulation: ❖ Sine: A soft, round vibration, e.g. for vibrato. ❖ Tri(angle): Also soft, but not as round as the sine wave. ❖ Puls(e): An alternating wave for trills or echo effects. ❖ S+H (Sample & Hold): Random values with sudden switches. Symm The possibility to freely determine the symmetry of the LFO waveform represents a very interesting feature that is rarely found in other synthesizers. It allows you to transform a sine or triangle wave into a positive or negative sawtooth, or to add inversion to a pulse wave. Try it out! We’ll be exploring this feature in depth a little later.

118

Sound Parameters in Depth

LFO Matrix You already familiarized yourself with this type of switch matrix in the “Filt-Env→Osc” module. It works exactly the same in the LFO module. 1 Load snapshot 35. 2 Alternately click the buttons in the upper row (Oscillator 1), as well as “Filt” and “Amp” while playing—this should demonstrate the possibilities. The table below shows which LFO settings generate which effect. Enter the parameter values one at a time, paying attention to the resulting effect. The basis for each setting should be snapshot 35. Effect

Rate

Amt

Vibrato

14.5

Trill

Sym

Matrix

0.12 Tri

0.0

P

10.5

0.2

Puls

0.0

P

Laser

20.0

0.6

Tri

−0.99

P

PWM

8.5

0.5

Sin

0.0

Sym

Wah-wah

11.0

0.5

Sin

0.0

Sym

Filter echo

5.0

0.7

Tri

−0.99

Filt

Tremolo

14.5

0.8

Sin

0.0

Amp

Reverse

8.0

1.0

Tri

0.99

Amp

Hacker

9.5

0.6

Puls

−0.6

15.5

0.8

S+H

0.0

Computer image

Wave

P, Filt, Amp P, Filt

Table of typical LFO effects.

Reproduce the effects in the table one at a time, and take the time to experiment with some variations. I guarantee: You will suddenly understand a few synthesizer sounds that until now seemed like pure magic.

119

7 Advanced Sound Design How do I get fat, warm, dirty, friendly, silky, lively, mean and nasty sounds? This has to be one of the most common questions posted by beginning sound designers in the numerous online forums worldwide. And you know what? Often, even seasoned pros have to pass on that one. Time and time again, letters from readers would contain requests for a sound-designing course that would teach others how the pros program their sounds. Regardless of the fact that this is a very difficult subject to begin with, this was clearly an impossible task without a platform that was accessible to all participants—like the SoundForum Synthesizer. You simply can’t learn sound design without hands-on examples and exercises. So now, after the extensive warm-ups in the initial chapters and the indepth exploration in the SoundForum Synthesizer Tutorial, I’d like to attempt to attain the unattainable: to teach you sound design capabilities that exceed even those of many who make their living as sound designers. “If it’s possible to teach sound design,” the critical reader may ask, “then why haven’t any of the hundreds of workshops and books published so far managed to do so?” My take is that most of these workshops and books have restricted themselves to explaining synthesizers as such. Always oscillators, filters, envelopes and LFOs. Sort of as if one was attempting to teach writing and focused entirely on spelling.

121

7 Advanced Sound Design

I have spent a lot of time thinking about how to teach sound design. And I agree with the traditional approach in so far as that a basic knowledge of the functions of a synthesizer is a prerequisite—one must know his tools or his “opponent.” If you have reached this point in the book, this requirement should be fulfilled. Now we come to the truly difficult and important part: The art of imagining sounds and realizing them on a synthesizer. Please pay good attention to this crucial message: A good sound designer must hear a sound before he programs it and knows how to get there before he starts. True: One can achieve many surprising results through trial and error. The fact that even trial and error requires a certain amount of ability to keep the error factor in check is proven by the “Frankenstein” sounds that are often produced by the random algorithms in sound editors. I’m going to do my best to teach you these abilities, which you should then be able to utilize with other synthesizers. As I’ve said before, your drivers license is not restricted to the car you take the test in.

From the Beginning Let’s start nice and easy by contemplating what is really important about sound design; which “secrets” separate a good sound designer from a mediocre one. The Challenge With the advent of modern electronic music, beginning with techno, classic synthesizer sounds slipped continuously more into the background. There are rarely models for the sounds used in genres like Techno, Trance, House, Ambient and so on.

122

From the Beginning

Even in Kraftwerk’s, Vangelis’ or Jean-Michael Jarre’s music, tangible parallels are hard to find. This is apparent in the names used for such sounds: “String Orchestra,” “Minimoog Bass,” “DX E-Piano” or “Synth Brass”. These days you can’t even describe the sounds on a Madonna CD using such examples, and Madonna is not exactly what you might call revolutionary sound-wise—we won’t even begin to talk about the typical instrumentation in contemporary electronic music. On the one hand, this makes the whole thing more interesting. On the other hand, however, it places new demands on sound designers. Sensual Perception Sound design has a lot to do with imagining sound. But how do you grasp or describe it? The oscilloscope is certainly no help here, nor is a parameter table. More often, terminology from the sensual world is used to describe sounds; what you hear can trigger a feeling or an image. Most terms used to describe sounds come from the realms of taste, sight and touch: ❖ You can virtually feel a punchy, wiry bass snapping. ❖ A warm, silky pad can trigger an image of a beach bathed in sunshine. ❖ Just about anyone can imagine what hard, fat power chords sound like. ❖ You also hear negative adjectives like tinny, thin or sterile in sound descriptions. How and whether such characteristics are perceived depends largely in my experience upon individual sound preferences. If a sound triggers a positive or pleasant sensual perception, it is considered a good sound—and vice versa.

123

7 Advanced Sound Design

The Power of Precedents This embodies one of the biggest paradoxes: Although everyone constantly screams for “never-before-heard” sounds, the most popular sounds are those that remind us of something else—consciously or sub-consciously. Piano, string orchestra, acoustic drums—these are the bestsellers among sample CD-ROMs. Good old herd instinct also plays an important role: Everyone who heard the DX7 electric piano on dozens—no, hundreds—of hits wanted to use it himself and thus participate in the success. The death-rattling “Shamus Theme” sound of the Roland D-50—one of the dumbest synthesis creations of all time in my humble opinion—won’t be put to rest; not even by much better substitutes. Why? Because it’s been heard in so many hit singles, film soundtracks and commercials that virtually everyone recognizes it immediately (and feels at home). What I’m trying to say is: The customary should not be underestimated. Only once in a while does a sound manage to become legendary (even among non-synthesizer freaks). Next to the above-named examples, one thinks of the Oberheim sound on Van Halen’s “Jump” or Phil Collin’s drum sound in the eighties. Taste Is No Secret Next to the customary, personal conditioning plays an important role in sound preferences. Different people react differently to the same sound. “Wiry” or “distorted,” for example, are not positive attributes for everyone. Nonetheless, there is consensus in many areas: Full, warm and lively sounds are pretty much loved by all. A fat horn sound generates the perception of something that can be chewed like toffee—just about everybody likes that. A reverberant sound can conjure up a pleasant, cool feeling.

124

From the Beginning

On the other hand, unintentionally distorted piano sounds can make you feel like you’re holding barbed wire, while a mushy bass sound is reminiscent of biting into a soggy pizza box. Remember: Sound preferences are determined by widely differing personal perceptions triggered by particular sounds. The perception itself— warm, metallic, cool, dirty—does not vary so much from person to person, but the effect it has varies greatly. The effect is also influenced by what the listener is accustomed to; all of this forms the personal sound preference. A good sound designer must be aware of these factors and know how to use them. I would even go so far as to say that this capability is twice as important and only a tenth as common as the basic knowledge of oscillators and ring modulation. A Brief History of Sound Synthesis In the beginning, the primary mission of the synthesizer was to simulate “real” instruments or noises. What an exciting proposition for someone who previously only had an organ or piano to work with—to be able to play guitar, trumpet or wind sounds on one of the first Moogs! In time, of course, other typical synthesizer sounds like Minimoog bass, Prophet synth horns or the Elka Synthex pad were introduced. Nonetheless, simulating “real” instruments as accurately as possible remained the most respected achievement for any sound designer. The advent of digital synthesizers did little to change this. Even with the (still) most popular DX7, only a small portion of the sounds were “new,” while the vast majority consisted of imitations. I remember well that I spent the entirety of my first days with a DX trying to program a realistic piano—that was the mark of a good sound designer in those days.

125

7 Advanced Sound Design

Why? Because there were no samplers. With the introduction of samplers, the synthesizer almost lost its right to exist. Samplers and their descendents, the PCM synthesizers (especially the Korg M1 and the Roland JV series), imitated not only real instruments much better than analog synthesizers, but also imitated analog synthesizers so well that the latter threatened to follow in the dinosaurs’ footsteps. Their saving grace came in the late eighties in the form of Techno, House and Acid, to name only a few of the currently popular modern electronic musical styles that largely ignored the general trends in instrumentation. Easily understandable: If there was anything that musicians and listeners could rightfully have had enough of by the end of the eighties, then it had to be sampled or “conventional” sounds. Techno and friends have not only survived, but have also continuously developed. A 12-year old today may find a drum loop more commonplace and natural than a piano. Just One More Thing … … namely the ability to translate everyday attributes into synthesizer parameters. You know, “How do I get a warm, fat, rich, punchy sound?” That will be the focus in this and the following chapters. Being a good sound designer does not mean being able to differentiate between the sawtooth in a Virus B and the sawtooth of a Minimoog, or knowing whether the TB-303 uses an 18-dB filter. These are merely words that can be memorized, but have nothing to do with feelings or concepts. And believe me, I’ve met enough “experts” who could listen to a Minimoog and guess its serial number but had no idea what to do with the thing. A good sound designer needs a good ear, but only for control. The actual art is translating an imagined sound into a real sound with the synthesizer he has at his disposal. The ability to turn an idea into a sound is important. Nothing else.

126

From the Beginning

This is proven by the fact that the few truly exceptional synthesizer programmers in the world can work with any synthesizer and deliver any sound to order—even an acoustic guitar with an analog synthesizer. The Levels of Perception In my experience, the reaction to a sound seems to take place on various levels of perception. The four most important are dimension, temperature, texture and animation. The impression that a sound makes on a person is a result of the interaction of these levels. Dimension This is the perceived size or spatial spread of a sound. Undulating pads or sounds with lots of reverb, multiple echoes and wide stereo or even surround spreads are perceived as being very big, while dry, thin sounds with little movement or modulation are perceived as being small. Temperature Is the sound warm or cold? Sounds with lots of low end, heavy modulation, little presence and large dimension are perceived as being warm, while thin, metallic sounds seem cold. Reverb and/or noise can emphasize this impression. Temperature is influenced by pitch, oscillators (i.e. waveform, modulation depth), the filter and possibly ambience. For example, reverb and delay are perfect for creating dimension. Texture You could substitute material or surface for texture, although they don’t do quite as good of a job at describing the internally experienced characteristic. Our perception compares the sound more to the effect on a material. A fat sound can be compared to chewing on something pleasantly soft, or to smacking your lips with enjoyment. A woody sound may be perceived like banging on a table (or a xylophone …), while an unnerving, metallic sound might remind one of fingernails on a chalkboard.

127

7 Advanced Sound Design

Considering that we constantly associate sounds with particular perceptions, feelings or experiences, it’s not really any wonder that the same happens with synthesizer sounds. If you’re looking to proactively shape texture on a synthesizer, reach for the oscillator and the filter. Animation This refers to the movement of the sound. Does it snap shut? Does it sound like it was hit, or does it softly fade in? Is it “stiff” or does it rotate in space? With acoustic instruments, the animation is usually governed by the technique: striking, bowing, blowing, specific articulation. With synthesizers, you’re dealing more with envelopes, LFOs and modulation. Terms used to describe animation include rotating, standing, fading or decaying, fading in, tremolo and vibrato. Interaction Obviously, these characteristics are not isolated, but rather interact; it is the combination of these perceptions that results in the final impression. In the case of a Hammond organ with a Leslie cabinet, the perception of a large file (texture) might combine with that of rotation (animation). Shaping or sharpening something dull with a file is generally a pleasant experience; thus the sound is perceived as pleasant. Distorted guitars are every bit as popular as the sound of frying fat. This also illustrates our premise: Sounds that remind of us things we find pleasant in everyday life are also perceived as pleasant. To this we add the above-mentioned factor of conditioning: What we’re familiar with, we automatically like better. How else do you explain the fact that even in the most complex digital synthesizers most sounds are still imitations, or that PCM synthesizers with ROM samples sell so well?

128

The Five Levels of Sound

The statements in this chapter form the basis for everything that follows in this book. Please internalize them. You may even want to take a break from learning for several days, then re-read them to be sure you have fully digested them before you continue. In the next section we’ll return to the synthesizer to sharpen your perception.

The Five Levels of Sound First off, a review of the core statements from the last two sections. Minimum Requirements The most important quality of a good sound designer is the ability to imagine a sound before it exists. Next, the sound designer must be able to translate this sound as closely as possible using whatever synthesizer is available. Case in point: Often, sound designers contracted to program sounds for a new synthesizer will receive requirements from the manufacturer; there are certain obligatory sounds upon which many synthesizers are judged, such as simulations of acoustic instruments. Regardless of whether sample-based, FM or additive synthesis, the sound designer has to realize the requirements on that particular synthesizer. Often, the synthesizer itself presents restrictions; more often, however, the limited capability or experience of the sound designer. That’s what we aim to change. If you cannot clearly define the destination, describing the route will be that much more difficult.

129

7 Advanced Sound Design

Sometimes, the excitement lies precisely in that which does not seem “possible” or “intended,” like using an analog synthesizer to create the perfect Rhodes simulation. Programming the perfect analog synth sound on an analog synthesizer is not exactly Nobel Prize material. Sound Preferences Result from Conditioning and Habit Conditioning through extensive music and sound experiences, combined with listening habits, are decisive factors in determining individual sound preferences. Positive and negative experiences, along with their associated sounds and music, dictate sound preferences like nothing else. Without wanting to go into too much detail, I have observed the following in my own experience and in my environment over the years: The individual preference “memory” is especially receptive in early years and/or in emotionally unstable phases, normally encountered in early childhood, puberty and with each new relationship with a member of the opposite (or personally preferred) sex. If, for example, one heard lots of Finnish folk music during a particularly pleasant vacation in his childhood, he/she is likely to be very fond of this type of music. The fact that he/she probably has a difficult time finding anyone to share this preference with outside of Finland does little to change this. At around 25 years of age, music and sound preferences are largely fixed, ready to influence the spontaneous reaction to anything new. New music or sounds no longer reach the subconscious as unfiltered as they did earlier. In other words: What you haven’t experienced by a certain point in time is unlikely to give you goose bumps afterwards.

130

The Five Levels of Sound

Which explains why 50-year olds consider techno and the like to be melody-free noise, or why 12-year old hip-hop fans associate the piano with funeral homes. Music and Sound Preferences Are Intertwined One inevitably prefers music in which his favorite sounds predominate. Similarly, one prefers to play—or program—the sounds that correlate with his favorite music. Among other things, this is the source of frustration for numerous nolonger-twenty-year-olds who power up their sequencers with the intention of producing a hard-core techno hit, only to end up with another REM tune … They simply can’t wear that other hat. As I mentioned earlier: Although people always seem to be looking for new, never-before-heard sounds, those with recognizable elements will always be received better; they simply touch the perceptive sensors more intensely. How else do you explain the phenomenon that star war films in which robots and laser swords play the leading roles are accompanied by orchestral music? Nobody seems to notice this actually fully obvious contradiction. While this may not really be a good thing, it is totally human. I can’t help but wonder why I’ve never seen a film on the Roman Empire with a soundtrack by Kraftwerk … The conditioning and habit factors must always be taken into account by sound designers. Whether they decide to utilize them or to fight them is their own decision. Remember, though: Know your friends!

131

7 Advanced Sound Design

Culture, Personality and Situation Culture, personality traits and listening situations also play important roles. Japanese manufacturer Roland recognized very early on, for example, that ROM players don’t sell very well when Japanese sound designers attempt to imitate American or European sounds—not to mention that nobody needs a workstation full of kotos or shamisens. The success of Roland’s JV series was not due to good looks or low distortion, but rather to the thousands of presets that satisfied Euro-American sound preferences. Depending upon personal conditioning, one will prefer aggressive or relaxed music and/or sounds. Folk music and speed metal fans tend to have relatively little in common. Then there’s the listening situation: Even choleric, violence-praising types aren’t likely to listen to Marilyn Manson records during autogenic training sessions, just as a pan flute is not the most effective instrument to get a club rocking. This, too, must be considered by the programmer of factory presets: “Usable” sounds do nothing to increase the sales of a synthesizer, but rather the sounds that are likely to cause accidents in front of the music store when demoed. Perception and Evaluation In my experience, the perception of music does not differ so much from person to person. Evaluation, however, underlies extreme differences. Probably just about everyone in the world perceives the sound of a koto as wiry, nasal and quickly decaying. While the Japanese are likely to associate the sound with a plentitude of pleasant childhood memories, however, the European is more likely to look for the nearest lighter. This influences the evaluation.

132

The Five Levels of Sound

Another take: Celine Dion is loved by many wonderful women because her music and voice tickle their well-developed tear ducts so well. For studio musicians, on the other hand, who have other criteria for measuring the quality of a voice, she may well represent the incarnation of Lucifer as a high-performance throat-voice artist. Having to sift through perception and evaluation makes it very difficult to define sounds and their character in any kind of usable and generally valid fashion. Yet these definitions are a prerequisite for teaching sound design. If you cannot clearly define the destination, describing the route will be that much more difficult. As I already implied in the last section, musicians and non-musicians alike naturally utilize attributes from the realm of the senses to describe their perception of sounds: warm, fat, wiry, wide, hard, cold. These perceptions vary little from listener to listener. The Levels of Sound Perception in Overview In the last section, we grouped the perception of sound into four levels of perception: 1 Dimension: The perceived spatial spread of a sound. 2 Temperature: Icy cold to hot. 3 Texture: Material characteristics that are “felt” when hearing a sound (wood, rubber, clay), or material processing techniques (striking, stroking, rubbing). 4 Animation: Curve and movement. Based upon what we’ve learned in this section, we could add one more: 5 Aggression.

133

7 Advanced Sound Design

Aggression More than anything else, aggression is the factor that separates a soft pad from a distorted techno snare. It could be defined as the “strength” of a sound, as in the power with which something is hit, whether it is more sympathetic (e.g. silky) or non-sympathetic (e.g. unpleasantly wiry or metallic). When designing sounds, one must pay close attention to the generation of aggression at each step along the way. No one appreciates a sound entitled “Soft Pad” that blows the listener’s fillings out of his teeth. An Analysis of GM Instruments In order to understand and practice dissecting sounds into the five levels of perception you need the smallest common denominator. This is exactly what the GM standard was originally developed for—a selection of the most-used sounds worldwide. The table on page 136 lists a selection of GM sounds and orders them in the five levels of sound perception. We’ll spare ourselves the temptation to once again make fun of GM sounds— we’re not concerned with honor here, but rather with a common basis. Besides, even Porsche drivers have to do their drivers license exam in a Golf … Let’s get active: 1 Go through the instruments one at a time, at first without listening to a keyboard. It’s very important that you first imagine the sound while reading its evaluation in the table. You’ll notice that my evaluations are subjective and don’t always coincide with yours. Remember also that the evaluation can only refer to an imagined average sound; depending on how it’s played, a sound can be very soft or extremely aggressive. 2 Enter your own evaluation, at least in the number columns, to record what kind of image the sound generates in your imagination. 3 Next, play the sound on one of your sound sources—preferably on a GM keyboard or module. Once again, we’re not concerned with quality here, but rather with the smallest common denominator.

134

The Five Levels of Sound

4 Now compare your current evaluation with the values you recorded in step 2. Don’t be concerned if your new evaluation differs from your first—in fact, this is desirable. The main thing is that you are actively evaluating on your own and thus training thinking in the five levels of sound perception. We’d all have it easier if sounds could be evaluated objectively. 5 Last but not least, try to create your own evaluations of sounds that are not included in the table. You may want to share evaluations with colleagues, or even organize weekend tournaments for your synthesizer club … No matter how you do it, spend as much time as possible analyzing sounds in this fashion. Just when you’re thinking just how diligent you’ve been, you’ll be hit just like in school: here’s another exercise … Homework Your assignment is to dissect the default snapshots 1 through 20 into the five levels. You’ll find an empty table for this on page 138. Outlook Next to sound design, the five levels can also be applied to other aspects of sound and music, such as technique/dynamics, mix, EQ or effects. There are, for example, effects that generate cold or warmth, emphasize ambience or transform texture, as well as gentle or aggressive EQ settings. Since effects and EQ are undeniably a part of sound design, they will be covered in the next section. Till then, have fun practicing!

135

7 Advanced Sound Design

A selection of GM sounds, divided into the five sound perception levels: Instrument

Dimension Temperature Texture 1: Small/narrow 1: Cool Perception 10: Large/wide 10: Warm

Animation

Aggression 1: Gentle 10: Aggressive

Grand Piano

3

3

Wooden strike; pleasantly metallic; vibrating

Slow decay, mildly undulating

4

Electric piano

3

6

Metallic click; soft cloth

Slow decay

3

Clavinet

2

4

Wiry, snappy

Quick decay

5

Glockenspiel

2

2

Metallic; short, hard strike

Very quick decay

5

Vibraphone

3

4

Soft metallic; bell-like

Slow decay; rotating

3

Xylophone

1

3

Wooden; short click

Very quick decay

6

Percussive organ

2

6

Metallic, wooden

Abrupt with click

5

Rock organ

3

6

Coarse sanding; metallic; smoky

Abrupt, rotating

8

Accordion

2

5

Compressed air; breathy; squawking

Softened organ curve; undulating

6

Nylon-string guitar

1

6

Picking; plastic; wooden

Quick decay

3

Jazz guitar

1

5

Rubbery; picking

Soft decay

2

Distorted guitar

2

5

Metallic; mill saw-like; sizzling

Slow decay

9

Slap bass

2

3

Punchy, wiry, snappy

Short, dry, quick decay

8

Fretless bass

2

6

Soft, silky

Soft attack and decay; mildly undulating

3

Violin

1

3

Wiry squeaking; wooden tone

Quick attack and decay

9

String ensemble

7

8

Soft, silky bowing; warm, wooden tone

Soft attack; undulating

3

136

The Five Levels of Sound Instrument

Dimension Temperature Texture 1: Small/narrow 1: Cool Perception 10: Large/wide 10: Warm

Animation

Aggression 1: Gentle 10: Aggressive

Synth strings

7

8

Silky; soft plastic

Soft attack and decay; undulating

2

Choir aahs

9

2

Breathy; singing

Soft attack; undulating

3

Trumpet

2

5

Powerful; compressed air; “like chewing toffee “

Quick blowing/ attack phase

6

Muted trumpet

2

2

Thin; cutting; metallic

Soft attack

8

Tenor sax

2

7

Smoky; wooden; compressed air

Soft, percussive attack

5

Flute

2

5

Breathy; soft; airy

Soft attack

2

Banjo

1

2

Wiry; tearing; wooden

Quick decay

6

Bagpipe

5

3

Cutting; metallic; squawking

Cardboard-like

8

Drums (kick, snare, toms)

8

7

Striking on a rebounding surface; punchy; pleasant

Punchy quick decay

7

Metallic screeching; strained

Slow decay

9

Breaking plastic

Punchy quick decay

6

Cymbals 808 drums

4

4

137

7 Advanced Sound Design

Empty table for your own analysis of the SoundForum Synthesizer snapshots. Snapshot

Dimension 1: Small/narrow 10: Large/wide

1 Default 2 Basic Square 3 Basic Saw 4 Basic Sine 5 Basic Tri 6 Basic Noise 7 Basic Detune 8 Basic PWM 9 Basic Reso Sweep 10 Basic Blip 11 Basic Bass 12 Basic Sub Bass 13 Basic Mini 14 Basic Brass 15 Basic String 16 Basic FM 17 Basic Sync 18 Basic Ringmod 19 Basic Thunder 20 Basic Modular Echo

138

Temperature Texture 1: Cool Perception 10: Warm

Animation Curve; Movement

Aggression 1: Gentle 10: Aggressive

Effects and Mix

Effects and Mix In this section we’ll continue to train thinking in sound perception levels while expanding the concept to include effects, arrangement and mix. In the previous two chapters we began practicing thinking in sound perception levels. As a reminder, the sound perception levels are not a meditation technique that I learned during a self-discovery vacation in Nepal, but rather a very effective method to transform anyone interested in sound into a sound designer. They provide a common language that allows us to characterize sounds relatively precisely and free of personal evaluations or preferences. The five levels are—yes, I’m repeating myself: Dimension, temperature, texture, animation and aggression. Each and every sound, whether natural or synthetic, can be analyzed using these levels.

Solution to the Homework Assignment The table below shows my analysis of the 20 snapshots. And here’s your assignment: ❖ Start the SoundForum Synthesizer. ❖ Load and listen to the snapshots, one at a time. ❖ Look at the analysis—especially the more subjective descriptions under Texture and Animation. ❖ Try to either confirm these descriptions or to find your own that more closely describe your perceptions.

139

7 Advanced Sound Design

My analysis of the SoundForum Synthesizer snapshots: Snapshot

Dimension 1: Small/narrow 10: Large/wide

Animation Curve; Movement

Aggression 1: Gentle 10: Aggressive

1 Default

3

3

Snappy; plastic

Blowing into the filter

3

2 Basic Square

2

2

Electronic

—

7

3 Basic Saw

2

3

Plastic

—

7

4 Basic Sine

1

6

Soft cloth

—

1

5 Basic Tri

1

3

Cloth

—

2

6 Basic Noise

3

2

Strong wind

—

7

7 Basic Detune

3

4

Plastic

—

6

8 Basic PWM

3

4

Sawing plastic

Rotation

6

9 Basic Reso Sweep

3

4

Soft plastic

Decaying

5

10 Basic Blip

2

2

Rubber band

Baring teeth

4

11 Basic Bass

1

6

Muted plastic string

Percussive

3

12 Basic Sub Bass

1

6

Plastic string

Punchy; percussive

4

13 Basic Mini

2

5

Touching clay

Blowing

4

14 Basic Brass

3

5

Touching rubber

Blowing, noticeable modulation

6

15 Basic String

3

5

Rubbing plastic

Bowing, noticeable modulation

5

16 Basic FM

2

2

Striking a bell

Slow decay

7

17 Basic Sync

2

2

Sawing metal

Decaying

9

18 Basic Ringmod

2

1

Electricity; sparks; buzzing

Stopping

9

19 Basic Thunder

6

3

Muted rumbling

Slow decay

3

20 Basic Modular Echo

4

5

Metallic, fat

Percussive echo

6

140

Temperature Textur 1: Cool Perception 10: Warm

Effects and Mix

Effects and the Sound Perception Levels It’s common knowledge that effects are an integral part of any sound. As early as the eighties, synthesizer smith Roland recognized that synthesizers with integrated effects sound better and thus sell better—one crucial reason for the success of their D-50, which still influences synthesizers and workstations today. An effect doesn’t sound like much on its own; it needs a signal to improve. Take a look at the table on page 143. It represents a selection of especially popular effects. The data in the table show the “direction” in which the effect transforms the sound, thus the plus/minus symbols. Since not every effect changes a sound on all five levels, several effects show no entry in certain columns. Where this is an empty cell, the effect behaves neutrally. What Effect Do Which Effects Have? Dissecting effects into the five sound perception levels is actually quite easy: ❖ Ambience effects like reverb or delay are there to enlarge the dimension of a sound. ❖ Modulation effects like chorus or flanger add warmth and movement to a sound. ❖ Distortion effects like overdrive are sure to add a certain amount of aggression. As you can probably deduce on your own, effects are often used—along with synthesizer parameters—to take a sound in a particular direction: wider, warmer, fatter or more aggressive.

141

7 Advanced Sound Design

Trying Out Effects Now you’re going to have to a little work again: 1 Using whatever parts of your setup work best, set up a piano sound you can play. This sound is overtone-rich and relatively neutral in regards to the sound perception levels. A mono sound is preferable. 2 Insert a multi-effect device into the signal path. You can, of course use the plug-ins for your audio/MIDI sequencer. If you own a workstation with built-in effects, simply use that. 3 Activate the effects in the table below, one after another. The basis for the values listed is always a standard setting. Be aware of the fact that some effects, i.e. overdrive or compressor, must be insert effects. 4 Try to confirm the descriptions in the table or to create your own that accurately describe your perceptions. How to Use the Table The following examples illustrate how to use and understand the table: ❖ Take the effect “Large Hall.” Imagine a small, harmless oboe with essentially zero ambience. If you add a long reverb, it will become somewhat wider, but more noticeably it gain depth, being placed further back in the room. This is represented in the table with an increase in dimension at the value 5. ❖ A phaser makes the sound spin. Even a monotonous waveform becomes animated. The perceived texture is a sort of “gentle whirlpool”—kind of like a water funnel. ❖ While an overdrive adds warmth to the sound and gives it a texture like simmering fat, a hard distortion effect generates a perception more like two pieces of metal rubbing against each other. ❖ A compressor reshapes the curve of a sound: Percussive sounds become punchier; sustained sounds get “squashed.”

142

Effects and Mix

Some essential effects and their influence on the sound: Effect

Dimension Temperature Texture +: Larger/ −: Cooler Perception further +: Warmer

Animation Movement

Aggression −: Gentler +: More aggressive

Large Hall

+5

−3

Lengthening

−3

Wood chamber

+3

+3

Lengthening

−3

Garage

+2

−3

Lengthening

Mono delay, long

+2

+2

Repeats

−3

Stereo delay, long

+6

+2

Repeats

−5

Chorus, wide

+5

+4

Fat

Undulating

−2

Flanger

+2

Sandpaper

Shimmering

+2

Phaser, slow

+3

Whirlpool

Swirling

−3

Ensemble

+5

Terrycloth

Complex modulation

−5

Wah-wah

+2

Rubber

Vocalization

Pitch shifter, detune

+5

Tremolo

+2

Vertical rotation

Auto-pan

+5

Horizontal rotation

Rotary

+3

+6

+3

−3

Butter

−3

+5

Greasy rubbing

Overdrive

+3

Frying fat

+4

Distortion

−3

Metallic rubbing

+7

Compressor Exciter

Complex rotation

Punchy, squashed −3

Breathy

−3

+2 +2

EQ and the Sound Perception Levels Equalization (EQ) can also be used to modify sound on at least four of the five levels. A table doesn’t make much sense here, since the exact frequencies to be boosted or attenuated vary widely from signal to signal. Nonetheless, you can pinpoint pretty accurately how changes in the frequency spectrum affect the four levels.

143

7 Advanced Sound Design

Aggression takes place in the upper mid-frequencies; these frequencies are perceived by the human ear as “invasive.” EQ and Dimension The perceived vertical dimension of a sound is governed by its frequency content. It’s no coincidence that we refer to “lows” and “highs”—this not only stems from the Hertz value, but also from the fact that low frequencies are perceived as coming from below, while high frequencies seem to come from above. Interestingly, bass speakers are almost always positioned below high-frequency speakers. When you add low frequencies to a sound, you pull it down in terms of sound perception. When you add high frequencies, you push the sound upwards in the acoustic image. EQ and Temperature Temperature is the level most noticeably affected by EQ, especially in the lower frequencies. A “warm” sound is always bottom-heavy, while “cool” sounds generally have a deficiency in low and lower mid frequencies. High frequencies, on the other hand, exercise very little influence on the temperature. A slap bass with sparkling highs can still sound warm, while a reverb-saturated high-pass pad will always seem cold, regardless of how much high end it has. Any sound will get warmer when low mids and lows are added. You can often make an extremely cool sound much warmer, although it’s always easier the other way around. EQ and Texture While EQ cannot create or completely change the texture of a sound, it can intensify or attenuate the texture. Texture perception takes place almost exclusively in the mid- and high-frequency bands, so that you can easily emphasize or mask a sound’s texture by simply boosting or attenuating the corresponding frequencies.

144

Effects and Mix

EQ and Aggression You’ve probably never heard of “aggressive” lows, but certainly of “aggressive” mids or highs. Aggression takes place in the upper mid-frequencies; these frequencies are perceived by the human ear as “invasive.” Hungry babies, ambulance drivers and recorder players know this and use this knowledge without mercy. By boosting frequencies between around 700 Hz and 4 kHz, you noticeably increase a sound’s aggression factor; attenuating the same frequencies makes the sound more pleasant and less invasive. End of the Lecture That was a lot of new stuff—especially a lot to read—in just one chapter. In the next chapter we’ll begin practicing thinking in the five sound perception levels on the synthesizer, and I hope to teach you how to bend a sound in any direction you desire with deliberation and good aim, regardless of whether you start with a raw waveform or are modifying an existing sound. By the way, this craft is called … Right: Sound design. Until then you can work out and get fit by: ❖ going through as many of your sounds as possible and try to analyze them according to the five sound perception levels; ❖ trying to determine why you like some sounds and don’t like others. Describe this as accurately as possible using the attributes dimension, temperature, texture, animation and aggression, and try to define what your personal preferences are; ❖ practicing thinking in sound perception levels whenever you hear music, whether it’s your favorite CDs or just something on the radio.

145

8 Advanced Hands-on Course After the previous three chapters of pure theory on the five sound perception levels, we’re finally ready for an advanced hands-on course with the SoundForum Synthesizer. The last chapter was more of a theoretical introduction to sound design based on the five sound perception levels dimension, temperature, texture, animation and aggression. Now we’re going to get down to business, sit down at the SoundForum Synthesizer and practice these techniques. Since I don’t want to overestimate the tendency to internalize vital information in this age of lightning-fast zapping, we’ll toss in refreshers on the contents of the previous chapters here and there. In case you should be gifted with the admirable tendency towards thoroughness, I can only recommend that you read the previous chapter several times and repeat all the exercises before you continue. Here again for good measure are the best of those exercises, some of which you can even do while washing dishes or working out: ❖ While listening, arranging or playing, always dissect sounds into the five levels. Whenever possible, make notes in a table. ❖ Analyze while listen to music: Which sounds influence the musical impression in what way and why? When does something sound cold, aggressive or lively? ❖ Get a hold of every synthesizer within a 100-kilometer radius and tweak knobs. Which parameters affect the sound on which level? ❖ Do the same thing while arranging and mixing; the levels affect the entire process, from composing to mastering. There are also cold and

147

8 Advanced Hands-on Course

warm, aggressive and friendly harmonies, as well as hectic and relaxed phrasings. Don’t worry: You won’t be filling out tables for the rest of your sound designing days … On the contrary: The better and more experienced you become, the less you will consciously think about and utilize the five sound perception levels. At some point you will delete it from your consciousness entirely; it will be permanently stored in your subconscious. This requires, however, that you get on your own nerves for at least six months by analyzing even cinema spots in regard to temperature and aggression. This way of thinking has to be literally burned into your mind—later you’ll be programming sounds on your synthesizer in your sleep.

Sound Perception Levels and Synthesizer Modules The crucial question is: How do I translate the five levels to the synthesizer once I’ve internalized them? For all of you who have asked themselves quietly whether the 5-level method isn’t maybe just some kind of esoteric trick I thought up, now the crucial argument for its use: It translates directly to the synthesizer and it always works, with any sound and with any synthesizer. Pretty convincing, huh?

A Brief Tour of the Synthesizer Modules I’d like to deliver proof of this statement right away. This brief tour of the synthesizer modules will give us a first, very rough idea of which module influences which aspects of the sound.

148

A Brief Tour of the Synthesizer Modules

Lest you get too euphoric: This is merely basic knowledge. Exactly how each individual parameter affects the sound and how the modules and parameters interact—that’s the true science, for which we’ll need a couple more chapters. Oscillators and Mixer Load snapshot 4, “Basic Sine,” into the SoundForum Synthesizer. This is an absolutely neutral sine wave: No overtones (thus very warm), no movement, no width. In five-sound-perception-levels-speak: Dimension



Temperature  Aggression



Texture

Soft cloth

Animation

—

Slowly turn up the “Symm” knob in Osc 1: The sound’s texture changes; its “edge” (aggression) remains static but it gets a little wider and more “cheeky.” The evaluation looks a bit different now:

Dimension

Before

After





Temperature 



Aggression





Texture

Soft cloth

Plastic

Animation

—

—

149

8 Advanced Hands-on Course

Now turn up Osc 2 in the Mixer module, switch Oscillator 2 to “Saw” (so that both oscillators sound alike) and increase the “Detune” value to 0.05. The result: The sound gets fatter, livelier and thus warmer. The movement sounds like a pleasant rubbing. Texture and aggression barely change at all. The evaluation at this point: Dimension



Temperature  Aggression



Texture

Plastic

Animation

Rubbing

Now set “Interval” in Oscillator 2 to 12. The sound becomes more powerful but loses warmth. Another change in the evaluation: Temperature  Aggression



Keep this setting for the following section; you may want to go ahead and store it as a snapshot. A brief summary for the oscillators: They determine the basic sound and thus the initial values for temperature, aggression and texture. They can also influence dimension or animation somewhat; however, they need help here from envelopes or LFOs.

150

A Brief Tour of the Synthesizer Modules

Filter The filter mode is set to LP4 (steep low-pass). Turn the cutoff control back and forth a couple of times. What changes? The sound becomes less aggressive when you lower the cutoff frequency; it gets warmer and transforms the texture from hard to soft. Dimension and animation do not change noticeably. The mode determines the basic characteristic, while the cutoff frequency actually controls the affect on the sound. For this reason, there are two ratings for two cutoff frequencies here: Cutoff

Low

High

Dimension





Temperature 



Aggression





Texture

Plastic

Soft

Animation

Rubbing

Undulation, Tremolo

Resonance Set the resonance to 0.6. The sound changes, most noticeably when you quickly turn the cutoff control. Two levels are extremely affected by this: The temperature falls by at least three values, while the aggression increases by four or five over the entire cutoff range. Filter Mode Switch the filter mode to HP2. In this mode, the filter doesn’t influence the high frequencies at all, but rather filters low-frequency information out of the signal. Now the influence of the filter looks entirely different:

151

8 Advanced Hands-on Course Cutoff

Low

High

Dimension





Temperature 



Aggression





Texture

Plastic

Electric current

Animation

Rubbing

Sizzling

To summarize: Starting with raw material from the oscillators and mixer, the filter shapes the most important sound-defining levels: Dimension, temperature, aggression and texture. Even the character of existing animations is emphasized. When you look at the comprehensive influence the filter has on the sound, it becomes easy to understand why the filter is such an important factor for synthesizer enthusiasts in judging the quality of a synthesizer. Filter Envelope We continue our experiment: 1 Set the filter back to LP4, “Cutoff” to 60 and “Env” to 100. Suddenly, the sound abruptly snaps down. The filter envelope generates— not surprisingly—an animation; in this case, a “snapping” movement. 2 Set “A(ttack)” to 20 and “D(ecay)” to 30 in the “Filer Env” module. Now the animation sounds more like lip-smacking; the sound’s texture changes from plastic to something more rubbery. Since texture and animation cannot be displayed in a gradual fashion, we’ll do without our little tables here. Later on, we’ll present a list of envelope settings with their resulting animations and textures.

152

A Brief Tour of the Synthesizer Modules

Amplitude (Volume) Envelope And now a brief experiment with the amplitude envelope (Amp Env): 1 Set “A(ttack)” to 50 and “D(ecay),” “S(ustain)” and “R(elease)” to 0 in the “Amp Env” module. The lip-smacking is transformed into a reverse effect—sort of like “lipsmacking while inhaling.” Since, as they say, the third time’s the charm, we’ll add a pitch envelope: 2 Neutralize the amplitude envelope (all parameters to 0), activate both “P” buttons in the “Filter Env→Osc” module and set “Amount” to 1. The lip-smacking has now turned into a sort of whipping. 3 Set “A(ttack)” in the “Filter Env” module to 50. That’s how easy it is to make a vinyl-scratching noise out of a whipping sound. 4 Now deactivate both “P” buttons in the “Filter Env→Osc” module and activate the “Sym” buttons. Set “Symm” and “Puls-Sym” in the oscillator modules to 0 and set both oscillators to “Puls.” Now you’ve essentially animated the oscillator’s symmetry controls. The animation varies the texture, as well as the dimension: the sound becomes wider. The brief experiments are only meant to give you an idea of the possibilities—you don’t need to understand anything yet. To summarize again: Both envelopes generate animation. The filter envelope has a much greater influence on the sound, since changes in tonal color are much more obvious than volume curves. The filter envelope can also be routed to pitch or pulse width. The envelopes generate and influence the animation, texture and to some extent the dimension levels. The aggression factor is almost totally dependent upon whether the envelope fades in softly or attacks in an abrupt, percussive manner.

153

8 Advanced Hands-on Course

LFO The last of the modules is the good old LFO. I believe we all know what’s coming now: If anything obviously generates animation, it’s the LFO. There are already good examples of this among the default snapshots: Load snapshot 20, “Basic Modular Echo.” This patch generates an echolike effect via a combination of repeats through the LFO and a decay through the envelope. Here we have two animations working together. Here are two very interesting, somewhat ambiguous experiments: ❖ Set the LFO waveform to “Puls” and the “Symm” parameter to 0.08. If you now play a few notes, you’ll notice that the sound is much more ambient and the echo is more clipped, less flowing. In this case, the LFO influences animation and dimension (more ambience). ❖ Activate the “Sym” buttons in the LFO module and set the waveform to “Tri.” Now the LFO also affects the texture. Here are a few general examples of how which levels can be influenced by the LFO: ❖ Dimension: When the LFO generates echo-like modulations, the sound is expanded along the Z-axis (to the rear). When the LFO generates undulations (PWM or Pulse Width Modulation), the sound becomes wider. ❖ Temperature: The LFO has almost no influence on this level. ❖ Aggression: The LFO can make a sound much more aggressive, especially through pitch modulation. Think for example of a fast, overdone vibrato. Filter or volume modulations don’t offer quite as much scope, unless the LFO is set to self-vibration in an audible range. ❖ Texture: Undulating modulations, one of the LFO’s specialties, can change the texture considerably, for example from smooth to very rough plastic. ❖ Animation: This is the LFO’s true domain. Regardless of how and where it’s used, animation always results.

154

Mission: Possible—Exercises in Goal-oriented Sound Design

This is how it could look in table form: Dimension

Medium

Temperature Minimal Aggression

Medium

Texture

Medium

Animation

Very strong

Mission: Possible—Exercises in Goal-oriented Sound Design After our brief analysis of a synthesizer based on the five-sound-perceptionlevel approach, we will now deepen our newly won knowledge through experimentation. To recapitulate: Sounds can be dissected into and described using five levels: Dimension, temperature, aggression, texture and animation. These five sound perception levels are influenced by the various synthesizer modules and parameters as illustrated in the table below:

155

8 Advanced Hands-on Course Module

Parameter

…

Level

Oscillator Mixer

Waveform Interval

Generate

Dimension Temperature Aggression Texture

Detune

Generates

Animation

FM, Sync, Ringmod Modify

Texture Aggression

Mode (LP/BP/HP) Cutoff

Shape

Temperature Dimension

Cutoff

Controls

Aggression Texture

Resonance

Increases

Aggression Temperature

Filter Env(elope)

Generates

Animation

Amp Env(elope)

Generates

Animation

LFO

Generates

Animation

Filter

Influence of the synthesizer modules and parameters on the sound perception levels.

In addition, we’ve seen that these terms can also affect other aspects of a music production: effects, EQ, mix, arrangement and even harmonies. From Levels to Adjectives Since introducing the five levels, we have described sounds by dissecting each into the levels and assigning a value to each level. In this fashion, a sound can more easily be identified via relatively clear, quantitative values. As sensible as this may be for the introduction to level-based sound design: Nobody really thinks in these categories. You want a sound to be “warmer” and “fatter,” not “Temperature −3” and “Dimension +4.”

156

Mission: Possible—Exercises in Goal-oriented Sound Design

For this reason, we will now take off our water wings and try to grasp the levels in the form of adjectives. Since we’ve been practicing the “quantization” of sound characteristics for some time now, it should be easy for you to use the common adjectives for sound description; now you will be able to assign and translate them more precisely than ever before. Missions We’ll practice just that in this section by manipulating the SoundForum Synthesizer snapshots step by step in a specific direction. I have prepared ten missions for you, all based on a particular snapshot. I’ll accompany you through the first steps; after that, you’ll be on your own. The missions increase in difficulty as you go along. Advanced students are asked to be patient if they should feel under-challenged at first for the sake of the less experienced. At the beginning of each mission, load the snapshot referred to in the heading, then follow the directions. At the end of each mission there will be assignments for you based on the final result of the written directions. The assignments also increase in difficulty as you go along. Mission 1: Soft Melody out of Basic Square Raw, overtone-rich waveforms always sound very aggressive, sterile and unpleasant. A square wave has the tendency to gnaw at tender eardrums with its unmistakable gaming-machine charm. That can be changed. We want to transform this aggressive sound into a warm melody tone with a soft attack and take away its aggression, while we mask its sterility via a light tremolo. Here are the steps:

157

8 Advanced Hands-on Course

Less Aggression, More Warmth ◆

Set “Cutoff” to 80.

By filtering the upper harmonics, you can take away the edge from any sound. This works best, of course, when there’s enough low-frequency content, which is almost always the case with raw “analog” waveforms. Soft Attack ◆

Set “A” and “R” to 35 in the “Amp Env” module.

A(ttack) determines the time between key depression and maximum volume; R(elease) the time between key release and minimum or no volume. By softening the volume curve we’ve done a bit more to tame the beast. Animation 1 Set “Rate” in the LFO module to 8. 2 Deactivate the upper “P” button and activate the “Filt” button. 3 Set “Amount” to 20. Now the LFO modulates the cutoff frequency in the filter, generating a light vibrato, which gives the sound a thicker texture and masks its sterility. Your Assignments Now we come to the assignments that you’ll be carrying out on your own. Each step builds on the previous one, and I’ll give you a clue for each step as to which module you’ll need: 1 Increase the aggression (Filter). 2 Transform the hollow tonal color into a fuller one (Oscillator; waveform). 3 Change the filter animation to a vibrato.

158

Mission: Possible—Exercises in Goal-oriented Sound Design

Mission 2: Juicy Bass out of Basic Saw In no time at all, we’re going to transform a similarly sterile, aggressive sawtooth into a bass whose animation is reminiscent of “whipping” and with a texture that conjures up images of Jamba Juice. Whip It 1 Set “Cutoff” in the Filter module to 70 and “Env” to 80. 2 Set “A(ttack)” in the “Filter Env” module to 13. This sets the percussively programmed filter envelope in motion, generating the animation. The actual “whipping” effect is due to the low attack time, which causes the filter to open abruptly. Juice It ◆

Set the filter resonance to 0.7.

With analog synthesizers, “lip-smacking” is almost always generated by an abruptly opening filter envelope combined with resonance. Due to the abrupt opening, the filter’s resonance frequency itself creates a water droplike sound that lends “moisture” to just about any sound. Bass ◆

Set “Interval” in the Oscillator 1 module to −12.

Your Assignments 1 Program the bass so that it decays in a percussive manner and doesn’t “smack” anymore (envelopes). 2 Make the sound softer without using the filter (Oscillator; Symm). 3 Soften the tonal color even more while leaving the sound static (Filter; Cutoff, Env). The result should be a nice sub-bass sound.

159

8 Advanced Hands-on Course

Mission 3: Opera Voice out of Basic Sine Here’s a sound with which you can impress at least your closest relatives— easy to program yet sounds very difficult: A cutting, killer-vibrato appointed opera diva voice can be created with just a few steps out of a simple sine wave. Pay attention to the order of the steps, which I will explain as we go along. Vibrato ◆

Set “Amount” in the LFO module to 0.2.

It already sounds like opera, even if entry and tonal color aren’t quite right yet. Smooth Entry ◆

Set “A(ttack)” to 26 and “R(elease)” to 44 in the “Amp Env” module.

This causes the voice to “rise to the occasion” rather than to “fall through the door,” and it decays slowly instead of abruptly going silent. Vocal Power ◆

Set Oscillator 1’s “Interval” to 24. Select the “Sin” waveform for Oscillator 2 and turn it up to 0.8 in the Mixer module. Set Oscillator 2’s “Interval” to 12 and “Detune” to 0.

With this move, you’ve added an additional sine wave an octave higher, giving the sound a good deal more cutting power. The Icing on the Cake 1 Set “Amount” to −0.2 in the “Filt-Env→Osc” module and activate both “P” buttons. Now the pitch is modulated by the filter envelope, creating a sort of lifelike intonation. Since the envelope doesn’t land at the null point, though, the final pitch is not yet accurate. 2 Set the “Filter Env” parameter to the following values: A = 0, D = 39, S = 0, R = 39.

160

Mission: Possible—Exercises in Goal-oriented Sound Design

Now each new note begins with an upward “entry,” which masks the sterility of the sound. The LFO and the pitch envelope generate the entire animation. Just for fun, turn them both off by deactivating all “P” buttons— sounds like a video game now, huh? You see how important animation is, and as in this case, it doesn’t have to be original, just create an original impression. Your Assignments 1 Program the vibrato so that it only affects the upper octave, creating the impression of two voices. 2 Change the interval of the two oscillators so that an organ-like tone is generated (tip: the difference must be at least one octave plus a harmonic interval). 3 Remove the “entry.” 4 Program a percussive amplitude and tonal color curve (Amp Env). The result should sound something like a percussive jazz organ. Mission 4: Electronic Delay out of Basic Tri Time for a little break: In this mission, we don’t change all that much, but merely add a small animation to the sound that gives it more ambient depth via a simulated echo. Waveform Animation 1 Set the “Symm” control in Oscillator 1 to 0.45. 2 Deactivate both “P” buttons in the LFO module, set “Amount” to 0.6 and activate the “Symm” button for Oscillator 1. Now the sound vibrates hectically between a soft and an overtone-rich waveform—not yet the animation we’re going for. Echo Repeats ◆

Set the LFO rate to a slow 6 and turn the “Symm” knob all the way to the left.

161

8 Advanced Hands-on Course

This changes the shape of the LFO animation from gentle rotation to percussive repeats—in other words, an echo. Echo Decay ◆

Set the “Amp Env” parameter to the following values: A = 0, D = 50, S = 0, R = 50.

As you can see on the graphic curve, the sound now gradually decays after the initial attack—just like an echo. Your Assignments 1 Intensify the echo effect using the filter. Remember that you have to activate the animation with the filter as well. 2 Make the sound wider and more undulating (Oscillator 2). 3 Invert the echo so that it sounds like a reverse effect. 4 Intensify the animation again using filter resonance. The result should sound about 200 percent more powerful, wider and clearer than in the first part of the mission. Mission 5: High Winds out of Basic Noise Wind is a classic synthesizer effect, whose genesis you can now follow step by step. It is based on white noise; as presented in the Basic Noise snapshot biting, sterile and choppy. We’ll need to give the sound a darker color, the characteristic whistling and the typical rise-and-fall effect. Actually three very simple steps, if you know how to go about it. High-wind Tonal Color ◆

Set the filter cutoff to 60 and resonance to 50.

The cutoff setting darkens the noise, while the resonance intensifies the color. Both changes give the noise more depth and dimension—that’s because they give it more of a reverb character. The aggression disappears.

162

Mission: Possible—Exercises in Goal-oriented Sound Design

The Rise and Fall As wind “rises and falls,” two things are happening: It gets louder and quieter, and the pitch of the whistling noise follows the volume. We’ll recreate this using the filter and the amplifier, in combination with the corresponding envelopes. 1 Set the “Amp Env” parameters as follows: A = 65, D = 65, S = 1, R = 65. Now the volume swells up and down when you press and release a key. 2 Set the “Filter Env” parameters to exactly the same values as “Amp Env” and turn the “Env” knob in the Filter module up to 30. Now the tonal color—the whistling—also follows the volume curve. That was pretty easy, wasn’t it? Have fun playing St. Peter. Only for a minute, though, then straight to the assignments based on the wind sound. Your Assignments 1 Program a crisp, decaying volume curve with a length of about 150 ms (Amp Env). 2 Do the same with the curve of the tonal color, but make the decay approximately equal to that of a bass drum (Filt Env). 3 Make the tonal color extremely dark and program a very distinctive filter curve (Cutoff, Reson, Env). The result should be a monstrously phat, punchy techno bass drum. Mission 6: Extra-wide Brass Pad out of Basic PWM As the name implies, this snapshot also serves up a pretty much still initialized sound. Just click on “Sym 1” in the LFO module, and you’ll hear a fully neutral pulse wave—absolutely free of any kind of sound or volume shaping. Reactivate the button, and the LFO modulates the pulse width of Oscillator 1. Back in the days when an oscillator was still an expensive piece of hardware, this feature was used as a substitute for a second oscillator; the effect adds a modulation-like animation to the otherwise sterile sound.

163

8 Advanced Hands-on Course

Let’s plan our dream sound: ❖ First of all, the sound has to be a good deal wider and fuller. We’ll accomplish this by adding Oscillator 2, somewhat detuned, and switching to Unison mode. ❖ Next we’ll create the brass effect by sending Oscillator 2 through a little pitch envelope at entry. ❖ A pad must sound warm and full, which means we’ll have to get rid of a lot of overtones: Enter the filter. ❖ Last but not least, we’ll program a soft curve using the volume envelope. Wider, fuller … 1 Turn “Osc 2” all the way up in the Mixer module and set its “Detune” value to 0.07. We already have a noticeably thicker and more stable sound. 2 Take a trip to the toolbar and set the UNISON value to 2. Now each note is played by two voices; the sound is doubled. 3 Select Instrument > Properties in the menu bar. In this dialog, you can determine the amount of detuning between the unison voices: Set “Unison Spread” to 0.1. Now each note is played by four oscillators; these are nicely detuned against each other, providing an extra-wide sound. Brass Attack In real life, every wind instrument needs a couple milliseconds before the tone stabilizes, causing the pitch of the tone to “swim around” slightly. If you recreate this phenomenon using a pitch envelope, the ear immediately recognizes due to its conditioning: “Aha! Gabriel’s blowing his horn!” (or whatever). Since we’re not dealing with a solo instrument, but rather with a section sound, we only want some of the participating oscillators to simulate this attack phase—that way we get some nice friction within the sound. We’ll program a pitch envelope for Oscillator 2—to this end, the SoundForum Synthesizer allows us to misuse the filter envelope.

164

Mission: Possible—Exercises in Goal-oriented Sound Design

1 Active the “P” button in the lower (Oscillator 2) row of the “FiltEnv→Osc” module. Right now it sounds horrible, for two reasons: The variation is too strong and, since the sustain level in the filter envelope is not at zero, the pitch lands somewhere in the middle of nowhere. We’re going to change that right now: 2 Set “S(ustain)” in the “Filter Env” module to 0. Still doesn’t sound quite like brass, but we now only have to soften the attack of the filter envelope a bit to get there: 3 Set “A(ttack)” to 20 and “D(ecay)” to 40. Now we’ll bring the still-too-strong detuning down a notch: 4 Set “Amount” in the “Filt-Env→Osc” module to 0.02. Done: Now the tone swims just right at attack. Remember: At this point I’m still walking you through each step. Once you’re standing on your own two feet, you’ll be able to handle this fine-tuning yourself by continuously tweaking the appropriate parameters. Hearty Lip-smacking The acoustic image is still to aggressive and edgy for a pad sound. We’ll chisel away at that with the filter: 1 Lower the cutoff value in the Filter module to 70 and set “Env” to 30. Now it sounds a good deal softer; the filter follows the filter envelope, simulating the typical “lip-smacking” of brass instruments. Brass sounds don’t call for such a harsh (e.g. steep) filter, though, so: 2 Set the filter mode to LP2 and “Reson(ance)” to 0.4. With that, we’ve “flattened” the filter from 24 dB/octave to 12 dB/octave and let more “silk” through. The resonance emphasizes the mids around the cutoff frequency, giving the sound more body.

165

8 Advanced Hands-on Course

Your Pad or Mine? Since we’re going for a pad sound, we want it to fade in and out gently. That’s an easy one: ◆

Set “A(ttack)” to 35 and “R(elease)” to 55 in the “Amp Env” module.

That’s it—time to store your new sound with a click on the camera symbol. Your Assignments 1 Using this sound, program a much fatter, undulating sweep. For a sweep, the filter has to slowly open and close with a relatively high resonance value. 2 Use the brass pad to create a basis for solo brass sounds and, out of this, approximations of the following instruments: Clarinet, oboe and trumpet. Tip: Try the band-pass filter for woodwinds. 3 How do you make a “techno-compatible” hook-line sound out of this sound? Give it a try. Here are a couple of clues: No filter curve, but high resonance value with a high cutoff frequency. Anything that generates modulation should be exaggerated. Mission 7: Pitched Reverb out of Basic String Any synthesizer can generate reverb—not the reverberation of an external audio signal, of course, but rather a type of “pitched” reverb that can be combined with a direct sound. The nice thing is that this reverb can be tweaked at will using the synthesizer’s parameters. We’re going to build such a sound out of the Basic String preset. Again, the planning comes first: ❖ More ambience means less direct sound, so we need to make the basic oscillator sound more diffused and softer. ❖ The filter is responsible for the ambience characteristic; therefore it can neither allow too many overtones to pass nor be modulated via envelope or key tracking. We also need it to generate a mildly cool impression via attenuation of low frequency content. ❖ To create a longer reverb, we must lengthen the decay.

166

Mission: Possible—Exercises in Goal-oriented Sound Design

Diffusion 1 Instead of pulse waves, as in the preset, we’re going to use triangle waves. Set both oscillators to “Tri” and the “Sym(m)” controls to 0.3. This already gives us a much softer basic sound. 2 Diffusion can also be generated via friction between pitches: Intensify this by increasing “Amount” in the LFO module to 0.17. Since the LFO is only affecting one oscillator, this creates an extremely wide, diffuse modulation. Shaping the Room In this case, we don’t want the filter to generate a tonal curve, but rather— similar to the oral cavity—lend a fixed character to the sound. 1 Set both “Env” and “K-Track” in the Filter module to 0. Now the filter remains stable. 2 Set the filter to BP4, “Cutoff” to 75 and “Reson(ance)” to 0.6. With the 4-pole band-pass, the filter now only allows a “cave-like” frequency range around the cutoff frequency to pass, while the resonance additionally emphasizes these frequencies. Reverb Time ◆

Set “R(elease)” in the “Amp Env” module to 60.

Done. But this is only the basis for a plentitude of room-like sounds that you will now program on your own. Your Assignments 1 Modify the sound using noise so that it sounds like a sonar. 2 Now change the sonar so that it is suited for use as reverb for electronic drums (less pitch). 3 Combine the sound with direct signals from other synthesizers and adapt the reverb to each respective direct signal.

167

8 Advanced Hands-on Course

4 Find out how to simulate reverb parameters like pre-delay, early reflections, decay time, room size or high-frequency damping with the synthesizer. Insight In this chapter you have seen—perhaps without being totally aware of it— that it is possible to realize an imagined sound. After all, that’s exactly what we’ve done: Imagine a complex sound sculpture and then chisel it out of a basic patch. One could say that sound designers have to work like computer scientists: The solution lies in dissecting the whole problem into several smaller problems that are easier to solve. That’s what we’ve practiced in this chapter. We analyzed each modification first and then carried it out step by step. Planning and executing the individual steps may still be a bit difficult for you—you’ll likely need a bit more practice first. What’s important is that you stick to the following steps at first. I’ll leave it up to you whether you write them down or memorize them: 1 Imagine the sound you want to create in as much detail as possible. 2 Thanks to our comprehensive introduction to the “sound perception levels” you should have no problem dissecting your imagined sound into the characteristics dimension, temperature, texture, animation and aggression. Use these characteristics to define the sound. 3 Find a snapshot in the SoundForum Synthesizer that most closely approximates this definition. 4 Next, plan the steps you’ll need to take to modify this patch in the direction of your imagined sound. 5 Start with the basic sound—just like we did in the previous exercises: Oscillator waveform and mix, detuning, pitch modulation. Continue with the filter and finish up with the volume curve.

168

Mission: Possible—Exercises in Goal-oriented Sound Design

Outlook Since synthetic sounds are often based on acoustic instruments or natural sounds, there will always be certain “building blocks” that repeatedly appear. You should actually be able to load these into each module of any synthesizer; as a matter of fact, you can find this type of envelope or oscillator presets in several sound editors. Every sound designer should carry some sort of “preset collection” for basic sounds, modulations, filter settings and amplitude envelopes. Even the wildest sounds are often nothing more than new combinations of preset building blocks with a little fine-tuning. If you always need to spend a half-hour contemplating how to create basic sounds like fat basses, silky strings or powerful leads before you get started, sound design just isn’t much fun—and takes way too much time. In the next chapter, I’d like to introduce you to exactly this kind of repertoire—a plentitude of sound building blocks, out of which you can create finished sounds with the SoundForum Synthesizer.

169

9 Puzzle Programming In plain English, puzzle programming simply means programming sounds out of “sound building blocks.” By combining such puzzle pieces you will at some point be able to create hundreds of synthesizer sounds with your eyes closed. After concluding the advanced course with the last chapter, we will now enjoy sort of a break for our intellectual capacity. So far we have become acquainted with several ways to transform an imagined sound into a synthesizer sound: ❖ Dissecting the imagined sound into the five levels dimension, temperature, animation, texture and aggression, and building the sound according to these characteristics. ❖ Modifying complex sounds in the desired direction. ❖ Developing basic patches into complex sounds. That’s what we did with the basic SoundForum Synthesizer snapshots. Now for an additional method:

Completing Puzzles We can pretty much assume that there are many synthesizer owners who have a difficult time programming a sound from scratch, but who have a knack for fitting puzzle pieces together. In the end, anything should be allowed that leads one to one’s goal.

171

9 Puzzle Programming

As experienced sound designer, the question always arises as to why one can immediately program any synthesizer set in front of him, seeing as how FM is so difficult and the Nord Lead has completely different parameters than a Minimoog. One reason for this certainly lies in the fact that in time a significant number of puzzle pieces is collected in the subconscious, from which one at some point simply takes what one needs. This also explains, by the way, why seasoned sound designers are able to program the same sounds on any synthesizer. We’ll begin with the basic tonal colors, starting with the fundamental ones: the basses. After that we’ll concentrate on polyphonic sounds. Then we’ll finish off with curves and modulations. Upon conclusion of the puzzle programming section, you’ll be able to prepare a treasure chest full of puzzle pieces for every synthesizer and to expand the personal sound collection stored in your subconscious. For this chapter we have a new Ensemble entitled puzzle.ens, which you might as well load right now.

Tonal Color and Tonal Image Don’t worry: We’re not going to start with didactic dissertations on sound and vibration. We just want to deepen our understanding of tonal color and tonal image. Among many others, one possible definition of a sound is that tonal color represents a snapshot of a particular sound; a combination of overtones at a particular point in time. Not until you add time-based progressions do you have a tonal image.

172

Basic Tonal Colors

Basic Tonal Colors The SoundForum Synthesizer snapshots that I’ll be presenting here are basic patches to which elements such as modulation, filter curves and volume curves can be added. Only then do you get a complete sound. The basic tonal colors are completely without animation. It’s your job to make finished sounds out of them, and that’s the focus of this part of the SoundForum course. Under the heading “Modifications” I have described the most important steps in programming a finished sound, although you should actually be quite familiar with these by now. What’s Missing in the Tonal Colors? In the snapshots for this chapter, everything is left out that isn’t part of the basic tonal color. “Cutoff” is set to the value that the sound should have with minimal envelope/modulation. You can activate the envelopes later via the “Env” parameter. Right now, of course, the control is at a neutral setting, or zero. Modulations are also deactivated, as long as they are not necessary for the basic tonal color (e.g. beating). In other words: What you hear when you load these snapshots is not exactly exciting, but rather a darkish, sterile basic sound—the wood out of which you have the honor to carve a masterpiece of the century. Just so there are no misunderstandings: The snapshots I will be presenting during the puzzle chapter are merely meant to be suggestions and stimulations. They do not represent a complete library—otherwise there’d be nothing left for you to learn or do. Your own motivation is much more important than my descriptions. Or, to cite a stupid example: If, during a photography course, I teach you how to photograph a house, you will be able to use the same techniques to photograph a tree or an automobile.

173

9 Puzzle Programming

Which Controls Do What? You may be asking yourself which control elements you have to tweak to make a dynamic, finished sound out of a boring basic tonal color. The table below should be of some assistance: Module

Control

Effect

Filter Env

All

Shape of the tonal color curve

Filter

Env

Intensity of the tonal color curve

Amp Env

All

Shape of the amplitude (volume) curve

LFO

Amount

Intensity of the modulation

Filt-Env→Osc

Amount

Intensity of the effect of the tonal color curve or filter envelope (Filter Env) on pitch, waveform or volume of an oscillator

Mixer

Osc 1, Osc 2

Volume balance between the oscillators or waveforms

Oscillator 1, 2

Waveform, Sym

Basic waveform of the oscillator

Oscillator 2

Detune

Modulation (undulation)

Oscillator 1, 2

Interval

Sound spectrum (octave, pitch)

Oscillator 1

FM

Number and volume of the FM-generated overtones (with FM-based sounds)

Mixer

RingMod

Portion of ring modulation in the sound spectrum

It might be a good call to copy this table and hang it over your bed, since it applies to all sounds in the SoundForum Synthesizer, as well as in any other synthesizer. Sub-bass Let’s start with a relatively easy example. The term sub-bass refers to a ultra-low bass sound—sort of a “license to exist” for subwoofers. Often, a sine wave will be added as sub-bass to an existing bass sound, since overtone-rich basses rarely have enough low-end punch.

174

Basic Tonal Colors

Our example combines both: Oscillator 1 generates a growling basic sound under which Oscillator 2 delivers a sine wave one octave below. Simple but effective. Modifications ❖ Basic tonal color: Using “Cutoff” in the Filter module and “Symm” in Oscillator 1, you can determine how much “color” the sound should have. The mix between Osc 1 and Osc 2 controls the low-frequency content better than any EQ could. ❖ Tonal curve: Turn up “Env” in the Filter module until the sound is bright enough. ❖ The “D(ecay)” parameter in the “Filter Env” module defines the length of the “pop.” Octave Square Bass Consisting of two square waves an octave apart, this hollow bass sound with extreme cutting power is very popular for everything from eighties to technoid styles. A light filter resonance sharpens the edges, making the sound more aggressive. Otherwise, there’s not much to it. Modifications ❖ Basic tonal color: You can vary the aggression via the mix between Osc 1 and Osc 2. ❖ Tonal curve: Turn up “Env” in the Filter module until the sound “whips” to your liking. ❖ The “D(ecay)” parameter in the “Filter Env” module determines the length of the filter pop. ❖ More or less “Reson(ance)” (Filter) gives the sound more or less of an electronic character. ❖ Cut the pulse width in half and you’ve got a seventies disco bass.

175

9 Puzzle Programming

TB Bass Technically speaking, the good old TB-303—one of the world’s most popular groin stimulants—is about as complex as an egg poacher, making its sound fairly easy to simulate on any synthesizer. Basically all you need is the following mixture of basic ingredients: ❖ The oscillator generates either a sawtooth or a square wave. You can choose between the two by turning up either Osc 1 or Osc 2 in the Mixer module of the corresponding snapshot. ❖ The filter should actually be an 18-dB low-pass; we’re going to use LP2, a 12-dB filter: The flatter, the fartier. ❖ The filter resonance should be turned up really high and the filter should “sweep,” e.g. be “whipped” by the envelope. Modifications ❖ Program a simple filter envelope and turn up “Env” in the Filter module as far as you can stand it. ❖ For more obscure tonal colors, set the filter to high-pass or band-pass. ❖ Insert some kind of distortion or overdrive device into the signal path for maximum credibility. Black Bass The politically correct will please forgive me for not naming this snapshot “Bass Sound Inspired By American Citizens Of African Heritage:” The short form simply fits better in the snapshot menu. This bass sound is actually a staple in every basic collection. Think of the seventies, the Bronx, boom boxes and overfilled trashcans … that’s where the sound comes from. The most crucial element of the Black Bass is the two-octave spread between the two oscillators. We’ll use a pulse wave for the lower oscillator because of its punch, and a clearly defined sawtooth for the upper oscillator. Basically, that already just about covers it.

176

Basic Tonal Colors

Modifications ❖ Basic tonal color: Oscillator 1’s waveform determines the cutting power in the mix. ❖ Tonal curve: Turn up “Env” in the Filter module. This kind of bass sounds most authentic with an extremely short filter decay (D) time. ❖ More or less “Reson(ance)” (Filter) gives the sound more or less of an electronic character. ❖ For robot soundtracks and nasty techno, try adding a vibrato to Oscillator 1. With this snapshot, a simple click on “P 1” in the LFO module will achieve this. Sync Bass Oscillator sync generates an overtone-rich, “sawing” bass sound with incredible cutting power. Since the synchronized oscillator (Osc 2) gains overtones but loses low end, we add Oscillator 1 to the mix. Both oscillators are set to “Puls” (pulse wave), since that gives us the wiriest sound. Oscillator 2 is tuned to a fifth above Oscillator 1; the desired metallic sync overtones occur only above this point. The pitch envelope must be modulated for the oscillator to reach this point. Modifications ❖ Saw blade: You determine the “sharpness” of the sound via the mix between Osc 1 (fundamental) and Osc 2 (sync/metal). ❖ Tonal curve: Turn up “Env” in the Filter module until the sound has enough bite. ❖ Sync curve: The actual sync effect comes alive when you activate the “P” button for Osc 2 (bottom row; “P 2”) in the “Filt-Env→Osc” module. ❖ Alternately, you can transpose Oscillator 2 a bit higher and generate a flanging effect with a slow LFO. The LFO is prepared for this in the snapshot; you just have to activate the “P 2” button in the LFO module.

177

9 Puzzle Programming

Fat PWM Bass As a reminder: PWM stands for Pulse Width Modulation, a technique for making a sound fatter, wider and undulating via quick modulation of the pulse width. This bass sound makes use of this effect. Oscillator 1 and 2 are set almost identically. Pulse width is set to a low value, since the modulations through the filter envelope or the LFO have an additive effect. To emphasize depth and dimension, the oscillators are tuned an octave apart and detuned against each other for animation. So much for the basic sound. Modifications ❖ Basic tonal color: You can remove the octave spread to achieve a more defined and less deep sound. However, you should not modify the mix or the waveforms. ❖ Tonal curve: The “Env” control in the Filter module adds a curve that is programmed as a slow sweep. ❖ Modulation 1: You can let the pulse width follow the filter envelope— the effect is similar to a flanger: Simply activate both “Sy(m)” buttons in the “Filt-Env→Osc” module. ❖ Modulation 2: The pulse width modulation can, either alternatively or additionally, be generated by the LFO; in this case it’s not a curve, but rather a sort of rotation. To this end, activate both “Sy(m)” buttons in the LFO module. The only way to get the sound any fatter with the SoundForum Synthesizer is by using Unison mode. E-Pianos Analog synthesizers display definite limitations when it comes to simulating electric piano sounds. Thanks to its simple FM capabilities, however, the SoundForum Synthesizer has the potential to create fully usable—if not entirely authentic—simulations, from DX-Rhodes to Wurlitzer. Sine waves form the basis here, to which the typical bell-like spectrum is added via FM.

178

Basic Tonal Colors

Modifications ❖ The basic tonal color is defined above all with Oscillator 1 by adapting the “Interval” and “FM” parameters to each other. Setting “Interval” to 12 and “FM” to 4000, for example, will result in a Wurlitzer sound. ❖ The intensity of the tonal color curve is defined via the filter (“Env”) and the amp control for Oscillator 1 (“A 1”) in the “Filt-Env→Osc” module. Plucked Strings This label denominates just about any type of plucked strings, from harp to clavinet to electric guitar—in other words, everything wiry. Another case for the FM feature in the SoundForum Synthesizer. The basic tonal color in this snapshot is based on a nasal pulse wave from Oscillator 1 (“Symm” at 0.5 or higher), with FM from Oscillator 2 supplying the “wire.” Modifications ❖ The basic tonal color is determined by the pulse width (Oscillator 1; “Symm”), between harp and cembalo. ❖ You can mix freely between a round (Osc 1 only) and a nasal (Osc 2 only) basic sound in the Mixer module. ❖ The tonal color of the “wire” portion can be varied using the “FM” parameter and/or the “Interval” control in Oscillator 2 (since the oscillators are synced, you can’t do anything “wrong” or non-harmonic here). ❖ For a more distinct “bite,” add a filter envelope by turning up the “Env” knob in the Filter module. Bells/Mallets Bell-like and metallic sounds are created by non-harmonic overtones, best realized with an “analog” synthesizer by way of ring modulation.

179

9 Puzzle Programming

Modifications ❖ Overtones vs. fundamental: The balance between “Osc 2” and “RingMod” in the Mixer module determines whether the sound is round or more metallic. ❖ Type of metal: The “Interval” parameter for Oscillator 1 defines the basic tonal color. It’s important to note that harmonic values will not generate metallic sounds; you need “crooked” or non-harmonic values. ❖ Strike dynamics: The “FM” parameter in Oscillator 1 controls the amount of overtones, which determines the brightness or “sharpness” of the sound. Again, the intensity of the tonal curve is controlled by the “Env” parameter in the Filter module. String Ensemble This basic tonal color already contains a “built-in modulation;” the sound of real strings feeds on the undulations and differences in tonal color between the various instruments. The analog waveform that most closely approximates the basic tonal color of a violin or a cello is pulse wave (with “Symm” at 0.5 or higher). Both oscillators are needed here. The impression of an ensemble is created via detuning and modulation of the pulse width (tonal color). Modifications ❖ The basic tonal color can be varied using the “Symm” parameter of both oscillators. ❖ Use the UNISON mode (at the right of the toolbar) to vary the perceived number of instruments—you’ll want to use a value higher than 2. ❖ The “ensemble factor” is generated by the LFO via modulation of the pulse width. “Rate” and “Amount” (LFO) can be used to vary the discrepancy between the instruments. ❖ To achieve the typical bowing sound, try additionally modulating the pulse width via the filter envelope. To this end, increase the “Amount” value in the “Filt-Env→Osc” module and the “Env” value in the Filter module.

180

Basic Tonal Colors

Brass Compared to metallic or voice-like sound, brass sounds are relatively easy to program on an analog synthesizer. You need two sawtooth waves and a little pitch modulation. That’s it. Using this example, you can program virtually any brass sound in solo and ensemble versions, from trumpet to tuba. Modifications ❖ The balance between Osc 1 and Osc 2 in the Mixer determines whether you get a solo or section sound. For solo you should only use Oscillator 2. ❖ The curve of the attack is controlled by the “A(ttack)” and “D(ecay)” parameters of the filter envelope (“Filter Env”). ❖ The “Amount” control in the “Filt-Env→Osc” module determines the intensity of the pitch variation in the attack phase. ❖ As always, the intensity of the tonal color curve is controlled by “Env” in the Filter module. ❖ To vary between horn and trumpet sounds, use different “Cutoff” and “Reson(ance)” settings (Filter module). Horns generally need higher resonance and lower cutoff values; in addition, you should use a bandpass instead of a low-pass (BP4 or BP2). Voices Simulating the tonal color of a human voice with the SoundForum Synthesizer is only possible through a complex combination of FM, ring modulation and band-pass filter. Using this combination correctly, however, will allow you to program even a solo voice with shaping possibilities you wouldn’t expect from an analog system.

181

9 Puzzle Programming

Interestingly enough, the formants remain constant over the entire keyboard, in spite of the transposition, just as the vocal chords don’t grow or shrink with changing pitch. Better not to ask yourself (or me) exactly how this works—just be glad it does, and learn how to achieve and modify it yourself. Modifications ❖ The “Symm” parameter in Oscillator 1 determines the vowel sound. ❖ “FM” in Oscillator 1 and “Interval” in Oscillator 2 serve to vary the color of the voice (you can only use octaves or fifths as intervals). ❖ You can achieve more depth by increasing Osc 1’s level in the Mixer module or by lowering the cutoff frequency of the band-pass filter. ❖ To simulate vowel curves, turn up “Amount” in the “Filter-Env→Osc” module, allowing the filter envelope to automate Oscillator 1’s “Symm” parameter. The corresponding LFO parameter generates more humorous effects. Choir Don’t expect a substitute for a church choir that’s stuck in traffic here, either. Vocal ensemble sounds that not only fulfill the function of a choir, but can also be used as pseudo-reverb under various instrument sounds, can however be achieved fairly easily with the SoundForum Synthesizer. Again, the basic tonal color is generated via FM, whereas the Unison mode determines the number of choir members. ❖ The “Interval” relationship between Oscillator 1 and Oscillator 2, as well as the “FM” value, determine whether it’s a male, female or mixed choir—or something entirely different. A male choir is achieved with −12:0 and FM = 2850, while 0:24 and FM = 1650 give you a female choir. ❖ The size of the choir is determined by the UNISON setting (right side of the toolbar). ❖ Adjust the balance of the basic sound with “Cutoff” and “Reson(ance)” in the Filter module. Care must be taken here to find the right compromise between dull and squawking.

182

Basic Tonal Colors

Breath This term covers almost any kind of tonal color created by air movement that generates a harmonic vibration. Examples reach from flute (more tone, less air) to a blown bottle (more air, less tone) to wind in a cave (lots of air, hardly any tone). Three components are required: Tone, noise and formants. Oscillator 2 generates the tone, the noise is supplied by—you guessed it—the Nois(e) waveform in Oscillator 1, and the filter’s cutoff and resonance are responsible for the shape of the formants. ❖ The tonal portion is determined via the “Osc 2” control in the Mixer module, whereas its color is defined by the waveform. ❖ You can add noise in two ways: The “Osc 1” knob in the Mixer module adds noise directly. A more subtle method is via the “FM” parameter in Oscillator 1, which serves to modulate the pitch of the tonal portion of the noise—a darker, more natural effect, especially with solo instruments, since it causes the tone to seem more “raw.” ❖ Tonal shaping is done with “Cutoff” (darker-brighter) “Reson(ance)” (rounder-more nasal) in the Filter module.

and

❖ The entry or attack phase is shaped by the filter envelope (“Filter Env”); its intensity can be adjusted via the “Env” parameter in the Filter module. ❖ Since the “Amp” button in the “Filter-Env→Osc” module is activated, the noise portion always follows the filter envelope.

183

9 Puzzle Programming

Envelopes So far we’ve focused on static basic tonal colors; now we’ll expand to include programmable time-based progressions or “envelopes,” which represent a sort of “cover” that is slipped over the static tonal curve. The synthesizer modules that generate envelopes are called, quite logically, envelope generators. The SoundForum Synthesizer features two of these: One for the filter (“Filter Env”) and one for the amplitude or volume (“Amp Env”). With the help of the somewhat cryptically labeled “Filt-Env→Osc” module, the filter envelope can be routed to the oscillators to influence pitch, pulse width or the volume of a single oscillator. There are 17 special snapshots for this section, which contain envelope presets relating to common and recognizable instruments. To make it easy to discriminate these snapshots from the others, their names all begin with the contraction ENV. Each envelope snapshot is a combination of: ❖ a filter envelope (“Filter Env”), ❖ an amplitude envelope (“Amp Env”) and ❖ mild adaptations of the filter (“Cutoff,” “Env”). With a few exceptions, the basic sound is always the same and is rarely suited for anything in particular, but that’s not really important: We’re only concerned with the envelopes. Go through the snapshots once quickly, playing them as appropriately as possible: The Piano snapshot like a piano, the Solo Brass snapshot like a trumpet solo and so on. You’ll notice that the sounds—even though the oscillator signal is completely wrong—approximate the original fairly closely due to the appropriate envelopes; they are even pretty playable.

184

Envelopes

One very important factor is that the filter and amplitude envelopes aren’t each doing their own thing with no regard to the other; rather it’s their interaction that “makes” the sound. The Four-phase Model Before we look at the examples, I’d like to introduce you to a thought pattern that intuitively helps me program envelopes. The diagram below is a schematic representation. Have no fear: It looks far more technical than it is. → Note On Tone entry phase Attack

Note Off← Initial receding phase Hold phase

“Initial Decay” Decay

Sustain

Final receding phase Release

This is how an envelope works: In the upper row note on and note off, the curve phases in the middle, and the ADSR parameters at the bottom.

To clarify: The starting point, length and end of an envelope are determined by two events: Note on (pressing the key) and note off (releasing the key). As the diagram shows, three envelope phases take place while the key is depressed: ❖ Tone entry phase: Blowing, bowing, striking. ❖ Initial receding phase: Generally gentle, gradual decay after the attack. ❖ Hold phase: As long as the key is held, the curve remains constant at some point. ❖ Final receding phase: Once the key is released, the final receding phase begins, which—at least in the case of the amplitude envelope—has to eventually land at the null point.

185

9 Puzzle Programming

ADSR—with a Handicap I personally despise the classic ADSR envelope, which is found in the vast majority of synthesizers, for the simple reason that one phase is missing: An initial decay time for the entry phase. Why? Imagine a vibraphone. The mallet strikes the metal bar—there’s an initial click, followed by a slow decay. Logically speaking, there is a short and a long decay time—the ADSR envelope, however, has only one. This is, in my humble opinion, a serious and fundamental deficit. Nonetheless, ADSR has survived and been carried over into even the best and newest synthesizers. Now, I’m not telling you this simply to get it off my chest, but because it’s important that you’re aware of this problem—and of the solution. There is a way to at least halfway simulate this second decay phase: By using different decay times for the amplitude and filter envelopes. As you can see in the diagram above, the decay cell is split into “Initial decay” and “Decay.” This represents the different decay parameters of the amplitude and filter envelopes. The Examples I know you’ve heard it before, but I have to repeat it: The examples won’t do you much good if you copy them and transfer them 1:1 to your sounds. Play with the sounds, modify the envelope parameters and pay attention to what happens, to get a feeling for how the envelopes work and what effect they have on the sound. Piano To make it clear at the outset: This preset is not called Piano because we want to simulate a piano with it, but rather because the preset’s tonal and volume curves are fashioned after those of a piano (and are thus generally associated with the instrument).

186

Envelopes

With a piano, the tone enters abruptly when a key is struck and decays slowly as long as the key is held. When the key is released, the tone does not stop abruptly, but has a minimal decay time (damper effect). Many programmers ignore the decay time, by the way, which is why so many piano sounds are somewhat unrealistic. Electric Piano Here the overall sound is somewhat darker than that of the piano. There is a click at the attack phase (hammer on tone bar), which we simulate with the decay time in the filter envelope (our initial decay). The amplitude envelope is very similar to that of the piano. Harpsichord This is a very quickly decaying instrument, which explains the extremely short decay time in the filter envelope. The rebound of the plectrum is simulated with a trick using the filter envelope: This envelope can be routed to the filter with negative polarity (“Env” in the Filter module set to a negative value). In this case, the attack parameter simulates the decay, and sustain must be set to the highest value (see what happens when you reverse this). The zero release time in the filter envelope causes the envelope to close immediately and cutoff to snap to 100%. Vibraphone The vibraphone needs no special control over the length of the note: You strike the metal bar with the mallet, and the sound immediately begins to decay—not unlike a cymbal or a triangle. This explains why the decay and release times for both envelopes are identical and the sustain times are set to zero. The result: No matter how long you keep the key depressed, the tonal and volume curves are always the same.

187

9 Puzzle Programming

Xylophone Basically the same as with the vibraphone, except that here the curve is much shorter and wooden. Bells Also basically identical to the vibraphone, but with a very slow decay of both tonal color and volume. Perc(ussive) Organ With an organ, there is generally no volume curve—there is simply on (key pressed) and off (key released). The percussion of an organ is simulated via the filter envelope: A short decay time causes the filter to generate a click at the attack phase. Pipe Organ First of all, the entry phase (blowing) of the organ pipes is simulated via a short attack time in the amplitude envelope, combined with a contrary decay time in the filter envelope. Then, a reverb effect as a gimmick: A slow decay in the amplitude envelope combined with an abrupt damping of the tonal color when the key is released (R = 0 in the “Filter Env” module). Acoustic Guitar The entry phase in the filter simulates the plucking noise made by the fingernails (short decay time), while the volume decays fairly rapidly. Electric Guitar Here, the amplitude envelope creates a compressor effect (brief click at attack); the actual decay is generated—due to the lack of a more complex envelope—entirely via decay of the tonal color (decay at cutoff).

188

Envelopes

Acoustic Bass The tonal color decays very quickly; the volume only slightly less quickly. This creates the typical plucking sound, which is additionally emphasized by the mild pitch modulation via the filter envelope (see the “FiltEnv→Osc” module). Slap Bass This is a combination of volume compression (see Electric Guitar) and rebound (see Harpsichord) with a slightly darker tonal color. Synth Bass The synth bass is one of the easier sounds: A simple, quick decay of the tonal color (filter envelope) combined with an organ-style amplitude envelope. The decay phase should smack somewhat (short release time in the filter envelope). Strings/Ensemble/Choir This is also a very simple envelope: Soft fade in and out of the volume (amplitude), combined with a brief bowing sound, generated by way of contrary amplitude entry and filter decay phases. Solo Brass Here, the most important element is the slight bend of the tonal color that influences the pitch (“Filt-Env→Osc” module). For trumpet or trombone you only need a sawtooth as a basic waveform and you’re done. Brass Section An analog simulation of a brass section is dependent upon at least two oscillators, one of which generates a pitch bend at entry, possibly an octave apart from the other(s). Otherwise this envelope is basically identical to that of the Solo Brass.

189

9 Puzzle Programming

Synth Brass Whereas the amplitude envelope resembles that of an organ, the tonal color (filter envelope) is decisive. The filter is more or less “raised out of the cellar” by the envelope (“Cutoff” = 43, “Env” = 84). Attack and decay times should form a soft hump. Set the waveform to sawtooth, add a chorus effect, and the sound is actually ready to go.

LFO In conclusion we’re going to examine the most important methods of making a synthesizer sound fatter with the LFO, creating an entire string ensemble with only two oscillators, or goading screaming techno riffs out of something as harmless as the SoundForum Synthesizer. Who Is LFO? The LFO generates the “oscillating, undulating” modulations in a synthesizer—everyone is familiar with vibrato or tremolo. The rather bulky abbreviation LFO is short for the even bulkier “Low Frequency Oscillator”—an oscillator that generates vibrations at extremely low frequencies. The LFO signal is not an audio signal, but rather a control signal. The speed of the vibration (rate), the waveform (sine [Sin], triangle [Tri], pulse [Puls], sample & hold [S+H] and Symm) and the intensity (Amount) are freely programmable. The effect of the LFO on the sound, however, is equally dependent upon which characteristic or parameter it influences. The same LFO setting can, for example, create a vibrato (→ oscillator pitch), a chorus (→ oscillator pulse width or Symm), an auto-wah (cutoff) or a tremolo (amplitude). In order to activate the LFO in the SoundForum Synthesizer for a particular parameter, a group of buttons is provided. Before we get lost in dry, ambiguous theoretical possibilities, let’s get to the snapshots.

190

LFO

Snapshots As with the previous ones, each of the snapshots for this section represent a half-finished sound. In other words, not only the LFO modulations, but also the other sound parameters are pre-programmed. Once again, go through the snapshots quickly to get an idea of what the LFO can do with each of these sounds. The more experienced and/or adventurous among you should feel free to go ahead and experiment with the LFO settings a bit. While you’re going through the snapshots: ❖ vary all of the LFO parameters randomly and fearlessly to get a feel for the elements that make up a modulation and ❖ deactivate the LFO modulations by clicking on the buttons in the LFO module to hear the sound without the LFO. PITCH—Vibrato Let’s begin with the simplest assumable case: A triangle waveform modulates the pitch, resulting in a vibrato. The effect can be varied via “Rate” and “Amount.” Switch the waveform to “Sin”—you’ll notice that a sine wave has a milder effect than the smoother triangle modulation. PITCH—Laser Gun Of course we don’t really know how a laser gun really sounds—outside of C movies. Here we’ve simply taken the vibrato from the PITCH—Vibrato snapshot, intensified it and turned the “Symm” knob in the LFO module all the way to the left. The triangle wave is now a falling sawtooth—a sound familiar to all of us who spent any time in a gaming room in the early eighties.

191

9 Puzzle Programming

PITCH—Trill If you use a pulse wave instead of a triangle as the LFO waveform and switch the modulated pitch around, you get a trill. The relationship between the high and low notes can be adjusted using the “Symm” control. This example illustrates very nicely that the SoundForum Synthesizer’s LFO functions in a bi-polar fashion. The signal contains positive as well as negative peaks, which means that the modulated oscillator must be brought back into tuning via its Interval control. PITCH—Chorus Pitch modulations get really interesting when multiple oscillators sound simultaneously while being modulated differently. Since there is only one LFO in the SoundForum Synthesizer, we leave one oscillator un-modulated in this and in the following example. Oscillator 1, however, is slowly “moved” around its original pitch, creating a chorus effect. The faster the LFO rotates, the less intensity (“Amount”) is needed; a faster rate is perceived as a more intense modulation. Actually, the amount should be decreased when playing in lower registers and increased in higher registers. However, the SoundForum Synthesizer does not offer this function. On the other hand, it can perfectly emulate “old-generation” sounds: Try setting “Rate” to 12 and “Amount” to 0.14—sounds just like a good old electronic string ensemble, huh? PITCH—Flanging The typical sharp, cutting flanging effect is generated via the combination of sync and LFO modulation. Try deactivating the “Snc” button. To vary the flanging effect, simply change the “Interval” value for Oscillator 2. To intensify the “sawing” of the flanger, turn down Oscillator 1 in the Mixer module—now you only hear the “synced” oscillator.

192

LFO

PITCH—Random The speech of the good old R2 unit R2D2 of Star Wars fame is no secret to sound designers. His special brand of squawk box can be imitated very effectively using random pitch. The LFO’s S+H waveform works like this: A random value is “recorded” from each cycle (Sample) and held until the next cycle (Hold); thus the term “Sample & Hold” or “S+H” for short. Add FM to the mixture and it almost sounds like an angry R2 unit. Routed to the filter, the S+H wave creates a random tonal color; its use with amplitude is limited. PWM—Chorus Now we’ve made it to pulse width modulations (PWM). Back in the days when one oscillator per voice still represented a real, discrete collection of condensers and resistors, many synthesizers had to make do with a single oscillator to keep costs in check—and a chorus was not always included. In those days, PWM was often used to fatten up the sound—successfully, as the snapshot demonstrates. There’s only one oscillator at work here, and it nonetheless sounds like two. If you happen to own a synthesizer with two LFOs, recreate this patch on it and program a vibrato for the second oscillator with the second LFO. Talk about phat! PWM—Ensemble With a somewhat faster modulation, a second oscillator tuned at an octave and a bit of detuning, you get a very nice ensemble effect—with just one LFO. PWM—Trance For all of you who have always wondered how the techno freaks get those insanely cutting leads, to which all of those typical ponytailed, carb-andcommon-sense-free blondes bathe in laser and strobe light while dancing their calories away, here’s a round of wisdom on me: Turn off all good

193

9 Puzzle Programming

taste, and cutoff and resonance up to 3/4 … everything else is freely variable. In this example, the two oscillators are tuned an octave apart and PWM is too intense and too fast. Sounding hip is not really much of a challenge … FILT—AutoSweep And now to the filter modulations. The counterpart of the vibrato for the filter is a sweep, except that it rotates more slowly. The filter paints the friendly frequency band with relatively high resonance. Waveform: Triangle; sine wave if you want the ends to be more contoured. FILT—Repeat If you set the LFO waveform to sawtooth (select “Tri” and turn “Symm” all the way to the left), you get a repeat effect—the note seems to be played repeatedly at the LFO’s tempo. Sir Paul McCartney, by the way, used this very effect in a tear-jerking Christmas song—do you recognize the sound? FILT—AutoReverse Turning the “Symm” control all the way to the right causes the LFO to generate a rising sawtooth wave, which always sounds like a reverse effect— most LFOs offer this waveform as a preset. AMP—Tremolo Enter the first of three amplitude (volume) snapshots: Tremolo is actually nothing more than a volume vibrato—instead of the “P” button, the “A(mp)” button in the LFO module is activated. I went ahead and programmed the tremolo in a vibraphone-like patch.

194

LFO

AMP—AutoTrig In this patch, the synth bass seems to be played with multiple attacks, while the filter slowly opens and closes. What you otherwise would actually have to play with a Parkinson finger while slowly nursing the mod wheel is being done for you automatically by the LFO and the filter envelope. AMP—Echo If you use the trill effect (LFO with pulse wave) on the amplitude or volume, an echo effect is generated. The difference in length between “tone on” and “tone off” can be varied using the “Symm” control. If you set the LFO waveform to “Sin(e)” and turn the “Symm” knob all the way to the right, the echo sounds in reverse. Observing LFO Waveforms Yes, you can even observe the LFO waveforms on the oscilloscope. To this end, load bonus snapshot “LFO Wave View.” It generates white noise that appears at first as a relatively even surface in the oscilloscope. Set “Level” to 10:00 and “Time” to minimum in the oscilloscope. If you now vary the LFO’s waveforms and “Symm” parameter, you can observe the resulting LFO waveform in the oscilloscope as a volume curve.

195

10 10 Factory Bank Project The time for sowing is finally over—now we shall reap. Become a professional sound designer and participate in the creation of a factory patch bank for the SoundForum Synthesizer. As already announced, I assume at this point that you have acquired a solid basic knowledge of sound programming and the SoundForum Synthesizer. One of the ways to achieve this lay in actually working through the previous chapters of this book—at least for the most part. Be that as it may, beyond this point no more concessions shall be made for beginner’s questions; the descriptions of individual steps will also be appropriately brief. In this section we will be pursuing completion of a clearly defined task, as is daily routine for the professional sound designer: We have been assigned the task of programming a factory bank consisting of 20 patches that will guarantee maximum virtual retail success for the SoundForum Synthesizer. The task is defined this way because these patches should demonstrate the sound creation possibilities of the synthesizer as impressively as possible. They should deliberately aim for “aha!” reactions. So we really have to get creative here. All the while you’re in the enviable position of merely having to look over my shoulder and follow on your SoundForum Synthesizer. It won’t be all that easy, though: After all, you are supposed to understand what we’re doing.

197

10 Factory Bank Project

Example Files The “Init” snapshot that’s included in various ensemble sets on the CD will serve as basis for all of the sounds in this chapter. You’ll also find the individual snapshot on the CD as “init.ssf.” The sounds we’ll be using in this chapter are also included in the “factory.ens” Ensemble. Naturally, we’ll be programming all of the sounds from scratch. You can’t really lean on something created by someone else when you’re programming factory sounds—after all, we are creating original sounds for a brand new synthesizer here.

The Factory Bank Sounds 1: Butterfinger The SoundForum Synthesizer is an “analog” synthesizer. One of the first things a potential user is going to want to know is whether it delivers a healthy portion of punch and phat; we want to convince him right away. Our idea is a fat, cinema-style killer bass that flies in when a key is pressed and slips away at release. We call it “Butterfinger” because cleverly named sounds whet the appetite and Butterfinger it associated with something sweet, tasty and fattening. First we have to program a nice, fat basic sound: 1 Using the Init patch, we start by turning up both oscillators all the way, setting them to square wave and detuning them by 12 cents. Not thin— but not really fat either. 2 Next we’ll use the LFO to modulate the symmetry of both oscillators with an Amount of 0.95 and a Rate of −9. 3 To make the modulation a little less monotonous, we’ll adjust the symmetry in Oscillator 2 (Puls-Sym) to 0.04. Pretty nice basic sound, but since we only need a monophonic sound anyway, we might as well stack a couple voices. 4 A UNISON value of 5 ought to do it.

198

The Factory Bank Sounds

Now we’re talking phat! Just a little animation … After all, we’re not just going for fat—it’s supposed to sound spectacular. So we’re going to program a sweep not only for the filter—but also for the pitch, to get that “flying in and out” effect we were talking about. 5 First we’ll program a slow filter curve (A = 50, D irrelevant, S = 1, R = 50) and an appropriate amplitude envelope with a fast attack and slower release. 6 Then we set “Env” in the Filter module to 100. Nothing is happening yet, because “Cutoff” is set to over 100. 7 So we decrease the cutoff value far enough so that the filter doesn’t open fully until the envelope peak is reached; this happens at around cutoff = 60. 8 To add a little edge to the sweep, we turn “Reson” up to around 0.7. Getting there, but we still want the filter envelope to modulate the pitch at an octave. 9 To this end, we activate both “Osc” buttons in the “Filt-Env→Osc” module and set “Amount” to −0.96 (which equals approximately an octave). 10 For a little more cutting power, we turn the “FM” knob up to 200. I’d like to see someone beat this with a measly little Minimoog! 11 Now the final touch: since potential buyers in music stores like to play with the mod wheel, we route it to the FM control. Aha! Number 1 is a winner. 2: Baked Beans Now that our virtual potential buyer has heard something spectacular, he might like to hear something seriously usable. A fat, undulating pad is another “trademark” analog synthesizer sound and a good contrast to our first creation. So let’s program a nice, warm, fat, gooey, filling sound— kind of like the name implies.

199

10 Factory Bank Project

Dark, animated and mildly filter-sweeping pads seem to go over well among users, and these are easily realized with the SoundForum Synthesizer. Once again using the Init patch, we turn both oscillators up and detune the second one by 9 cents. Both should be set to sawtooth waveforms (in Oscillator 1, that’s “Tri” with the “Symm” knob turned all the way up). The good old unison trick will serve us here again as a thickening agent, but in this case we’ll leave it at two layered voices per note. For the pleasantly undulating modulation, we’ll program an LFO triangle vibrato for Oscillator 1 with a “Rate” of −4 and an “Amount” of 0.09. Now for the soft tonal color, for which we’ll once again activate LP4. A cutoff value of 73, combined with a resonance value of 0.31, creates a dark, moist tonal character. The already slow filter envelope (A, D and R = 70) needs only be enhanced via an “Env” value of 30, so that the sound opens up a bit. Then we might want to route the modulation wheel to the filter’s “Env” control to add some MIDI controllability. The amplitude envelope should fade in and out slowly. And we’re done with our easily programmed but very convincing pad. 3: Syncomator The bass was big and punchy, the pad sweet and silky—we’ve got our “fish” on the hook. Now we have to feed him enough acoustic energy to keep him going through the next ten sounds. How about a little shake, rattle and roll? A tonal curve (in this case sync) that gets chopped up by the LFO, creating constantly new harmonic patterns, should raise an eyebrow or two, don’t you think? To start out with, we only want to hear the synchronized oscillator, so we only turn up Oscillator 2. Sync is at its meanest in combination with square waves, to which we’ll set both oscillators.

200

The Factory Bank Sounds

We’ll generate the sync curve with a slowly decaying filter envelope (A = 0, D = 75, S = 0, R = 75), along with a hearty turn of the “Amount” knob in the “Filt-Env→Osc” module to 1 and a click on the pitch button for Osc 2 (“P 2”). Now let’s get chopping. As mentioned, that’s the LFO’s job, which we’ll route to Filt(er) and Amp(litude). To ensure that it chops, we’ll set “Symm” in the LFO module to −0.99, transforming the soft sine wave into a percussive repetition. That chops pretty well, but the filter isn’t right yet. Cutting filter curves are best realized with a bandpass. We’ll choose BP4, set cutoff to 60, resonance to 0.75 and Env to 90. That way, the filter also follows the sync curve a bit. The volume envelope (Amp Env) should decay slowly (A = 0, D = 70, S = 1, R = 70). 4: Metallurgy The first three sounds may not have been entirely commonplace, but they can be created with any analog synthesizer with a halfway respectable feature set. Our potential customer will want to know whether the synth is capable of producing “non-analog” sounds, and we’d like him to have an experience that leaves a solid impression—a sound that he’d never expect from the harmless-looking SoundForum Synthesizer. After all, we do have access to FM and ring modulation, making metallic sounds a very real and easy-torealize possibility. We’ve chosen the name “Metallurgy” to motivate us to create something metallic and morphing. In order to get the most out of the metallic FM sound, we start off by only turning up Oscillator 2. With the SoundForum Synthesizer, the metallic tonal color is a combination of three factors: The oscillators’ waveforms, their intervals and the value of the FM parameter in Oscillator 1. After a bit of experimentation, we end up with the following settings:

201

10 Factory Bank Project

Both oscillators generate a pulse wave with a symmetry value of 0.5. The intervals are set to 29 and 38 (Oscillators 1 and 2, respectively) and FM is set to 1450. Here again we utilize Unison mode with 4 voices per note. This generates a nice, fine “grinding” sound, as if you were sanding a church bell using a belt sander with fine paper. We’re going to emphasize that using a high-pass filter that lets the upper harmonics pass while attenuating the lower portions. We choose HP2 with “Cutoff” at 50 and “Reson(ance)” at 0.78. With “Env” at 100 and “Filter Env” at A = 0, D = 55, S = 0 and R = 55, the sound is enhanced with an additional “sanding.” Now we’ll program the LFO to add continuous spectral movement. It’s going to modulate the symmetry of Oscillator 1 using a triangle wave, a rate of −13.5 and an amount of 0.66. Now it sounds as if the belt sander was being moved and turned on the bell. The amplitude envelope should fade in and out softly (A = 40, D irrelevant, S = 1, R = 60). 5: Ciaoup After all that rattling, whistling and morphing our potential customer should be convinced that the SoundForum Synthesizer is capable of a thing or two. Now we want to present him with something solid—something he actually might use in a song. Since we have no way of knowing where his musical tastes lie on the scale from slow fox to warp trance, we’ll play it safe with a true classic. A very slender yet extremely low and contoured bass that underscores the synthesizer’s sonic quality. Using the Init patch, we’ll start working on the basic sound. We activate “S(y)nc” (slenderizes) and turn both oscillators all the way up. Oscillator 1 gets transposed by −24 half steps and Oscillator 2 by −26 (creates friction). We want a pulse wave in Oscillator 1, since it delivers more metal than a sawtooth. “Symm” is set to 0 (square wave). Now we’ll program the filter to generate a clear “chowoop” effect (that’s where the name comes from): “Cutoff” at 33, “Env” at 94 and “Reson(ance)” at 0.77.

202

The Factory Bank Sounds

The whip portion is supplied by the filter envelope (A = 11, D = 40, S = 0.3 and R = 23) and a slightly lengthened release time in the amplitude envelope. The latter creates the noticeable “snapping shut” at the end of the tone. We’ll create a bit more overtone development by activating pitch for Oscillator 2 (“P 2”) and setting “Amount” to 0.26 in the “Filt-Env→Osc” module. Here’s a trick for creating a little “fake beating” that doesn’t muddy up the low end like normal modulation tends to: The LFO slowly (“Rate” = −6) and gently (“Amount” = 0.5) modulates the pitch of Oscillator 2. Combined with oscillator sync, this sounds a little like mild flanging. 6: PPG 2002 Many roads lead to Rome. Similarly, an analog Synthesizer can deliver a convincing PPG-style sound using FM. Since the user more quickly digests something that has been pre-chewed via the name, we’ll call this patch “PPG 2002”—a PPG in a millennium cloak. Our only route to this sound is via FM; this is the only way to generate the typical gargling tonal colors that made the PPG so popular (and vice versa). To get this sound, we’ve got to experiment with oscillator combinations: First turn up the FM control, then tweak the interval and waveform parameters until you get the right basic sound. In this case we end up with “FM” = 1900, “Symm” = 0.14 and “Interval” = −12 for Oscillator 1. Then we add Oscillator 2 (the actual FM color) with “Puls-Sym” at 0.24 and “Interval” at 36. The typical PPG-style wavetable sweep (but even nicer) is realized via the “Filt-Env→Osc” module, where we set “Amount” to 0.5 and activate symmetry for Oscillator 1 (“Sy(m) 1”).

203

10 Factory Bank Project

Next we program the filter and filter envelope departments for a moderately percussive approach: “Cutoff” at 54 and “Reson(ance)” at 0.33 for a soft, dark basic color; with “Env” at 84 we add the filter envelope, which delivers the moderately percussive effect with A = 0, D = 56, S = 0 and R = 60. The amplitude envelope gets a similar treatment (A = 0, D = 20, S = 0.61 and R = 56). That’s it. No connoisseur would be offended if we mention that this sound beats the pants off of any PPG … 7: Logical Sound The electric piano is gradually replacing the piano as obligatory part of a factory bank. Not all synthesizer customers are 17-year old CD spinners who became acquainted with music over Fruity Loops. So we should try to support those among our potential customers who have had three years of piano lessons in the illusion that it was not all wasted time, which they had better spent with Fruity Loops. For these candidates, a nice, playable epiano sound is possibly the best medicine. Once again, FM is the basis here. But instead of an everyday Wurlitzer piano, we’re going to create a shimmering, expensive-sounding electric piano with lots of metal. We’ll start by setting FM to a healthy 800. Oscillator 2 is responsible for the metallic portion of the sound, so we turn it all the way up and set “Interval” to 60. Then we activate the “Amp” button for Oscillator 2 in the “Filt-Env→Osc” module so that the metal portion decays a bit more quickly than the basic sound (Oscillator 1). Now for some tonal color shaping: The filter gets a slightly more moderate treatment than with the PPG sound, with “Cutoff” = 71, “Reson(ance)” = 0.36 and “Env(elope)” = 58—we don’t want it to sweep, but simply dampen the sound somewhat in the decay phase. Since the SoundForum Synthesizer’s “Env” parameter is permanently routed to velocity, the piano is automatically velocity sensitive—the same goes for volume.

204

The Factory Bank Sounds

The filter envelope gets a medium-strong percussive treatment: A = 0, D = 55, S = 0 and R = 44; the amplitude envelope gets a slightly longer decay time (D = 64) and a shorter release time (R = 24). A small but important detail: Many programmers make the mistake of setting the release time to zero with piano sounds, which gives them an unrealistic feel that reminds me of cardboard. All we have left is a bit of finest-tuning: The LFO modulates Oscillator 1’s symmetry with “Rate” = 1.5, “Tri” as a waveform and an “Amount” of 0.44, which in this case creates a mild, pleasant flanging effect. 8: Fjord Lord After three “usable” sounds we should maybe wake up our imaginary test listener with something a little different. “Fjord Lord” is a whale-song-like background lead sound with a moderate overblowing effect in the attack phase. We turn up both oscillators, set FM to 1450 and Interval to 12 (Oscillator 1) and 24 (Oscillator 2) respectively. After setting the Symmetry control (“Symm”) to 0, our whale-song-like basic sound is finished. Tweak the FM knob a little to see what it does. The overblowing effect is, thanks to FM, very easy to achieve: One of the oscillators (Oscillator 1) is minimally modulated via the “Filt-Env→Osc” module (“P 1” button activated, “Amount” at 0.08). Now we program a filter envelope with a short entry: A = 35, D = 34, S = 0 and R = 77. “Cutoff” at 82 and “Reson(ance)” at 0.41 in the Filter module supply a dampened basic tonal color. We’ll cause the envelope to negatively affect the cutoff frequency with “Env” at −22; here, you can experiment with the setting and choose the one you like best—they all sound good. The LFO generates a slow vibrato. The “Tri” waveform is crucial here, as is a “Rate” of 11 or lower and a mild “Amount” setting of 0.13. The “P” button for both oscillators should be activated; you can, however, later deactivate one of them if you like—three different types of vibrato effect are possible. The amplitude envelope gets a gentle attack and slow release with A = 15, D = irrelevant, S = 1 and R = 58.

205

10 Factory Bank Project

9: Machine Head Any passionate test listener is going to want to bring the roof down—at least once. So let’s program a big, fat, mildly animated low-end power saw. First, we need an earth-shaking basic sound; once again we’re going to make use of FM. We’ll go ahead and turn it up to 3350, and turn up Oscillator 2 in the Mixer module so that we can hear the FM effect. To get a fat razor sound, Oscillator 2 should be transposed down a good deal; say to −26. The interaction of the two oscillators’ waveforms plays an important role with FM; I’ve chosen “Tri” (triangle) with “Symm” at 0.6 for Oscillator 1 and “Saw” (sawtooth) for Oscillator 2. The animation comes from the LFO set to a pulse wave with a short impulse. To this end, we set “Symm” to −0.6, “Rate” to 14 and “Amount” to 0.5 in the LFO module, and select two modulation destinations: The pitch of Oscillator 1 (“P 1”)—this modulates not only the basic pitch, but also the frequency relationship between the oscillators—and the filter (“Filt”). We want the filter to generate a slow, cutting sweep. “Cutoff” at 62 and “Reson(ance)” at maximum, combined with BP2 (bandpass), gives us the basic sound we’re looking for. Then we set “Env” at 96 and “K-Track” to 0 for a prominent sweep that remains constant over the keyboard. Finally, we program slow filter (A = 69, D = 74, S = 0, R = 65) and amplitude envelopes (A = 0, D = 69, S = 1, R = 62), with curves that cause the sound to build and diminish when a key is held. Done. 10: OBY 16 Even with contemporary sound banks you have to be careful to not only think of teenagers—there are still working keyboard players out there. We know our virtual client will sell the SoundForum Synthesizer to anyone who puts his money down, so we shouldn’t just chase trends that will be obsolete tomorrow, and be sure to toss in a true classic now and then. And what could be more classical for an “analog” synthesizer than the sound

206

The Factory Bank Sounds

whose middle name could be “Fat Fat Fat”—namely the typical Oberheim brass sound? Oddly enough, many of the people who made this sound famous could share its middle name—maybe that’s why it sounds so f… Never mind. We’ve touched on elements of this sound several times in the course of this book, so the ingredients should be familiar by now. The basic waveform for brass sounds is the fat, greasy, overtone-rich sawtooth wave. We’ll set both oscillators accordingly. Next we have to emulate the typical brass entry or attack phase. This is the job of the filter envelope, which we’ll route to only one oscillator in the “Filt-Env→Osc” module. The resulting friction in the attack phase tricks the ear into thinking it’s hearing horns. We’ll program a filter envelope with short attack and release phases. Sustain must be set to zero, otherwise Oscillator 2’s pitch will land somewhere in nirvana (try it out). To get a silkier sound, we switch the filter from LP4 to LP2, making it less steep and letting more “shine” through. With that, we’ve basically covered it—the rest is fine-tuning of the individual parameters; I encourage you to experiment with all of them and try to adapt the sound to your own personal taste. 11: FlexPad One simple trick is to present the user at regular intervals with a sound that he wouldn’t expect from that particular synthesizer because it seems to lie outside the possibilities of the implemented type of synthesis. If the sound also earns the rating “not completely useless,” it’s also a very effective trick. FlexPad is a pad sound with a complex, morphing overtone structure. Not too long ago, a surprising number of Yamaha fans with more money than common sense lightened their accounts by several thousand dollars for a brand-new VP-1 because of just such sounds. But hey, that’s OK: Otherwise they probably would’ve just spent it on something even more questionable …

207

10 Factory Bank Project

This sound is based on the SoundForum Synthesizer’s welcome ability to dramatically modify the sound spectrum of an FM sound via modulation of oscillator 1’s pulse width (Symm). This is exactly what the “Sy(m) 1” button in the “Filt-Env→Osc” module does when activated: The filter envelope automatically and invisibly tweaks the “Symm” knob in Oscillator 1. The basic sound can be widely varied using the FM control. Except for the above-described trick, the sound is a no-brainer: The undulation is generated by the UNISON mode (value: 3), whereas Oscillator 1 also gets a mild vibrato effect courtesy of the LFO. Determining the appropriate filter and amplitude envelopes shouldn’t really represent a challenge to any of us. To program your own special variation of the sound, I recommend tweaking the following parameters (next to FM, of course): Oscillator 1 Interval, Oscillator 2 Waveform and Oscillator 2 Interval. Good luck! 12: Chime On As we know, the potential customer often has something with him in the music store; something he generally refers to as “honey.” This “honey” is only predictable in one respect: It will be yawning in five minutes or less. It is of primary importance that we take this social component into consideration when programming factory sounds: “Honey” will inevitably exercise pressure on our test listener with the goal of getting him away from the synthesizer as quickly as possible and pulling him into the next perfumery, where our potential customer will end up spending his synthesizer budget on a collection of liquids that smell like they were made by filling helpless rodents with soap and then pressing them … This is to be avoided at all costs, which is why we need to appease “honey” now and then during the test listening phase. The wind-chime emulator we call “Chime On” fulfills precisely this task. Accompanied by the (predictable) sentence: “Listen, honey—just like the chimes that the Reiki teacher gave you,” our test listener glisses down the keyboard. The resulting “Oh! It can do that?…” signals say that we’ve been granted another five minutes to hook our guy.

208

The Factory Bank Sounds

This essentially important sound is very easily programmed: Two oscillators are set into frequency modulation via the FM parameter; their intervals are set very high and tuned slightly apart. The trick is tweaking the relationship between the FM and the two Interval controls so that a windchime sound results. The bandpass filter ensures that only a small portion of the sound spectrum is audible, while the envelopes supply the appropriate decay. To keep the whole thing from sounding too sterile, the LFO mildly modulates the cutoff frequency. If you want to experiment with variations, I once again recommend FM, Oscillator 1 Interval and Oscillator 2 Interval. 13: Dream Theam There is a style of music that developed entirely around one single sound. It’s that style with the soothingly identical structure, best experienced on German television made for pubescent teens, for example the NBC music/ lifestyle show “Giga.” It starts with a suspiciously young and suspiciously blonde teen female presenter type with nuclear warheads where women typically have breasts warbling a nursery rhyme into a cosmos-sized reverb. Then the camera pans to something that resembles a gigantic sweatshirt with ski goggles. That’s the DJ whose creation we’re marveling at, and with him enters this sound, accompanied by a bass drum that pins the needles on every seismic counter within a 50-mile radius. As they say in Germany, simple means lead to simple ends, and so it is that this sound can not only be perfectly realized on any SoundBlaster card, but is also perfectly easy to program. All you need is two oscillators tuned at octaves with fast pulse width modulation for width, edge and hectic movement and lots of detuning. That’s it. Well, OK: To make it a little more interesting, you can open the filter a healthy amount and add some bite via the resonance knob. Of course, you can do the same thing with EQ.

209

10 Factory Bank Project

One of the most popular “arrangement techniques” in the Trance-DreamDance-Ambient-Rave scene, by the way, is to slowly open the filter while the nursery rhyme is playing back. A nice touch that doubles as a warning, allowing more intelligent listeners to bring their brain cells into safety before the true fans’ remaining vacuum is shaken out. 14: AmbientraK Even the patch name contains a favorite sound designer trick. The name— as so often, a play on words using common “Americanisms”—suggests to The Guy Next Door that this sound comes from another planet and is so amazing that he’s got to like it, or be riddled with doubt regarding his own street credibility. AmbientraK is one of those sounds that anyone—even those who have trouble associating “finger on key” with “tone”—can do something impressive with. Give it a try: Play a major or minor chord (for those described above: three white keys with one in-between simultaneously—there’s only one tricky one) every few seconds. You’ll get a cool chill-out phrase every time, to which you only need to add a bass drum and a bass to become rich and famous. And this is how the sound works: The two oscillators are tuned a fifth apart (“Interval” in Oscillator 2 at −7). The cool, dark trance factor comes from the almost closed filter with a medium resonance setting. Here, a sawtooth waveform is used, but that’s a matter of taste and/or musical context. The delay effect is generated by cutoff modulation via the LFO using a pulse wave for its distinct peaks. The delay “feedback” is controlled by the release time in the amplitude envelope. The filter envelope supplies a crisp attack and a mild damping of the sound in the decay phase.

210

The Factory Bank Sounds

15: Chick Magnet We don’t want to forget an important fringe group among keyboard-testing customers, which—for lack of a better name—we’ll call John Joseph. Sort of an intellectual wannabe-Casanova that, around twenty years ago or so, would take near-sighted, natural-wool crocheting women with names like Abigail to Emerson, Lake & Palmer concerts in hopes that the rumble of the organ’s bass pedals might produce something like a hormone rush— even in a body as organically neutralized as that of his date. Once in a while these attempts were successful—much to the dismay of all nearby who were then subjected to a pitiful chain reaction: Batik and parka-covered vegetarian bodies broke into graceless belly dances, accompanied after a few minutes by the unmistakable bouquet of “Brise d’ Be Eau”… Meanwhile, John Joseph is a social worker in whom the sound of an organ awakens desire: “Organ” means “I want;” our task is therefore “Here you go.” The name for our organ sound not only suggests the opposite of John Joseph, but also the magnetic sound generation of the Hammond we’re emulating. Analog synthesizers with two oscillators give you three “stops” to play with. One common combination: Oscillator 1 generates the tonic (16'), Oscillator 2 the 5-1/3' and the filter resonance the 2-2/3' stop. This is exactly how our sound is structured. The key click comes from an extremely short filter envelope; the “Leslie” effect is generated by the LFO modulating the pulse wave in Oscillator 1. Oscillator 2 generates a sine wave. If you route the LFO rate to the modulation wheel you can even control the speed of the Leslie.

211

10 Factory Bank Project

16: ShakAtak At 17, John Joseph toured India in a VW bus with his CO buddies; since then he’s an incurable world music freak. Now that we’ve warmed him up with our organ, he’ll flip out over our lead sound. This sound name is hard to beat in terms of multiple entendres—it contains hints of “shakuhachi” as well as “shark attack” while contributing to the creation of new slang by artistically leaving out one “t” and the “c” in “attack.” The ability to see through this fills John Joseph with a warm feeling of intellectual humor and profound civil intelligence—both of course purely virtual, in the case of the programmer as well as the target person. Thanks largely to Peter Gabriel’s mega-hit “Sledgehammer,” the shakuhachi (www.shakuhachi.com) has long since become part of everyone’s musical vocabulary; since then the typical asthmatic yodel is an obligatory protagonist in love-film soundtracks and ethno-pop songs. Once again, we take advantage of FM or frequency modulation. An interval of a fifth between the two oscillators, combined with square waves, generates the breathy-metallic basic tonal color. The filter envelope supplies the attack, which negatively influences the pulse width (Symm) of Oscillator 1, creating a second effect: At the end of the decay phase, the tone morphs into a kind of feedback. If you want this effect to be constant, simply deactivate the “Sy(m)” button in the “Filt-Env→Osc” module. The programming of the FM and Symm parameters is the key to this sound—the rest is sound shaping for household use, so to speak.

212

The Factory Bank Sounds

17: Revell Yell Every synthesizer freak knows sci-fi films and associates them with positive things—some more, some less. Case in point: whereas others see Worf when they watch Star Trek: Voyager, I see a bad actor on whose forehead someone stapled a model of the Black Forest out of a 1975 garden railway catalog … On the other hand, I love those old B-movies in which invisible hands control Revell model spaceships on seemingly intentionally not-well-hidden nylon strings in front of a Discovery Channel-style “Wonder of the Milky Way” poster, making “voom” and “cussshhh” noises and holding Bunsen burners. Before I was 12, the appropriate soundtrack for these movies was the synonym for “synthesizer,” and that’s how it’s stored in my ROM. Which pretty much explains the name for this aural homage: Revell Yell. The basic sound needs to be in octaves and pretty thick, which we’ll achieve via pulse width modulation of both oscillators (“Sy(m) 1” and “Sy(m) 2” activated in the LFO module). Additionally routing the LFO to volume (Amp(litude)) creates the rotation effect that one instantly associates with movement or drive train. To simulate the sound of the spaceship approaching and/or passing, we use a combination of a slow attack/release curve in the amplitude envelope and a slow filter sweep. The extremely steep bandpass filter (BP4) and high resonance value make the old space ride sound wonderfully shrill and electronic at the peak of the filter curve. 18 to 20: Sound FX By now I think even the most skeptical among our virtual test listeners has noticed that the SoundForum Synthesizer is capable of a broad spectrum of analog and digital synth sounds. The most important “standards” were included, as well as several complex sounds and real surprises—all in all, a nice, colorful mixture. As dessert—sort of an after-meal wake-up—we’re going to serve our potential customer (and don’t forget the “honey,” sitting on a guitar amp a couple yards away, filing her nails) a couple of sound effects, which:

213

10 Factory Bank Project

1) should create a certain “aha” effect, 2) should be easily modifiable for personal use, 3) should underscore the infinite sound-creating possibilities of the synthesizer and 4) have to come now, because sound effects always come at the end of the factory sounds—that’s what our test listener is accustomed to and is going to expect. Let’s make it short and summarize the three effects here: ❖ 18: Miniplane is a sport plane flying overhead (air battle with Revell Yell, anyone?). The filter envelope controls the pitch; the amplitude envelope also fades in and out slowly. The sound character comes from two oscillators slightly detuned against each other (propellers), while UNISON emphasizes the undulation. The LFO is in charge of damping the propeller sound to a muffled mumble. ❖ 19: Cannonball is the sound of a cannon being fired far away, and can be used as a universal electronic drum reverb. And here’s proof that impressive sounds can be technically simple: Noise in Oscillator 1 is transformed to a subwoofer-worthy rumbling via FM. The FM control determines the distance of the cannon from the listener and should be routed to the mod wheel. You’ll discover the rest on your own. ❖ 20: Kick Yer Butt is the sound with which you can turn the speaker into confetti just before you leave the store—proceed with caution! The filter resonance is responsible for the low-end punch here; the oscillators are totally irrelevant. The filter envelope generates the click (A = 0, D = 28) and the slow downward bend (S = 0.25, R = 56); the amplitude envelope should follow.

214

Factory Bank Overview

Factory Bank Overview Every good factory bank should be documented. If you have serious plans to join the circle of professional sound designers, take my advice and never deliver sounds without documentation. Otherwise, you (or your sounds) may experience some unpleasant surprises during the manufacturer’s evaluation session: Your sweep pad that takes 20 seconds to fully develop might be played staccato, or your first-class solo saxophone could be mutated to an inferno by a ten-finger piano figure. That’s going to happen at the music store anyway … The human psyche works in such a way that when you’re offering something to someone you have to tell them what he should like about it, why he should like it, and how he should use it. And so, just as the synthesizer should have an owner’s manual, the factory sound bank should have playing instructions. And the concise it is, the more success your sounds will have. Whether you are pitching a sound bank to a manufacturer or delivering one already contracted, a brief, concise documentation displays professionalism and enhances your sounds much like a high-quality reverb enhances a voice. What should such documentation look like? ❖ Be brief: Never more than a sentence; if possible almost in shorthand style. ❖ Include all important facts: Number, name, and brief description. ❖ Playing instructions work wonders: Whereas almost everyone knows how a piano should be played, the manner in which some synthesizer sounds are played is so crucial that you shouldn’t risk leaving it up to the individual. The table below shows the documentation for our factory sound bank, just like you might include with a finished job.

215

10 Factory Bank Project Name

Description

Playing Instructions

1

Butterfinger

Fat, undulating cinema-style bass.

Long single low notes.

2

Baked Beans

Wide, undulating pad as contrast to Stately chords in middle register. previous sound.

3

Syncomator

Cutting, rattling sequencer sound.

4

Metallurgy

Morphing metallic sound like a belt Slow melodies or single notes in sander on a church bell. middle or higher registers.

5

Ciao Oop

Low, punchy cinema-style bass.

Typical sequencer-style bass lines.

6

PPG 2002

Typical, gargling PPG sound in “millennium” cloak.

Slow arpeggios or melodies in middle to higher registers.

7

Logical Sound

Electric piano emulation with digital Staccato chords with varying ringing. dynamic (e.g. “Logical Song” by Supertramp).

8

Fjord Lord

Whale song.

9

Machine Head Automated low-end power saw.

Long single notes.

10

OBY 16

Typically fat Oberheim brass sound (caution: greasy!).

Any typical brass-style chords or melodies.

11

FlexPad

Pad sound with complex, morphing overtone structure.

Long chords in lower or middle registers.

12

Chime On

Wind chime emulator; doubles as love lure for hippie chicks.

Random sparkling figures in any register.

13

Dream Theme

Fat rave-style lead sound à la rural disco.

Simple bumber car-style melodies.

14

AmbientraK

Chill-out phrase for people with low Long, long chords. blood pressure (that want to keep it).

15

Chick Magnet

Simulation of an electromagnetic organ with rotor enclosure.

Organ-style chords and phrases.

16

ShakAtak

Ethnic flute sound with distinctive entry or attack phase.

Slow melodies with lots of embellishments in middle to high registers.

216

Long single notes.

Long single notes in middle register.

Factory Bank Overview Name

Description

Playing Instructions

17

Revell Yell

B-movie spaceship simulation

Long single notes or chords.

18

Miniplane

Sport plane flying overhead.

Long single notes in low to middle registers.

19

Cannonball

Cannon being fired far away; can be Short single notes. The lower the used as a universal electronic drum note, the darker the sound. reverb.

20

Kick Yer Butt

Punchy sub-bass with slowly falling Short single notes. The lower the pitch. note, the darker the sound.

217

11 11 Box of Tricks As brilliant culmination of the SoundForum course, I will open my box of tricks for you in this chapter, putting to paper what I have learned throughout my programming years. It’s a widely held misconception that sound design represents a cornucopia of creative passion, exorbitant lifestyle and repulsively high income. The first point is only half true: Sound design is sometimes pure craftsmanship with a little accounting. Even with synthesizer sounds there is something similar to mastering. I’d like to teach you some tricks, with which you can make your own as well as others’ sounds sound just that bit better and achieve the reaction in users that makes the difference between “That’s nice” and “I’ll use that in every track.” Perfect sound and good playability are not always the result of cocaine-induced creativity peaks, but often of the interaction between perfectly harmonizing components. And that’s exactly what we’ll be focusing on here. The basis for the following tricks will once again be the Init snapshot, which—like the other examples for this chapter—can be found in the trick_collection.ens file on the CD that came with this book.

Beating between Oscillators Done Right Frequency Beating between oscillators in a synthesizer sound is the aural counterpart to salt in the kitchen. Beating determines whether a sound is lively and fat or simply horribly detuned.

219

11 Box of Tricks

Two factors are crucial to the sound of a beat: Its frequency and its depth. Beating is generated by pitch friction between the oscillators. The more they are detuned against each other, the faster the beat—you can hear that. Here are a few tips on choosing a frequency: ❖ Slow beating: Lead sounds and “thin” basses should undulate very slowly; slow beating lets a sound be “harder” with more cutting power, and doesn’t get mushy even when you crank up the vibrato. For slow beating, set the “Detune” control in the SoundForum Synthesizer somewhere between 0.01 and 0.03. ❖ Medium-fast beating: This is what you want for polyphonic sounds (chords, pads), arpeggios, sequencer figures and soft lead or melody sounds. Medium-fast beating makes the sound fuller, but also less contoured. Detune range: 0.04 to 0.07. ❖ Fast beating: Appropriate for catchy hook lines, effects and choppy riffs. Fast beating creates a mild to heavy detuned character, causing the sound to stand apart from others. Sounds playing choppy parts can handle detuning better than those playing stately lead lines. Range: 0.08 to 0.23. You can try this out using the snapshot “Detune Tester.” Depth In the course of a beat, the oscillator signals cancel each other out, causing a “hole” or a distinct phasing in the sound. You can, however, level out this “up and down”: ❖ For maximum perceptibility of the beating, both oscillators must be set as identically as possible (waveform etc.), tuned to the same pitch and equally loud. ❖ If you want the sound to undulate but still be stable in lower frequencies, either set one oscillator to a somewhat lower level or tune it down an octave. ❖ Another possibility is to set the two oscillators to different waveforms. Here’s a generalized rule of thumb: The less identical the oscillator signals, the less deep the phase cancellation will be.

220

Pulse Width Modulation

If your synthesizer offers a key-track parameter for oscillator pitch, use it for one of the oscillators at 99% or less, so that the speed of the beating remains constant across the keyboard. Otherwise the beat frequency will increase with the pitch. The SoundForum Synthesizer does not offer this parameter. Special Case: Bass Generally speaking, the bass should not “swim” in the mix. Otherwise, individual notes may disappear. If your track is built on an animated bass sound, double it with a dry, dark 1-oscillator bass (e.g. triangle, sine or filtered sawtooth) to add fundament. Depending on the register, this fundament can also be tuned an octave lower. Several synthesizers offer a dedicated sub-oscillator for exactly this purpose. If you want to fatten up your bass using a chorus, flanger or other modulation effect instead of oscillator beating, try the following trick to keep a healthy bass fundament: Use an external chorus fed through an aux path (send). Insert a high-pass filter into the aux path, so that the lower frequencies don’t reach the effects processor. The sound will be nice and fat, but the bass fundament will stay dry and concise.

Pulse Width Modulation In the good old days, or rather in the good old analog synthesizers, pulse width modulation (PWM) was actually a stopgap (just like the chorus, by the way): Many synthesizers had only one oscillator due to cost, yet had to compete with the big guys. So they were given a chorus module or a PWM circuit. Pulse width modulation can generate many nuances, from “mildly undulating” to “totally detuned.” There are two factors that must be considered: Contrary to standard modulation, pulse width modulation allows you to define range and depth. The higher the PWM frequency, the less depth should be used to avoid detuning.

221

11 Box of Tricks

The lower the played note, the more intense pulse width modulation becomes. You should always use key tracking when available so that the modulation speed adjusts to the pitch. To achieve a well-balanced sound, you shouldn’t use the entire range between square wave (50%) and a practically non-audible nasal sound, since the PWM sound wobbles in this range. Be sure to carefully adapt pulse width and modulation depth to each other. The PWM snapshots are good starting points for practicing PWM programming: ❖ “PWM Range:” Here, the PWM range is set too high; the sound has a distinct hole. Decrease either “Symm” in Oscillator 1 or “Amount” in the LFO module. With the former technique, the modulation retains its depth; with the latter, it becomes milder. ❖ “PWM Speed:” This modulation is programmed so that it sounds good in a middle register (around C3). Play the sound in lower and higher registers. You’ll need to adjust the LFO rate and amount to get a consistent sound in all registers.

Which Type of Fat for Which Recipe? Almost every synthesizer offers four different ways to add fat and width to a sound: ❖ Detuning via the Detune control, ❖ vibrato on one oscillator, ❖ pulse width modulation and ❖ chorus (external in this case). The question is: When do I use which one? Should I combine two or three? Here again a few rules of thumb, which should ease the decision-making process. For a direct comparison, use the “Fat Check” snapshots:

222

Which Type of Fat for Which Recipe?

❖ Detuning (of both oscillators) sounds the most natural and doesn’t rob the sound of its stability, but is usually less spectacular sounding. If you don’t want the sound to be too synthetic, or if you want it to be “true analog,” this kind of undulation is your best bet. ❖ PWM sounds harder and edgier, always adds an easily perceptible movement and is more flexibly controllable than pure detuning (depth and range in addition to frequency). PWM works best if you want the beating to be distinctly audible and/or to make the sound thicker and fuller—fast PWM creates a much wider sound image than fast detuning. You can achieve a “happy medium” between the two by detuning the oscillators, then adding PWM to one. ❖ For strings, choirs or other ensemble-type sounds, try a medium-heavy detuning of both oscillators, combined with a vibrato on one oscillator. This results in a fast, wide modulation that sounds like lots of oscillators. If you have two LFOs available, you can even modulate both oscillators, but you should set the two LFOs at different rates (the ensemble or “symphonic” effect algorithms in effects processors function similarly, with several chorus circuits using different settings running simultaneously). The more different the modulations, the fewer audible repetitions and phase cancellations you will have. For example, you could set one LFO to sine wave and the other to triangle. ❖ Chorus/Ensemble: Modulation effects always “mush up” the sound, since they work with delays of up to several dozen milliseconds, but— like a good shampoo—they can also add a silky shine. Whereas ensemble should be reserved for more “swimming” sounds (slow attack and release times), chorus can also be used on percussive sounds. However, you should only use chorus in the place of detuning if you really have exactly that chorus sound in mind or if it frees up an oscillator for something else. And yet more general rules of thumb: ❖ Detuning should always be the preference for typical analog sounds. ❖ PWM should be used for powerful sounds with more movement. ❖ For ensemble sounds, program different vibratos for each oscillator (assuming you have two LFOs).

223

11 Box of Tricks

❖ Only use separate/external effects to fatten a sound if you want that particular sound character or if it frees up an oscillator that you want to use in another way.

Which Waveform Should I Start with? The sawtooth is always a good waveform as a basis for programming. All other waveforms are portions of a sawtooth. I use square waves, for example, to program hollow sounds, pulse waves for string-like sounds, PWM for fat strings or pads and sine or triangle for especially soft sounds. Try taking any synthesizer sound and setting the waveform to sawtooth— somehow it always sounds good. Filter curves and filter quality can also be judged best with the sawtooth. The Snapshot “Filter Harmonics Test” lets you hear which harmonics are present in a particular waveform: Hold a note and let the filter close slowly. Since the high resonance setting causes the frequencies around the cutoff frequency to be amplified, distinctively audible and visible (in the oscilloscope) peaks will result at those frequencies, where the cutoff frequency meets a harmonic. Switch to square or triangle wave and compare.

Where is the Tonal Color Variety Hiding? You can only generate a few truly different basic tonal colors with the limited number of analog waveforms (sawtooth, square/pulse wave, sine, triangle and noise). Two oscillators extend the palette by allowing you to mix various waveforms and registers. And even in the simply structured SoundForum Synthesizer, FM adds a multitude of variation potential. Thus, you can create a broad spectrum of basic tonal colors with even a simple analog synthesizer.

224

Filter Slope

Simply mixing different waveforms to achieve dramatically different tonal colors with two oscillators only works if you want a subtle variation. For more distinct variations, you need to tune the oscillators to different registers. It makes a big difference in a bass sound, for example, whether a sawtooth is added to a square wave an octave higher or vice versa (snapshots “Bass Saw/Square” and “Bass Square/Saw” demonstrate this).

Filter Slope Just as with the SoundForum Synthesizer, most filters in other synthesizers offer various filter slopes. 12 and 24 dB are common; often you’ll also find 6 dB or 18 dB. Steeper slopes like 36 dB and higher are rare. The practical rule of thumb: The steeper the slope, the more dramatically the filter modifies the sound; the edgier and fatter it sounds—but also the more artificial. This fact can be incorporated into a simple sound designer’s golden rule: The more natural the sound should be, the flatter the slope should be used. This becomes especially important when working with a sampler, sample player or sample-based synthesizer (like most workstations). If the filter is responsible for shaping the sound of a piano or string multisample, even 12-dB filters will often sound artificial. You can hear how the sound is “cut off” and doesn’t decay naturally. Tonal color curves—like the gradual decay of the harmonics in the release phase of a piano tone—are very subtle in acoustic instruments; a 24-dB filter is simply to harsh to simulate this. See for yourself, using snapshot “natural → synthetic” how dramatically different the filter modes sound: Select the various low-pass modes LP1 (6 dB), LP2 (12 dB) and LP4 (24 dB) one after the other. Whereas LP1 very closely resembles a natural decay, LP4 gives the sound a distinct synthetic shape. It is precisely this sound-shaping character that determines the quality and appropriateness of a filter. The sound of an old Moog synthesizer is largely a result of its Moog filter, whereas a Prophet 5 has an entirely different, equally characteristic filter.

225

11 Box of Tricks

Key Tracking Most synthesizers offer a key tracking control, which enables the filter frequency to be controlled by the keyboard or played pitch. Without this, low notes sound too bright and high notes too dark—if they don’t disappear entirely. You can test this with the snapshot “Filter Key Track:” ❖ Play the sound in broad registers. It sounds narrow, undefined and boring. ❖ Now set K-Track in the Filter module to 1 and play again: now it sounds well-balanced across the entire keyboard. For this reason, you should always use key tracking in the filter with sounds that should sound natural and/or be well playable. And here’s the trick: Generally, key tracking is neutral at C3. That’s why I always set the cutoff frequency while playing notes around C3. Afterwards, I test the other registers and adjust key tracking accordingly. You can also use key tracking to influence the character of filter sweeps: Without key tracking, filter sweeps are the same for all simultaneously played notes; the sound evolves in one synchronized movement. Key tracking enables you to break the synchronicity and add more movement to the sound. Give it a try using the snapshot “Sweep Key Track,” playing in broad registers with and then without key tracking.

Setting the Filter Range Correctly In order for a sound to be truly playable, the range that the filter emphasizes must be chosen correctly. Here’s the fastest way: 1 Envelope depth and velocity off; deactivate or neutralize any other modulation sources. 2 Play in the range around C3 and set cutoff to its minimal value, but not so low that the sound disappears. 3 Program key tracking so that the sound is audible and well-balanced across the entire keyboard.

226

Resonance as a Third Oscillator

4 Now set attack in the filter envelope to zero and sustain to its maximum value. Turn up envelope depth until the desired maximum value (the brightest desired sound) is achieved. 5 Finally, program the envelope, velocity and other modulations. If you stick to this method, your sounds will be much more playable and well-balanced. You’ll find an example in the snapshot “Perfect Filter Setting”—the filter sounds balanced over the entire keyboard and dynamic range.

Resonance as a Third Oscillator Many synthesizers allow you to use the filter resonance as a third oscillator. When fully open, it generates a sine wave via self-oscillation, whose frequency is dependent upon the cutoff setting. Some synthesizers require the filter to be triggered by an oscillator to be set to vibration—this is not necessary with the SoundForum Synthesizer. This signal can be used as a third oscillator—especially effective for creating organ sounds with analog synthesizers. In the snapshot “3Drawbar Organ” an organ sound is generated using three sine waves (registers 16, 8 and 2-2/3); the highest of these comes from the filter. A special trick: Since the coarse resolution of the cutoff control often makes it difficult to get exactly the desired tone, as in this case, we fine-tune it with a combination of a very low sustain value in the filter envelope and the envelope depth control. The key click is also generated by the filter envelope (extremely short decay). By the way: Cave-like or other noisy, reverby sounds (sonar, flute) can be simulated quite effectively via a combination of noise from an oscillator and a self-oscillating filter—try it out! The Snapshot “Tuned Noise” was created out of the snapshot “3Drawbar Organ” by modifying just a few parameters and demonstrates exactly this effect.

227

11 Box of Tricks

What Are Band-pass and High-pass Good for? Although multimode filters are standard these days, a low-pass filter is used for probably 99% of all synthesizer sounds. This is not due to a lack of creativity on the part of sound designers, but simply to the fact that a low-pass filter shapes the type of curve that most acoustic instruments possess, and you need this type of curve for most sounds. After all, the human being is a creature of habit. But what can you use band-pass and high-pass for? An acoustic impression is available in the snapshot “HP BP Test.” Hold chords while switching between the various BP and HP modes of the filter. BL4 is a special case: It’s actually a BP (band-pass) filter, but with a 6-dB high-pass and 18-dB low-pass, and sounds more musical than BP4. A high-pass filter—which shouldn’t be surprising considering the name— is exactly the opposite of a low-pass filter: It lets high frequencies pass, attenuating the portions below the cutoff frequency. Therefore, high-pass sounds are thin and cutting, because the highs are always fully present, while the lower harmonics are missing in part. The snapshot “HP Intro Sweep” shows how a high-pass filter can be used to create a sweep sound that evolves essentially out of nothing into an ice-cold pad. A band-pass filter, on the other hand, is a combination of low-pass and high-pass. It lets only the harmonic content around the cutoff frequency pass unaltered and is therefore especially effective for shaping formants— for example the resonant characteristics of an instrument body (i.e. guitar, piano or woodwind). In this case, the bandpass filter works like a parametric midrange equalizer. The snapshot “Oboy” demonstrates just that: A sawtooth wave is cut here in such a way that the sound has the same thin, nasal character in all registers. Why? Because key tracking is deactivated, causing the cutoff frequency to remain the same regardless of pitch. Not that I want to imply with this patch that an analog synthesizer is capable of simulating even vaguely realistic brass sounds … The idea should come across, though.

228

Juicy or Creamy?

Juicy or Creamy? In previous chapters I have philosophized now and then about the subconscious association between sounds and perceptions. Sounds that one likes trigger pleasant associations, which is why people generally like warm, fat or cool sounds, whereas thin, mushy or screeching sounds are less popular. With this last little trick, I’d like to show you how you—using almost only the filter—can realize sounds that automatically trigger associations from “juicy” to “creamy,” as if one would bite into a ripe peach or a piece of toffee, depending of course on individual taste. For some reason, sounds like this are especially popular among the audience at large and are widely used, among other things, in songs that seem to always have something to do with reproduction or its preliminary activities. Give the snapshot 20 “Juice or Cream” a test drive, and you’ll see exactly what I mean. You get these sounds automatically with a certain combination of filter, filter envelope and amplitude envelope. The better the filter, the more distinct the “wellness” effect. The core of the whole thing is the effect of letting the frequency of a lowpass filter snap open quickly and then slowly close again—similar to the time curve with which a cream pie covers a victim’s face (attack) and frees it again (release). Cutoff and envelope depth settings must be set according to the description above. The liquidity is determined by the filter resonance. At a lower setting, it sounds like cream or toffee; at a higher setting it approaches syrup or juice. (In case you might suspect that I was under the influence of a South American “medication” purchased in a dark alley while writing these lines, check the description again. The association is generally similar among musicians with a similar social background …)

229

11 Box of Tricks

It’s also important to avoid a “cardboard”-like effect when the key is released. To this end, the release time in both the filter and amplitude envelopes must be very short. The thickness of the associated substance can be influenced by the attack time in the filter envelope—longer attack times create a riper, mushier tone. At this point, feel free to tweak everything on the SoundForum Synthesizer except the described modules. You’ll discover that regardless of what goes into the filter, toffee or syrup comes out. Which proves once again just how crucial the filter is to the sound.

230

Contents of the CD-ROM

Windows ❖ Soundforum Synth Setup.exe This is the installation program for the SoundForum Synthesizer. Start the installation with a double-click; it will automatically take you through the installation process step by step. ❖ Sounds folder This folder contains the ensemble and snapshot files that you’ll need during the course.

Mac OS ❖ Soundforum Synth Installer This is the installation program for the SoundForum Synthesizer. Start the installation with a double-click; it will automatically take you through the installation process step by step. ❖ Sounds folder This folder contains the ensemble and snapshot files that you’ll need during the course.

231

Index A

D

Acoustic bass 189 Acoustic guitar 188 ADSR envelope 111, 186 Amp Env module 111 Amplitude envelope 111 Attack 111 Auto sweep 194

Decay 112 Detune 105 DX-Rhodes 178

B Band-pass filter 108 Bass Acoustic ~ 189 Slap ~ 189 Synth ~ 189 Beating 50, 219 Bell sound 72 Bells 179, 188 Brass section 189 Brass sounds 181, 189 Breath sounds 183

E Echo 195 Electric guitar 188 Electric piano 187 Ensemble 189, 193 Envelope ADSR ~ 111 Amplitude ~ 111 Attack 111 Decay 112 Filter ~ 114 Release 112 Sustain 112 Envelope generators 184 Envelopes 97, 184 EQ 143 Equalization 143

C Choir 182, 189 Chorus 192, 193 Chowning, John 61 Control voltage 96 Cutoff frequency 107

F Filt-Env→Osc module 115 Filter 106 Band-pass 108 Cutoff frequency 107

233

Index High-pass 108 Low-pass 107 Resonance 109, 227 Slope 107, 225 Filter Env module 114 Filter envelope 114 Filter self-oscillation 75 Flanging 192 FM 57, 104 Frequency Modulation 57

G Guitar acoustic ~ 188 electric ~ 188

H Hammond organ sound 211 Harpsichord 187 High-pass filter 108 Horn sounds 181

K Keyboard tracking 110 K-Track 110

M Mallets 179 Metallic bass sound 69 MIDI Learn 87 Mixer 105

N Noise 104

O Organ Percussive ~ 188 Pipe ~ 188 Oscillator 41

P Perc(ussive) organ 188 Pipe organ 188 Plucked strings 179 Pulse wave 102 Pulse width modulation 52, 221 PWM 52, 221 PWM bass 178

R L Laser effects sounds 75 Laser gun sound 191 Latency 22 LFO 42, 117, 190 LFO waveforms 195 Low Frequency Oscillator 42, 190 Low-pass filter 107

234

Random Pitch 193 Release 112 Resonance 109, 227 Ring modulation 65, 71, 105 RingMod 105

S Sample & Hold 118 Sawtooth wave 103 Sine wave 103 Slap bass 189

Index Slope 107, 225 Solo brass 189 SoundForum Synthesizer 19 ASIO 25 Audio Drivers 22 Compare 32 Configuration (Mac OS) 24 Configuration (Win) 21 CPU load 31 De-installation (Win) 24 Features 20 File Menu 27 Help Menu (Mac OS) 30 Instrument Menu 30 Latency 22 MIDI Learn 32, 87 Number of voices 89 OMS 25 Playing the ~ 26 Properties 34 Remote control 87 Snapshots 27, 36 System Menu 29 System Requirements 20 Toolbar 31 Unison 33, 90 User interface 33 View Menu (Mac OS) 30 SoundSchool Analog synth 16 Square bass 175 Square wave 102 String ensemble 180 Strings 189 Sub-bass 174 Subtractive synthesis 41 Sustain 112 Sync 65 Sync bass 177 Sync lead sound 68 Synth bass 189 Synth bass sounds 77

Synth brass 190 Synth horns sounds 78 Synth string sounds 80

T TB bass 176 TB-303 45, 176 Trance lead sound 193 Tremolo 194 Triangle wave 103 Trill 192 Trumpet sounds 181

V Vibraphone 187 Vibrato 191 Voices 181

W Waveforms Noise 104 Pulse 102 Sawtooth 103 Sine 103 Square 102 Triangle 103 White noise 104 White noise 104 Wurlitzer 178 Wurlitzer electric piano 54

X Xylophone 188

Y Yamaha DX synthesizers 57

235