The following contains excerpts from the book, Electronic Music and Sound Design (Alessandro Cipriani, Maurizio Giri).
This book alternates between teaching concepts of sound design, theoretically, and showing how it is done using Max/MSP. As David Zicarelli, the publisher of Max/MSP, states in the foreword: “This book provides the language for expressing the concepts in terms of the underlying theory (of sound design). Knowing the theory is essential…to be more than someone who can turn a volume knob.”
At our music school in Dallas, Texas we teach our composition students the value of using Max/MSP in their creative work.
Part 3 of this book deals with the subject of Noise Generators, Filters, and Subtractive Synthesis.
Sound Sources for Subtractive Synthesis
“A filter is a signal processing device that acts selectively on some of the frequencies contained in a signal, applying attenuation or boost to them. The goal of most digital filters is to alter the spectrum of a sound in some way. Subtractive Synthesis was born from the idea that brand-new sounds can be created by modifying, through the use of filters, the amplitude of some of the spectral components of other sounds.”
It is important for students at our music school in Dallas, Texas to understand noise generators, and how they work.
Noise Generators: White Noise and Pink Noise
“One of the most commonly used sources sounds for subtractive synthesis is white noise, a sound that contains all audible frequencies, whose spectrum is essentially flat (the amplitudes of individual frequencies being randomly distributed)…Another kind of noise that is used in similar ways for subtractive synthesis is pink noise. This kind of sound, in contrast to white noise, has a spectrum whose energy drops as frequency rises. More precisely, the attenuation in pink noise is 3 dB per octave; it is also called 1/f noise, to indicate that the spectral energy is proportional to the reciprocal of the frequency.”
Understanding how noise is used in subtractive synthesis is important for students at our music schoolin Dallas, Texas.
Simple pseudo-random sample generators “generate random values at a given frequency, maintaining a constant value until it is time to generate the next sample.”
Interpolated pseudo-random sample generators “use interpolation between each random number and the next…Interpolation between one value and the next can be linear…or polynomial, implemented using polynomial functions to connect the values using curves rather than line segments…The kinds of polynomial interpolation most common to computer music are quadratic (which use polynomials of the second degree) and cubic (which use polynomials of the third degree).”
Oscillators and Other Signal Generators
“Band-limited oscillators are often used in digital music. Such oscillators, which produce the classic waveforms, are built so that their component frequencies never rise above half of the sampling rate.”
“Beyond subtractive synthesis, one of the everyday uses of filters and equalizers is to modify sampled sounds.”
Second-order filters “provide twice the attenuation of first-order filters.” Examples of a second order filter are Butterworth filters (named after Stephen Butterworth, the physicist who first described them).
Filters help shape the sounds students in our music school in Dallas, Texas create.
Subtractive Synthesis
“Subtractive synthesizers are generators of audio signals that can be controlled by a musician using a control paned, a keyboard, or some other device. These synthesizers are driven by one of more oscillators, also called VCO (Voltage controlled Oscillators), which can produce classic waveforms (sawtooth, square, or triangle), as well as white noise, which is used to create specific effects. The outputs of these oscillators can be mixed together, and can also be slightly detuned in order to create beats.”
As students in our music school in Dallas, Texas learn how to create their own subtractive synthesizer, they add to their palette a new array of soundscapes.
“The sound produced by the oscillators is processed using an amplitude envelope (VCA: Voltage Controlled Amplitude) which defines its attack, decay, sustain and release. After this, the sound is sent to a lowpass filter (or multiple cascaded filters) whose cutoff frequency depends upon the note played (it might be, for example, a multiple of the fundamental). Key follow is the term used for this kind of parameter, whose value tracks the frequency of the key pressed…Such a filter may also have its own envelope for cutoff frequency (VCF: Voltage Controlled Filter), independent of the key frequency or the amplitude envelope, and Q can also be either set and/or controlled by another envelope.”
It is possible, with two oscillators, “to set hard sync between the first and second oscillator; by doing this you get an often-used feature in this kind of synthesizer that synchronizes the cycle of the two oscillators, even when they have different frequencies. When hard sync is active, every time that the first oscillator stars a new cycle, the second oscillator will be forced to restart, independent of its frequency.”
Equations for Digital Filters
“Digital filters are divided into two categories: FIR (finite impulse response) filters, which operate using what is called feedforward, and IIR (infinite impulse response) filters, (called also ‘recursive filters’) which operate using feedback coupled with feedforward.”
“When referring to the input and output routings for a filter, you use the term feedback when one or more copies of the output signal are delayed, then multiplied by a coefficient, and finally combined with the ongoing input signal. The term feedforward is used when one or more copies of the input signal are delayed, then multiplied by a coefficient, and then combined with the continuing input signal.”
“FIR filters (which only use feedforward) are characterized by their stability and the fact that they cannot be driven to oscillate, since they don’t use a Q factor to produce resonance.”
“IIR filters (which use feedback coupled with feedforward) can be very powerful filters. Although the calculations needed to implement them are simple, they can be subject to several undesirable effects such as ringing, phase distortion, and inaccurate propagation of high frequencies and transients.”
“A digital filter carries out three simple operations on the samples that it receives: it multiplies them by given values (called coefficients), it delays them, and it sums them together.”
“The algorithm for a non-recursive filter…needs only input samples to calculate its filtered output signal…A first-order filter will use the current input sample along with the immediately preceding sample, while a second-order filter will use the current sample plus two preceding samples…a sample is being produced every 1/44, 100 of a second, or about every 0.023 milliseconds (a period that we will call the processing cycle)…the sample from the previous processing cycle is stored into a dedicated memory cell, called a delay line.”
[Important audio programming languages: CSound and Max/MSP]
As students at our music school in Dallas, Texas begin to acquire skill in manipulating oscillators, filters and envelopes, they put at their fingertips a wide array of possible tonal sonorities.