1.4.1 Analog vs. Digital

With the evolution of computer technology in the past 50 years, sound processing has become largely digital. Understanding the difference between analog and digital processes and phenomena is fundamental to working with sound.

The difference between analog and digital processes runs parallel to the difference between continuous and discrete number systems. The set of real numbers constitutes a continuous system, which can be thought of abstractly as an infinite line of continuously increasing numbers in one direction and decreasing numbers in the other. For any two points on the line (i.e., real numbers), an infinite number of points exist between them. This is not the case with discrete number systems, like the set of integers. No integers exist between 1 and 2. Consecutive integers are completely separate and distinct, which is the basic meaning of discrete.

Analog processes and phenomena are similar to continuous number systems. In a time-based analog phenomenon, one moment of the phenomenon is perceived or measured as moving continuously into the next. Physical devices can be engineered to behave in a continuous, analog manner. For example, a volume dial on a radio can be turned left or right continuously. The diaphragm inside a microphone can move continuously in response to changing air pressure, and the voltage sent down a wire can change continuously as it records the sound level. However, communicating continuous data to a computer is a problem. Computers “speak digital,” not analog. The word digital refers to things that are represented as discrete levels. In the case of computers, there are exactly two levels – like 0 and 1, or off and on. A two-level system is a binary system, encodable in a base 2 number system. In contrast to analog processes, digital processes measure a phenomenon as a sequence of discrete events encoded in binary.

Aside:  One might think, intuitively, that all physical phenomena are inherently continuous and thus analog.  But the question of whether the universe is essentially analog or digital is actually quite controversial among physicists and philosophers, a debate stimulated by the development of quantum mechanics.  Many now view the universe as operating under a wave-particle duality and Heisenberg’s Uncertainty Principle. Related to this debate is the field of “string theory,” which the reader may find interesting.

It could be argued that sound is an inherently analog phenomenon, the result of waves of changing air pressure that continuously reach our ears. However, to be communicated to a computer, the changes in air pressure must be captured as discrete events and communicated digitally. When sound has been encoded in the language that computers understand, powerful computer-based processing can be brought to bear on the sound for manipulation of frequency, dynamic range, phase, and every imaginable audio property. Thus, we have the advent of digital signal processing (DSP).

1.4.2 Digital Audio vs. MIDI

This book covers both sampled digital audio and MIDI. Sampled digital audio (or simply digital audio) consists of streams of audio data that represent the amplitude of sound waves at discrete moments in time. In the digital recording process, a microphone detects the amplitude of a sound, thousands of times a second, and sends this information to an audio interface or sound card in a computer. Each amplitude value is called a sample. The rate at which the amplitude measurements are recorded by the sound card is called the sampling rate, measured in Hertz (samples/second). The sound being detected by the microphone is typically a combination of sound frequencies. The frequency of a sound is related to the pitch that we hear – the higher the frequency, the higher the pitch.

MIDI (musical instrument digital interface), on the other hand, doesn’t contain any data on actual sound waves; instead, it consists of symbolic messages (according to a widely accepted industry standard) that represent instruments, notes, and velocity information, similar to the way music is notated on a score, encoded for computers. In other words, digital audio holds information corresponding to a physical sound, while MIDI data holds information corresponding to a musical performance.

In Chapter 5 we’ll define these terms in greater depth. For now, a simple understanding of digital audio vs. MIDI should be enough to help you gather the audio hardware and software you need.