What makes a good instrument?
What makes an instrument good?
What makes a sound good?
What makes a sound bad?
A good instrument is a resonating system that is able to express the emotion of the player/composer by manipulating pitch, timber and volume in such a way that the listener perceives the same emotion transmitted through the music. A good instrument is expressive. Changes of pitch, harmonic content and amplitude need to be controlled through controllers like velocity where playing hard creates a brighter louder sound. and playing soft elicits darker softer sounds, From the earliest moments of our childhood we have learned that soft dark gentle sounds are associated with crooning, comfort and peace. Once we learn to pull the family dinner off the table by tugging on the table cloth we learn to associate tension with sharp bright sounds and sudden quick movement.
Your ears are only interested in one thing: Where is the noise of the attacking saber tooth tiger coming from and how fast is it closing in on me? Good sounds provide the ear with lots of information – harmonics and phase relationships which change over time. The ear can only extract useful information from the way a sound changes. The only really bad sound is the sine wave because it provides the least information possible to the ear . In a crowning irony every sound on Earth is made from just sine waves and noise. Ears were born to register changes in sound – not the sound itself.
There are seven kinds of synthesis:
Analog/Subtractive – start with a bright sound.
Additive – many sine wave harmonics.
Frequency Modulation – modulate one sine wave with another.
Wave table – many single cycle loops.
Sample Play – Subtractive with acoustic waveforms.
Granular – cut the sound into 100ms grains and fire them out.
Physical Modelling – a flight simulator for instruments.
Logic Pro X does all of them.
Click on the images to hear a demo of that synth on YouTube.
Here is a list of the major developments in synthesis during the last century to today. New ones are coming online every day.
1.Analog – Monophonic:
These things have been around since the beginnings of synthesis technology – after all, they are the simplest to create bearing in mind that they need only to have the facilities to produce 1 voice, and that they use less complex (but many would argue better sounding) analogue components. Examples: MiniMoog and ARP Odyssey.
Subtractive Synthesis 1968- 1985
Includes all analog synthesizers and most modern sample players. Starts with a bright sound and cut away what you don’t want. MiniMoog, Arp2600,
The grand daddy of them all – Moog 55 $35,000h
2.Analog – Polyphonic:
A polyphonic version of the above, meaning that they are able to play more than one note without compromising the other voices. In the early days these were gargantuan beasts with a sound to match, but the Prophet 5 revolutionised analogue polyphonic synthesizers due to its use of a computer microprocessor, giving it the ability to digitally store patches. My first job when I ditched band life and got off the road was with Sequential Circuits – makers of the Prophet 5.
Example: Prophet 5 $3500
Logic Equivalents: Click on Synth name for manuals click on Tutorial for a YouTube Tutorial of that synth.
In 1984 Yamaha bought Sequential Circuits and I joined Yamaha during the Glory Days of the DX7.
3. Digital Synthesis
A relatively short-lived group of synthesizers, mostly from around the 1980s featuring alternatives to subtractive synthesis. Yamaha pioneered this idea with their DX range of FM-based synthesizers, as well as contributions from Casio with their phase distortion technique, and Roland with their still very popular D-50 synth, which used a new type of synthesis called ‘Linear Algorithmic’. These synths, although responsible for the downfall of analogue machines, helped give musicians the ability to create completely new and unique sounds never heard before. Examples: Yamaha DX-7
Modulates sine waves to create harmonics
The Yamaha DX7 was the most successful synthesizer ever sold.
4.Wave table synthesis.
Stored large numbers of single cycle loops. Ensoniq ESQ 1
5. Additive synthesis
Stacked up to 127 sine wave harmonics.
Best description “the good news is that you can do everything. The bad news is that you have to do everything” Walther/Wendy Carlos.
6 Vector Synthesis
After the Prophet 5 Sequential came up with another idea – Vector Synthesis
There was a joystick with four poles and we could add a sound to each of the poles. The sounds were single cycle wave tables. It was possible to record the motion of the joystick through 127 segments. When you played a note it would repeat the sequence.
It sounded great but the imminent demise of Sequential and its purchase by Yamaha. Yamaha issued a low cost version called the SY22 but it was judged to be a niche market
6. Phase Distortion
Introduced by Casio as the CZ101 series phase distortion operates by changing the phase angle of a sine wave thereby distorting a sine wave into a sawtooth, square wave and triangle wave forms. Synthesizer parameters that follow the engine are typical subtractive synthesis parameters.
7. Granular synthesis
The basic principle behind granular censuses is that song can be broken down into tiny segments often less than 50 ms long. The segments or grains can then be recombined to create entirely new sounds. Long beloved by the academic world granular synthesis never made it into main stream synthesis however it is now available in the iPad
8. Sample Player
in 1996 Korg introduced the first sample play instrument, the
KORG M1. It was an instant success.
In an ongoing effort to b everything to everyone Yamaha and Roland both came up with Hybrids. Roland introduced the D50 which was a combination of analog and sample playback. Yamaha introduce the SY99 which was a hybrid of FM and sample player
The SY99 was the first project of my Sound Design Office.
10. Physical Models
As flight simulators are to airplanes so Physical Models are to instruments. The Yamaha VL1 was the world’s first physical model of a wood wind. Click on this picture of my friend Katsu Ujie demonstrating the VL1.
Physical models are now the wave of the future because they imitate the sound of instruments but also the behavior. Unlike samples which are dead sounds physical models are alive and respond in a living manner.