This post is the first in a series called Making Music with MIDI. What I’ll be talking about applies to all MIDI instruments (including hardware) and much of the software that includes MIDI, regardless of the computer platform, no matter how old or new.
This series is not called “Making MIDI music”.
First of all, MIDI doesn’t make any sounds at all. Like a telegraph key used to send Morse code, or e-mail, MIDI just sends characters over a wire – no audio.
Second, if “MIDI Music” exists at all, it’s not very musical. A lot of the stuff out there might be called “MIDI music”, but it’s not very musical. Just playing notes without ever changing the timbre isn’t very musical. Think of cheap handheld-game “music” or early chiptunes … every note’s the same volume, the same tone. Strictly beginner stuff. Like playing the piano without using the pedals!
If MIDI instruments don’t sound as good as acoustic instruments, that’s not entirely MIDI’s fault. It’s not Morse’s fault if the dude pounding the key can’t keep time or swing his dits and dahs. MIDI has been and is used to make scores for major films. You might not have known, because the composers really knew MIDI.
High Quality Midi soundfont demonstration
Acoustic musical instruments playing the note “Middle C” all sound different. That’s because none of them generate a simple sine wave. When you play a piano key or bow a violin, pluck a guitar string or mallet a xylophone bar, they all produce a different “Middle C”. What’s different between them? The timbre (whether you say tam-ber or tim-ber or tom-brr).
What ‘timbre’ is exactly is a complicated subject. Let’s start with two important parts of “a” note:
1. Attack, Decay, Sustain, Release
2. Which harmonics (aka partials, overtones) are present, which aren’t, their relative volume levels, and how that changes over time … all summed up in the term “spectral envelope.”
Basics of timbre at Hyperphysics
“Artistry” or “expressiveness” is part of what sets musicians apart. Timbre is another part. Whether they play strings, keyboards, winds … players select their instruments with care to ‘get a sound’ that pleases them and suits the music they play.
How to listen to the harmonics of a particular timbre
Trumpet note’s overtones evolve.
(Rotten old-school graphic, looking for better)
It’s hard to find good diagrams showing the overtones/harmonic series of a single note.
It’s a lot easier just to hear it if you have a synth with a FILTER.
Choose or make a patch that plays a sustained note that uses a PULSE waveform (the richest kind). Set the filter for lowpass mode. Set the filter cutoff so that it’s “wide open” … you can hear the whole sound. Now slowly turn the knob of (or otherwise adjust) the cutoff filter. As you do so, the sound will slowly become “less rich”. As the lowpass “reaches bottom” it will erase even the fundamental pitch, and you hear silence.
If you switch to using a SINE waveform, you’re listening to the simplest possible timbre … made of only ONE frequency playing at a constant volume. The lowpass filter will reduce the loudness of that one frequency, but that’s all.
Most basic synths have Oscillators that produce at least 4 simple waveforms. The other two are usually TRIANGLE and SAW. Their “richness” (number of overtones) is between a SINE and a PULSE. For such synths, that’s the entire basic sound palette you have to paint with. (Others may add a NOISE waveform.)
Square, saw and triangle waves at Hyperphysics
BANDPASS: If your synth has a bandpass (“frequency window”) filter that can be set to a really narrow bandwidth, you can sweep the “window” across a sound and hear each individual harmonic.
FM: In 1977 John Chowning got a patent for a new kind of synthesis called FM (frequency modulation). It creates many more kinds of waves, and made the DX-7 possible … a synthesizer people didn’t have to sell an arm for.
Sound synthesis at Hyperphysics (short)
Of course, other kinds of oscillators can produce other waveforms (eg WAVETABLES). And then, there are waveforms recorded from actual acoustic instruments (found in eg SOUNDFONTS or “ROMplers” – collections of samples MIDI can control). Today’s SOFTSYNTHS can be more versatile, but often lack the ‘character’ of classic hardware synths.
Basics of sound synthesis article at Pro Audio Files (longer)
Sound Synthesis article at SOS (longest)
In part 2 of this post, we’ll explore a few of the -many- ways MIDI can be used to control timbre … while playing/performing or afterwards with an editor.
In the meantime …
If you need to just catch up on the terminology here, Wikipedia will probably be enough for you.
If you really want to dig down in detail, there are many books; the one I like best is Charles Dodge and Thomas Jerse, “Computer Music” (I have the 2nd edition from 1997. ISBN 0-02-864682-7) Some other things to look at: ADSR and CCs (continuous-controllers). If you want to know A LOT about MIDI, I’ve never found a finer book than Robert Young, “The MIDI Files” (mine’s from 1996. ISBN 0-13-2620403-6)
Helpful Youtube videos (Dec. 2015)
IMPMOOC Timbre class
Partials, Overtones, & Harmonics Explained
Overtones, harmonics and Additive synthesis
Frequency Modulation Synthesis (part 1 of 2)