Hi there, in this post I’ll be going over the basics of synthesis and sound design. It’s meant to help you understand how synths work, and give you the foundation to start creating sounds.
Before we begin, I’ll briefly introduce the types of synthesis.
There are multiple types of synthesis, among the basics and popular ones are :
- Subtractive (Sylenth1, GMS, 3xOsc, Synth1)
- FM (Sytrus, FM8, Operator, Dexed, DX7)
- Additive (Harmor, Harmless, Loom)
*both Harmor and Harmless are actually additive synthesizers that functions like subtractive synthesizers - Wavetable (Serum, Massive)
Subtractive synthesis functions by taking a basic waveform and running it through envelopes, filters, and effects.
FM synthesis functions by taking a waveform, and then modulating it with another waveform. You can also use a modulated waveform to modulate another waveform. It also runs through envelopes, filters, and effects.
Additive synthesis functions by adding on multiple sines to create a waveform. It too runs through envelopes, filters, and effects.
Wavetable synthesis is similar to subtractive synthesis, except that instead of having only basic waveforms, you get more waveforms in addition to them. And yeah, it too runs through envelopes, filters, and effects.
What do they all have in common?
They all have waveforms, and they all run through envelopes, filters, and effects. That’s what we’ll be going over.
Let’s get started.
Waveforms
We’ll start with waveforms, as that’s the basis of every sound. Without a waveform, you wouldn’t get any sound.
What’s a waveform?
To put it simple, a waveform is just a shape that’s constructed by points to make a sound. Inside synths though, they’re also oscillating.
While there are many waveforms, we’ll only be learning the 4 basic waveform shapes as most sounds are made based on these 4 waveforms.
Here’s a video of what they look like in analyzers and how they sound like :
Let’s talk about them in order of appearance.
First up, the sine.
The sine is the purest waveform, and if you look at the video, it only occupies one frequency. It is the foundation of waveforms as all waveforms are built from it.
Unfortunately, because it only occupies one frequency, some synths don’t have it as a basic waveform. This is because it doesn’t fit the “subtractive” nature of subtractive synths since it doesn’t have anything to subtract from, and it can easily be created by low passing a triangle wave.
Speaking of triangle waves.
It sounds very similar to a sine, except that it has a bit more frequencies above it. The frequencies are made with only odd harmonics, similar to the square wave. The difference though, is that the harmonics drops off faster than that of a square wave.
I mentioned harmonics, but what are they?
To put it simple, harmonics are these things :
Notice how there’s more sines on the higher frequencies in addition to the first sine. These are sine waves that are multiples of the fundamental sine wave (which is almost always the first one on the left). They’re what determines the timbre of a waveform.
A timbre is just the characteristics/colour of a sound. Descriptors like bright, dark, warm, harsh, dull, etc. are describing the timbre of the sound. Think of it like textures but for audio instead.
Enough about that, let’s get onto square waves.
As I mentioned earlier, they’re similar to a triangle wave. They’re made with only odd harmonics, but with the higher harmonics rolling off much slower than that of a triangle.
Square waves are a bit unique, they can be controlled by something called Pulse Width Modulation. Pulse Width Modulation (or PWM) controls the spacing of the “squares”.
Here’s a sample of what it does when I gradually turn it up and down :
Onto the next one, saw waves.
Saw waves are different. They’re like square waves, but instead of just odd harmonics, it’s made with both even and odd harmonics.
Saw waves are the most common waveform that’s used to make lots of sounds simply because it’s rich in harmonics.
Right now, you may be wondering which waveforms should be used for which sounds?
You can use any waveform to make any type of sound, it’s important to experiment! Though, the general usage are as follows :
Lead : Square, Saw
Pad : Square, Saw
Basses : Triangle, Square, Saw
Sub-Bass : Sine, Triangle
Now that we’ve covered waveforms, let’s move onto envelopes.
Envelopes
Envelopes are bread and butter for synthesis. They’re prevalent in all synths and could control pretty much any parameter should the synth allow it. Generally though, they control amplitude.
Envelopes trigger on a note by note basis and they mainly consists of 4 parameters, ADSR.
ADSR stands for :
- Attack
- Decay
- Sustain
- Release
In addition to that, they might also contain a “hold” parameter between the attack and decay. Let’s understand what each parameter does.
Attack
Attack is a parameter that controls how fast a sound ramps up to full value. It starts from 0 and goes to 100.
In some synths, you can change the curve shape but usually they don’t allow you to. Synths will usually default to using a linear curve for parameters such as this.
Hold
Hold is a parameter that controls how long a sound would hold at full value after the attack before going to the decay. Think of like the decay with 100 sustain, except that it’s another parameter.
It’s pretty rare to find in synths, but it’s good to understand how it works.
Decay
Decay is a parameter that controls how fast a sound goes down from full value. It will start from 100 and instead of going only to 0, it goes to the volume that is set by the sustain parameter.
It behaves like the attack, but instead of going up, it goes down.
Sustain
Sustain is a parameter that controls the end point of the decay. 0 will make the decay end at 0, 50 will make it end at 50, 100 will make it end at 100.
Notice in the clip that when the sustain is at 100, the decay will not do anything. This is because the decay only drops to the point the sustain is set. 100 being full value, it won’t drop anywhere.
Release
Release is a parameter that controls how fast the envelope fades out after a note is released. It will always go from the value the key is released, to 0.
Envelopes are helpful in controlling a sound’s parameters. Sometimes, automation is not needed and would only clutter things up, envelopes help solve this problem.
Now let’s move onto something that’s a bit similar, LFOs
LFOs
Low Frequency Oscillators (or LFOs) are kind of like envelopes, but they use waveforms as a controller instead. LFOs play indefinitely, but you can control them with envelopes should the synth allow you to.
The parameters that you can modify on an LFOs are the waveform and rate.
Here’s a brief explanation.
For waveforms, LFOs usually give you only the basic shapes but sometimes a bit more like a noise or a PWM Square.
As for the rate, LFOs usually let you change it either using Hz or sync it with the tempo.
A bit on Hertz (or Hz)
Hz is a unit to measure frequency. 1 Hz means that something is oscillated 1 times per second. If you set an LFO’s rate to something like 4Hz, that LFO will repeat 4 times every second. Likewise for any value.
Here’s a video of an LFO in action :
LFOs are pretty similar to envelopes, except that it triggers indefinitely rather than on every note. Though you can make it retrigger on every note by turning on retrigger.
AM, RM, and FM.
Now these aren’t actually parameters for an LFO, but I think it’s a good time to segue into this as they’re very similar.
AM, RM, and FM are like LFOs, but they operate at a higher frequency. So they’re really fast. All three of these has a carrier (the modulated waveform) and a modulator (the modulating waveform).
I’ll only quickly go over it as to not overwhelm you.
AM
AM is amplitude modulation. It takes a waveform, and modulates the carrier’s amplitude with that waveform. Pretty much like an LFO.
AM is rarely found in synths nowadays, as there’s RM which is similar to it.
RM
RM is ring modulation. It’s very similar to AM, but it goes into the negative amplitude as well. If you know about phases, this just means it inverts its polarity too unlike AM.
FM
FM is frequency modulation. It’s similar to RM in terms of it going into the negatives as well but instead of modulating amplitude, ̶i̶t̶ ̶m̶o̶d̶u̶l̶a̶t̶e̶s̶ ̶f̶r̶e̶q̶u̶e̶n̶c̶y̶ ̶(̶a̶s̶ ̶t̶h̶e̶ ̶n̶a̶m̶e̶ ̶i̶m̶p̶l̶i̶e̶s̶)̶ it actually modulates phase instead (sorry, not as the name implies haha). There’s some math to it so go look it up if you’re curious!
If you set these to really low frequency, it actually acts like an LFO. LFO does mean “Low Frequency Oscillator” afterall.
Now that we’re done with that, let’s get into filters.
Filters
If you’ve been making music for a while, you would probably know what filters are, but let me touch this a bit.
Filters filter out frequencies based on the filter type. The frequencies they filter are determined by the cutoff point. A high cutoff point will be at the higher frequencies, and vice versa.
There are many different kinds of filters, but the ones most commonly used for synths are :
- Low Pass
- High Pass
- Band Pass
I’ll explain each of them.
Low Pass & High Pass
Both low pass and high pass are similar. A low pass filter out the high frequencies, and let the low frequencies pass. A high pass does the opposite of that. They both get their name from what they do.
Band Pass
A band pass filters out frequencies in a specific range. The reason why it’s called a band pass is because in equalizers, a band controls the cutoff point for a filter type and a band pass only allows the frequencies around that band to pass.
Those are it for filter types, now onto resonance.
Resonance controls how much area around a specific band effects. A high resonance would increase the volume at and around that band, and a low resonance would do the opposite. However, for a band pass, resonance instead affects the range of frequencies could pass through around that band.
Resonance actually originates from artifacts in the original cutoff filters used in old hardware synths.
Filters are important in shaping the timbre of a sound, they can make a bright sound dull, or a harsh sound soft. If you use high resonance, they can also make something sound twangy. Think the TB-303.
You can also use LFOs and envelopes on filters too.
Let’s move onto the last one, effects.
Effects
If you’ve been making music for a while, you’d also probably know what effects are, but again, let me refresh it for you.
Effects are just that, effects. They can be applied on any sound. In sound design, they provide more tools in shaping a sound.
While some synthesizers don’t include them, it is still crucial.
There are many effects, but here are some of the most used in sound design :
- Compression
- Expansion
- Saturation
- Distortion
- Bitcrusher
- Flanger
- Phaser
- Chorus
- Unison
- Stereo Separation
- Delay
- Reverb
I’ll explain them.
Compression & Expansion
These two are kind of similar. Compression can decrease the dynamic range of a sound. Expansion on the other hand, is the opposite. It increases dynamic range instead. They can make a difference for certain sounds on certain genres.
Personally though, I have never used expansion and I’ve rarely used compression. Some producers use them a lot though, maybe there’s something I’m missing out on.
Saturation & Distortion
Saturation & distortion are kind of the same thing, they both give a sound more higher harmonics. Distortion however, is a much more intense version of saturation
They’re both useful on basses and other sounds to give them some extra grit, or to just give them some warmth.
Bitcrusher
Bitcrusher crushes the sound and downsamples it. It can give a sound an 8bit or lo-fi kind of sound.
I like using it to break sounds just to make them more edgy and dark.
Flanger & Phaser
Both flanger & phaser are very similar. They both make copies of a sound and delays them. Phaser however, is a more intense version of flanger.
They both could could give interesting results.
Chorus
Chorus is similar to flanger and phaser, it makes copies of a sound. Though in addition to delaying them, it also detunes them.
This is what the 80s sounds like.
Unison
Now unison isn’t actually available as a plugin, instead it’s actually built into most synths. It behaves similarly to a chorus, but it doesn’t delay the copies of the sound and you can also control how much copies it can make.
Stereo Separation
Stereo separation controls the volume of the mid and side channels. It can merge or separate those channels. Merging will turn a sound into mono, while separating will increase the volume of the side.
This and unison are what makes a supersaw sounds so full.
Delay
Delay makes delayed copies of a sound to give it echoes. Sometimes, they allow an effect to affect the sound over time.
They’re useful for filling up a mix, or just to give something a dreamlike feeling. Very prominent in trance genres.
Reverb
Reverb helps give a sound a sense of space. It can help something sound like it’s coming from a physical space. It can also help give a sound depth, or give it some tail if you want to.
Some reverb modules allow you to crank the decay crazily high. If you like ambient, you could give that a try.
Where to start?
That’s it for the tools, where should you start?
You can start by picking up a synth and try messing around or recreating some sounds. Any synth is fine, but the easiest one would be a subtractive synthesizer. It has all the things you need to create a sound. All without adding any more complexity than needed.
If you don’t know which subtractive synth to start on, I recommend Charlatan since it has a friendly interface for beginners and decent functionality. If you don’t think you’ll be overwhelmed by complex interfaces, Synth1 would be the best choice since it has even more functionality.
Side note : For Charlatan, don’t forget to change the voice mode in the global panel to “Poly” if you want to play more than one note.
We’re not done just yet though.
Experimentation
Before I end this post, I want to state that the most important thing is to experiment.
To fully understand sound design, you should try everything and every possible combination. It helps you get an idea of how something affects a sound. This in turn, will help you recognize which effects are used in which sounds.
Also, a good thing to do is to use a spectrum analyzer and study the sounds. It will aid in imitating or recreating sounds if you know how they look like with certain things applied to them.
Anyways, that’s all for this post. I hope it has benefited you somehow.
I am planning to do a video for a more hands-on practical approach towards synths. I’ll also go and try to create prominent sounds used in electronic music. If you want me to recreate something specific, do leave a comment and I’ll do my best to recreate them.
Until then, happy sound designing.
