Synth Programming — How Do You Design a Synth Sound?

Synth programming is using a synthesizer to create your sounds. Typically, the idea is to create sounds (beyond synth factory presets) for use in songs and sounds tracks.

Synth programming is a core skill for a producer, a recording engineer. It is a valuable skill for a composer, a songwriter... any creator works with sound.

The following fundamentals are essential for sound design with synthesizers.

The Fundamentals of Synth Programming and Sound Design

Let's look at the fundamentals of sound design so that no matter which synth or software you use you understand what's going on and start creating the sounds you want to create.

Basic Waveforms

Almost every complex synthesized sound stems back to one of the four basic waveforms:

1. The sine wave.
2. The triangle wave.
3. The square wave.
4. The saw-tooth wave.

Without getting too technical, let's have a brief look at each of these.

#1 — The Sine Wave

The sine wave is perhaps the simplest waveform. You may have seen this in maths or physics.

This waveform is very smooth.

It is particularly useful in bass sound design. If a lower it a couple of octaves, you can get a nice sub-wave.

Often, kick drums are reinforced with the sine wave.

If you look at a sine wave on a frequency analyzer, it's got one note — one harmonic.

#2 — The Triangle Wave

This waveform has pyramids with two sloped sides.

A triangle wave has got that main fundamental note but there're many more harmonics added, which gives it that slight buzzing sound.

It's harmonically richer than a sine wave.

#3 — The Square Wave

As the name suggests, this is a perfectly squared-off waveform.

There're many more harmonics added onto the fundamental note, so it is richer than the triangle wave.

There is a lot more buzz, especially on the top end.

The four and final waveform is the saw-tooth wave.

#4 — The Saw-tooth Wave

The saw-tooth wave can be mistaken for the triangle wave but it sounds very different.

There are saw-teeth in the waveform — only one slanted side.

Of all the waveforms we've seen so far, this is the richest harmonically.

It's very buzzy and full at the top end.

Different synthesizers have different ways of changing the waveshape in the oscillator.

There're different ways to select your basic waveform but the oscilloscope will show you the waveform when you make a selection.

If you record sounds for each of these waveforms, and put them on your playlist, you'll see that on the actual playlist, the waveforms are sines, triangles, squares, and saw-teeth.

Learn those four basic wave shapes, try to identify what the sound like, try playing them with single notes and chords to get used to them.

The next thing we need to look at is additive and subtractive synthesis.

Additive and Subtractive Synthesis

Additive Synthesis

The idea behind additive synthesis is that you start with one waveform, and then make it more complicated by bringing in more waves on top of it.

Say, for instance, you start with a square waveform and then add a sine wave to it, the tops of the square waves will be rounded off.

That would result in a combination between a sine and a square waveform.

If, say, you start with a sine wave and then introduce a saw-tooth wave to it, you combine the two while also adding buzziness.

This is the basic principle of additive synthesis.

Subtractive Synthesis

Now let's look at subtractive synthesis, which is the opposite of additive.

In subtractive synthesis, you start with the complex waveform and take away from it to create a new sound.

You can start with a saw-tooth wave, for instance, and put EQ and filter off the top end subtracting frequency away so that it's no longer as harsh on the top end.

Instead of having to use external services like EQs, many synthesizers have this built-in.

They have various filters and effects, which allow you to modify the sound from within the synthesizer itself.

The next thing we need to look at is unison.

Unison

This is a really quick way to completely change the sound.

Unison adds more voices to the sound. Instead of having just one saw-tooth wave buzzing, for instance, it duplicates, pans, and detune saw waves around the stereo field.

You can see this stereo information change in the vectorscope in Serum, for instance.

You can add unison to any of the basic waveforms.

That goes a long way into making the sound a lot more rich — more professional, if you like.

The next thing closely tied to unison is detuning.

While more voices of unison sound good when you start (it is common to add 16), definitely experiment.

Some of the best sounds have 7, or even just 3.

Adding too much unison can be resource-intensive on your computer, so be careful with that.

Detune

So far, we've created all these extra sounds and we've panned them with unison.

To make them sound thicker and fuller, you slightly detune them from the original voice.

You have to be careful not to overuse detune. At some point its starts sounding bad.

Filters

Almost every single synth has a filter — they all function in slightly different ways, but they all follow the same basic principle.

If you don't know what filter is you probably have heard of an EQ where you can remove frequencies from a sound, a filter does pretty much the same thing.

We have a good article that introduces what is EQ. Have a look if you haven't already.

Usually, you select a filter type. In that article, I explain the different types of filters.

So is you have your synthesizer's oscillator producing the sound and a low pass filter, moving the cutoff will allow the low frequencies to pass, and it's going to cut away the high frequencies.

As well as those normal filters I mentioned, most synthesizers will also have all sorts of crazy combing and phasing filters.

There're all sorts of different possibilities.

The next thing we'll look at is something I've mentioned extensively on the mixing blog — ADSR.

ADSR (Attack-Delay-Sustain-Release)

In addition to this, some synths (such as Serum) also have a value called hold.

ADSR determines the volume and shape of the sound.

Attack

If you have an incredibly short attack, the sound will jump up to its maximum value instantly. If you have a much longer attack, it's going to take more time for the sound to develop and grow to its maximum.

Release

Release determines how the sound decays once you release the keys or once the MIDI stops.

If you have a short release, the sound cuts away immediately. If you have a longer release the sound decays gradually and gently.

Usually, you can adjust the shape of the attack and release curves as well — you can fine-tune all of these using Serum, which is why I like using Serum.

The decay and sustain are closely related.

Sustain

After you reach the maximum value, the sustain is the value you want the sound to hold on.

Say, for instance, you want the sound to rise the top and then cut down to a certain value and then just sustain there whilst you're holding the keys.

The decay here would be about how long it takes to decay from that top value to where it's going to hold.

Adjusting the unison, filter, and ADSR is where you create the majority of your sound — at least 80 or 90 percent of it.

If instead of having sustained sounds you adjust the attack and release to stay short, you end up nice plucked chord sounds.

LFOs (Low-Frequency Oscillators)

This is where people quite often go quite wrong when starting.

Low-frequency oscillators doesn't mean that it's affecting the low end of a sound (the bass), it means that it's oscillating in relatively low frequencies.

We can have a range of say, 0 to 30 or 40 Hz — something that we can recognize, and then the LFO is going to keep going backward and forward.

LFO adds movement and automation to most (if not all) parameters on the synth.

In Serum, LFO can be used on any parameter, but it's usually used on the filter.

Instead of having the filter stay static, you assign an LFO to the filter and then have that cut off move all the time.

LFOs are great because to achieve this sort of modulation or automation without them, you would usually have to create an automation clip on your playlist and assign points and then link the two controls together.

This can be done, but using an LFO saves you an awful lot of time.

In Serum, you don't even have to have an LFO follow a shape. You can set any sort of shape you want — which is awesome for sidechaining or any sort of sound design.

FX Processors

The final thing I want to look at is adding extra effects to this because sound design doesn't just stop with the synth engines.

In your DAW or the synth, there is usually an effects chain where you can add distortion, EQ, delay... all sorts of stuff including... you guessed it REVERB!

It's easy to add too many effects to a sound.

I like to keep things simple.

In Conclusion

That is it for this simple introduction to sound design.

The key takeaway is that no matter which synth you use (Serum, 3x Osc, even an analog synth), you should now start recognizing what some key sound design terms are, how they are linked, and what they do to the sound.

I want to make more guides like this. I want to keep getting more and more into these topics, so drop your email in the subscription form below to keep up with more helpful music production tips.

That's it for this article.