Table of Contents
Table of Contents
Table of Contents
Table of Contents
Table of Contents
Synthesis is the art and science of creating sound electronically. Every synth you have ever heard on a record uses one of four fundamental methods: subtractive, FM, wavetable, or granular. Understanding these core types unlocks the ability to create any sound from scratch rather than scrolling through presets hoping to find something that works.
Table of Contents
- Subtractive Synthesis: The Foundation
- FM Synthesis: Complex and Metallic Tones
- Wavetable Synthesis: Evolving and Moving Sounds
- Granular Synthesis: Textural and Atmospheric
- Practice Plan for Learning Synthesis
Key Takeaways
- Subtractive synthesis uses filters to carve away frequencies from harmonically rich waveforms
- FM synthesis creates complex, metallic, and bell-like tones through frequency modulation
- Wavetable synthesis scans through a table of waveforms to create evolving sounds
- Granular synthesis chops audio into tiny grains for textural and atmospheric effects
- Start with subtractive, then expand to the other types as you understand the fundamentals
Subtractive Synthesis: The Foundation
Subtractive synthesis is the most widely used synthesis method and the best place to start. The concept is simple: start with a harmonically rich waveform and use filters to remove frequencies you do not want. Think of it like starting with a block of marble and chiseling away to reveal the sculpture inside.
Oscillators generate the raw waveforms. The four basic waveforms are sine (pure fundamental, no harmonics), triangle (odd harmonics, soft sound), sawtooth (all harmonics, bright and buzzy), and square (odd harmonics only, hollow sound). A sawtooth wave is the most common starting point because it contains the full harmonic spectrum.
The filter is the heart of subtractive synthesis. A low-pass filter (LPF) allows frequencies below the cutoff point to pass through while reducing frequencies above it. As you lower the cutoff frequency, the sound becomes darker and more muffled. Resonance boosts the frequencies around the cutoff point, creating a sharp peak that can make the filter itself sing. High-pass filters (HPF) do the opposite, letting high frequencies through while cutting lows.
Envelopes shape how parameters change over time. An ADSR envelope has four stages: Attack (time to reach maximum level), Decay (time to drop to the sustain level), Sustain (level held while a key is pressed), and Release (time to fade to zero after the key is released). A sharp attack on the filter envelope creates a pluck sound. A slow attack creates a swelling pad.
LFOs (Low-Frequency Oscillators) create cyclic modulation. An LFO applied to pitch creates vibrato. Applied to filter cutoff, it creates a wobble effect. Applied to volume, it creates tremolo. LFOs typically run below 20 Hz and are synchronized to the tempo in modern DAWs.
To create a classic subtractive bass: start with a sawtooth wave, apply a low-pass filter with the cutoff around 200 Hz, add a quick filter envelope with a short decay, and finish with a tiny bit of resonance. This is the foundation of countless house, techno, and dubstep bass sounds.
FM Synthesis: Complex and Metallic Tones
Frequency Modulation synthesis uses one waveform to modulate the frequency of another. This produces incredibly complex harmonics that would be difficult or impossible to achieve with subtractive synthesis. FM was the defining sound of 1980s pop, thanks to the Yamaha DX7, and it remains essential for modern electronic music.
Operators are the building blocks of FM synthesis. Each operator is a simple sine wave generator with its own ADSR envelope. In a basic FM patch, one operator acts as the carrier (the sound you hear) and another acts as the modulator (which modulates the carriers frequency). The ratio between the carrier and modulator frequencies determines the harmonic content. Integer ratios (1:1, 2:1, 3:1) produce harmonic sounds. Non-integer ratios produce inharmonic, clangorous tones.
Modulation index controls the intensity of the frequency modulation. A low modulation index produces subtle, warm harmonics. A high modulation index creates bright, complex, and potentially dissonant tones. The modulation index can be controlled by an envelope, allowing the timbre to evolve over time. This is how FM creates those distinctive evolving bell and electric piano sounds.
Common FM algorithms arrange operators in different configurations. A simple two-operator FM can produce convincing bell tones, brass, and bass sounds. More complex algorithms with four or six operators can create layered pads, evolving textures, and aggressive leads. The classic DX7 used six operators arranged in 32 different algorithms.
For your first FM patch, try a simple two-operator setup with a 1:2 carrier-to-modulator ratio. Set the modulator envelope to have a sharp attack and medium decay while the carrier has a slower attack. Raise the modulation index gradually and listen to how the timbre shifts from pure sine to complex metallic. This is the basis of classic FM electric piano and bell sounds.
Wavetable Synthesis: Evolving and Moving Sounds
Wavetable synthesis uses a table of single-cycle waveforms that can be scanned through over time. Moving through different waveforms creates sounds that continuously evolve, making wavetable synthesis exceptionally good for bass sounds that change character and pads that breathe and move.
Wavetables are collections of waveforms arranged in sequence. A single wavetable might contain 64 or 128 individual waveforms, each slightly different from the last. The oscillator reads through these waveforms as it plays, creating a constantly shifting timbre. The position within the wavetable can be modulated by envelopes, LFOs, or note velocity.
Morphing between waveforms is what gives wavetable synthesis its signature sound. Wavetable position can be modulated slowly for evolving pads or rapidly for aggressive wobble bass sounds. Serum and Massive are the most popular wavetable synths, with Serum being particularly known for its high-quality wavetable editing and visual feedback.
Wavetable synthesis versus subtractive: In subtractive synthesis, the waveform is fixed and you shape the sound with filters. In wavetable synthesis, the waveform itself changes over time, creating harmonic movement that filters alone cannot achieve. Many modern synths combine both approaches, using wavetable oscillators with subtractive-style filters and envelopes.
To create a classic wavetable bass, start with a wavetable that moves from a smooth sine-like wave to a harsh sawtooth, apply a low-pass filter, and modulate the wavetable position with an envelope that opens after the initial attack. The bass will start warm and clean, then grow aggressive and distorted as the note sustains.
Granular Synthesis: Textural and Atmospheric
Granular synthesis takes a different approach entirely. Instead of generating sound from waveforms, it chops audio into tiny grains typically 1 to 100 milliseconds long and plays them back in various arrangements. The result is a dense, shifting texture that can sound like anything from a gentle pad to a chaotic glitch storm.
Grains are the basic units of granular synthesis. Each grain is a short slice of audio with its own amplitude envelope (typically a windowing function that fades in and out to avoid clicks). Parameters include grain size, density (how many grains play per second), pitch, pan position, and playback position within the source audio.
Time stretching is one of the most practical applications of granular synthesis. By controlling grain density and playback position independently, granular processors can stretch audio to extreme lengths without changing pitch. A one-second drum hit can become a 30-second atmospheric texture. This technique is widely used in ambient, experimental, and film music.
Granular effects can transform any audio source. Load a vocal sample into a granular processor, set the grain size to 50 ms with high density and random pitch variation, and you get ethereal vocal clouds. Process a guitar strum with small grains and randomized position, and it becomes a shimmering cascade of fragments.
For your first granular experiment, take a short audio clip (even just a single chord or drum hit), load it into a granular processor like Granulator II or The Mangle, set grain size to 40 ms, density to 30 grains per second, and slowly modulate the playback position. The result is a rich, evolving texture that bears little resemblance to the original sound.
Practice Plan for Learning Synthesis
| Week | Focus | Exercise |
|---|---|---|
| 1 | Subtractive fundamentals | Create a bass, lead, and pad using only one oscillator, a low-pass filter, and an ADSR envelope |
| 2 | Filter and envelope mastery | Recreate the same sound with three different filter types (LPF, HPF, BPF) and different envelope shapes |
| 3 | LFO modulation | Create a wobble bass by modulating filter cutoff with a tempo-synced LFO at different rates and waveforms |
| 4 | FM basics | Build three FM patches: a bell, a brass stab, and a bass using 2-operator FM |
| 5 | Wavetable exploration | Scan through a wavetable while holding a chord, create an evolving pad that shifts timbre over 8 bars |
| 6 | Granular texture | Take a drum loop and transform it into a 60-second atmospheric texture using granular processing |
Frequently Asked Questions
Which synthesis type should I learn first?
Start with subtractive synthesis. It is the most intuitive, widely used, and well-documented. Most hardware and software synthesizers are based on subtractive principles. Once you understand oscillators, filters, and envelopes, learning the other synthesis types becomes much easier because you already understand the fundamental concepts.
Do I need a hardware synthesizer to learn synthesis?
Not at all. Every major DAW includes built-in software synthesizers that are more than capable. Ableton Live has Operator (FM) and Wavetable. FL Studio has Sytrus and Harmless. Logic Pro has Alchemy and ES2. Free options like Vital (wavetable) and Dexed (FM) are world-class tools used by professional producers.
How long does it take to learn sound design?
You can learn the fundamentals in about 2 to 4 weeks of dedicated practice. Basic proficiency where you can create the sounds you need for a track typically takes 3 to 6 months. Mastery is a lifelong journey. Experienced sound designers often spend years refining their understanding of a single synthesis type.
Conclusion
The four core synthesis types are subtractive, FM, wavetable, and granular. Each offers a different approach to sound creation, and each excels in different contexts. Subtractive gives you the fundamental building blocks. FM provides complex, evolving harmonics. Wavetable delivers morphing, animated textures. Granular creates vast, atmospheric soundscapes. Start with subtractive, build your foundation, then explore the other types as your sound design vocabulary grows.