Introduction to Synthesis Methods

Electronic music production relies on various synthesis techniques to shape sound. Among the most foundational approaches are physical modeling and subtractive synthesis. While both serve the same ultimate purpose—creating audio that can be arranged into music—they differ dramatically in philosophy, complexity, and sonic character. Understanding their inner workings, strengths, and limitations is critical for producers, sound designers, and engineers who want to make informed decisions in the studio or on stage.

Subtractive synthesis has been a staple since the early days of analog synthesizers, while physical modeling emerged later as digital processing power increased. Each method offers a unique palette of textures, and many modern instruments blend them. This article provides a comprehensive comparison, with detailed pros and cons, to help you determine which approach (or combination) suits your creative workflow best.

What Is Physical Modeling?

Physical modeling synthesis replicates sound by simulating the physical processes that occur in real acoustic instruments. Instead of using static waveforms or samples, a physical model generates audio in real time based on mathematical equations that describe how a physical object vibrates, resonates, and radiates sound.

How Physical Modeling Works

A physical model typically comprises several interconnected components: an excitation source (like a mallet, bow, or breath), a resonator (such as a string, tube, or membrane), and a radiation system (the air or soundboard). The algorithm solves differential equations that model these elements’ behavior. For example, a bowed string model might simulate the friction between bow and string, the string’s stiffness, and the body’s resonant modes. Changes in parameters like bow pressure, bow position, or string damping directly affect the sound in a way that mimics real playing techniques.

Examples of Physical Modeling Instruments

  • Modal Synthesis – Models individual resonant modes of an object (e.g., a bell, a drumhead). Used in software like AAS Chromaphone.
  • Waveguide Synthesis – Simulates wave propagation in a medium (string, tube). Used in Yamaha VL1 and many software synths.
  • Digital Waveguide – A specific type of waveguide synthesis widely used for realistic plucked strings and woodwinds.
  • Mass-Spring Models – Simulate complex interactions of masses and springs, often for non-linear sounds like rubbing or scraping.

Key Characteristics of Physical Modeling

  • Dynamic Expression: Parameters like attack, bow pressure, or embouchure can be modulated continuously, creating evolving timbres.
  • Realism: Can replicate subtle artifacts of acoustic instruments, such as sympathetic string vibrations or body resonance.
  • Computational Cost: High CPU usage, especially with complex models or polyphony.
  • Programming Complexity: Requires understanding of physics concepts; not as tactile as manipulating a simple filter cutoff.

Notable software synthesizers that rely heavily on physical modeling include Applied Acoustics Systems Chromaphone, Steinberg The Grandeur (piano modeling), and Korg OASYS (hybrid engine).

What Is Subtractive Synthesis?

Subtractive synthesis is one of the most intuitive and widely used synthesis methods. It starts with a harmonically rich waveform (such as a sawtooth, square, or pulse wave) generated by an oscillator. The core idea is to subtract frequencies using filters, amplifiers, and modulation envelopes to shape the final sound.

How Subtractive Synthesis Works

A classic subtractive synthesizer signal flow includes oscillators that produce raw waveforms, a mixer, a voltage-controlled filter (VCF), a voltage-controlled amplifier (VCA), and modulation sources (envelopes and LFOs). The oscillator generates a sound with many overtones. The filter removes specific frequency bands—most commonly using a low-pass filter to cut high frequencies, but also high-pass, band-pass, or notch filters. An envelope (ADSR) controls how the filter cutoff and amplitude change over time, giving the sound shape and motion.

Examples of Subtractive Synthesis Instruments

  • Analog Classics: Minimoog Model D, Roland SH-101, ARP 2600.
  • Software Emulations: Arturia Mini V, u-he Repro-1, Diva (hybrid analog modeling).
  • Digital Subtractive: Ableton Analog, Native Instruments Massive (wavetable-based subtractive approach).

Key Characteristics of Subtractive Synthesis

  • Ease of Learning: The basic concept (start with a rich sound, then carve it) is simple and visually intuitive.
  • Low CPU Usage: Compared to physical modeling, subtractive synthesis is computationally light, even with many voices.
  • Versatility: Can create basses, leads, pads, effects, and percussive sounds by adjusting the waveform, filter, and modulation.
  • Character: Often produces a warm, analog-style sound due to the filtering of harmonically rich waveforms.
  • Limitation in Acoustic Realism: Hard to convincingly emulate acoustic instruments like a violin or grand piano because subtractive synthesis lacks the complex behavior of physical models.

Subtractive synthesis is the foundation of countless classic hardware and software synths. For a comprehensive introduction, see Sound on Sound’s guide to subtractive synthesis.

Pros and Cons of Physical Modeling

Advantages of Physical Modeling

  • Unmatched Realism: Physical modeling excels at reproducing the nuanced behavior of acoustic instruments. The interaction between bow and string, air column length, and body resonance can be modeled with remarkable accuracy.
  • Expressive Control: Many physical parameters (e.g., bow pressure, wind speed, string damping) can be modulated in real time using MIDI controllers, breath controllers, or touch surfaces, resulting in highly expressive performances.
  • No Static Loops: Unlike sample-based instruments, physical models do not suffer from loop points or velocity layers. Every note is generated afresh, with variations depending on how the performer plays.
  • Unique Sound Design Potential: By breaking the laws of physics (e.g., making a string infinitely long or altering the speed of sound), physical modeling can create otherworldly sounds that retain organic unpredictability.
  • Low Storage Requirements: Physical models are algorithms, not sample libraries. A comprehensive physical modeling synthesizer can occupy just a few megabytes yet produce an infinite variety of sounds.

Disadvantages of Physical Modeling

  • High Computational Demand: Real-time physical simulation is processor-intensive. Polyphonic instruments with complex models can max out CPU cores, limiting real-time performance, especially on older hardware.
  • Steep Learning Curve: Understanding the physical parameters (stiffness, damping, material density, etc.) requires some knowledge of acoustics and mathematics. This can be intimidating for beginners.
  • Limited Palate for Classic Synth Sounds: Physical modeling is not designed for creating standard subtractive bass or lead sounds. Its strengths lie in acoustic emulation and experimental textures, not in replicating the classic analog synth aesthetic.
  • Potential for Unstable Sounds: Some parameter combinations can cause the model to become unstable (e.g., feedback loops in waveguides), producing chaotic or unwanted noise. While sometimes desirable for experimental music, it can be a problem for conventional uses.
  • Requires Specialized Hardware or Plugins: Although many modern DAW plugins offer physical modeling, the most advanced implementations (like piano modeling) often come in dedicated software that may require significant investment.

Pros and Cons of Subtractive Synthesis

Advantages of Subtractive Synthesis

  • Intuitive and Quick: Load a saw wave, open a filter, add an envelope—you have a usable sound in seconds. This direct workflow is ideal for fast ideation and live performance tweaking.
  • Low CPU Impact: Even dozens of voices using simple waveforms and filters place little strain on a modern CPU. This makes subtractive synthesis perfect for dense arrangements and live setups.
  • Huge Variety of Sound: With different waveforms, filter types, modulation routings, and effects, subtractive synthesis can create an enormous sonic range: from massive basses to airy pads, from sharp leads to evolving noise textures.
  • Well-Established Ecosystem: Countless hardware and software instruments are based on subtractive synthesis. Preset libraries, tutorials, and community knowledge are abundant, making it easy to learn and share.
  • Analog Warmth: Even in digital emulations, subtractive synthesis can produce pleasing harmonic saturation and non-linearities that mimic vintage analog circuits, giving a rich, musical character.

Disadvantages of Subtractive Synthesis

  • Limited Acoustic Realism: Subtractive synthesis cannot convincingly emulate the complexity of physical instruments. It can approximate a plucked string or a brass-like tone, but the results lack the dynamic nuances of real physics.
  • Tendency Towards Synthetic Timbre: Without careful modulation, subtractive sounds can feel static or “electronic,” which may be undesirable for genres requiring organic textures (e.g., orchestral, folk, ambient soundscapes).
  • Need for Extensive Tweaking for Complex Sounds: Creating evolving, multi-part sounds often requires layering multiple voices, using complex modulation routing, and programming detailed automation. This can be more time-consuming than using a physical model or wavetable synth for similar results.
  • Waveform Static: Traditional subtractive synthesis uses fixed oscillator waveforms (saw, square, triangle, sine). While filters alter the harmonic content, the fundamental shape remains constant. Wavetable synthesis partially solves this, but classic subtractive lacks the evolving spectral variety of physical modeling.

Choosing Between Physical Modeling and Subtractive Synthesis

Consider Your Genre and Goals

If your music requires realistic acoustic instrument sounds (piano, strings, brass, woodwinds), physical modeling is almost mandatory for achieving convincing expressiveness. Pianists using modeling software like Pianoteq or Modartt’s instruments can achieve a dynamic range and response that even high-end sample libraries struggle to match. For electronic genres like techno, house, or synthwave, subtractive synthesis is a better fit because it provides classic analog textures and quick sound design.

Consider Your Workflow and Learning Style

Subtractive synthesis rewards a more experimental, hands-on approach. Twisting filter cutoff and resonance knobs yields immediate and pleasing results, which can be very satisfying for beginners. Physical modeling demands a more analytical mindset; you need to understand how parameters like bow velocity, string stiffness, and body material interact. If you enjoy deep engineering and are willing to invest time in learning the science, physical modeling can be incredibly rewarding.

Consider Hardware and Resource Constraints

For live performance, subtractive synthesis is generally more reliable because it uses fewer CPU resources. Many hardware synthesizers (e.g., Moog Subsequent 37, Dave Smith Instruments OB-6) are subtractive and offer immediate tactile control. Physical modeling instruments suitable for live use exist (e.g., Yamaha Montage, Korg Kronos with their modeling engines), but they often require more powerful hardware or careful voice management. If you are working on a laptop with limited processing power, subtractive plugins will allow you to run many instances without glitching.

Consider the Sound You Can’t Get Elsewhere

Physical modeling excels at producing sounds that are alive, breathing, and unpredictable. If you want a synth that evolves with every note, responds to subtle controller changes, and can mimic the feel of a real instrument, physical modeling is the way to go. Subtractive synthesis gives you immediate access to the classic electronic sound palette—think Reese basses, Hoover leads, string machine pads. Neither is inherently better; they are different tools for different jobs.

Combining Syntheses: Hybrid Approaches

Modern synthesizers and plugins increasingly blur the line between physical modeling and subtractive synthesis. For example, Ableton Live includes Collision, a physical modeling instrument, and Analog, a subtractive synth, allowing users to layer them. Some independent products, like u-he Diva, use analog circuit modeling (which is a form of physical modeling) combined with classic subtractive architecture. The Korg Prologue includes a digital oscillator capable of physical modeling alongside its analog VCOs.

Hybrid synths offer the best of both worlds: you can use a physical model for the attack and body of a sound, and then process it through subtractive filters and envelopes to sculpt the tone further. This approach is especially powerful for creating unique textures that no single method could achieve.

Practical Tips for Exploring Both Methods

  • Start with a subtractive synth if you are new to synthesis. Learn the basics of oscillators, filters, and envelopes. Online resources like Ableton’s Learning Synths provide interactive tutorials.
  • Once comfortable, try a free physical modeling plugin such as Plogue Chipcrusher (though more of a lo-fi effect) or Airwindows Synth. More comprehensive demos of commercial products like Applied Acoustics Systems Lounge Lizard (electric piano model) can also be trialed.
  • Experiment with layering: Use a physical modeling patch for the body of a sound and a subtractive patch for the sustain or harmonic richness. This can produce incredibly full and complex timbres.
  • Explore modulation: Both synthesis types benefit from deep modulation. In physical modeling, assign aftertouch to bow pressure; in subtractive, map a slow LFO to filter cutoff for movement.
  • Read the manual. Physical modeling synths often have non-obvious parameters. Understanding the documentation can save hours of frustration and unlock creative possibilities.

Conclusion

Physical modeling and subtractive synthesis represent two fundamentally different approaches to sound creation. Physical modeling excels at replication and organic expression, while subtractive synthesis offers accessibility, versatility, and classic electronic character. The choice between them depends on your musical context, performance needs, and willingness to dive into technical depth.

Neither method is obsolete or superior; they complement each other. Many of the most compelling modern sound designs emerge from a hybrid workflow that uses both techniques. By understanding the pros and cons outlined in this article, you can better navigate the vast landscape of synthesizers and select the tools that will bring your sonic vision to life.

Whether you are sculpting a bass line in a subtractive synthesizer or modeling a resonant flute, the ultimate goal is the same: to express yourself through sound. Mastering both methods will make you a more versatile and effective music producer, capable of crafting anything from realistic acoustic emulations to otherworldly electronic textures.