What 92 Parameters Actually Mean — Sound Design Without Fear

92 sounds scary, but it is not

Most audio plugins take one of two approaches. Either they give you three knobs and a preset menu — simple, but you quickly hit a wall when you want something specific. Or they bury you in a dense interface where every parameter feels equally important and equally confusing. Neither of these ever felt right to me.

Ninety-two parameters spread across 14 modules is a different idea. It is not about complexity for its own sake. It is about having a complete set of tools available, organized into clear categories, so you can reach for exactly what you need. Think of it like a synthesizer. A good synth might have hundreds of parameters. Nobody touches all of them for a single patch. You use the oscillators, maybe a filter, an envelope or two. The rest is there for the day you need it — and that day always comes.

The synthesizer logic

I come from a background of loving synthesizers. What always drew me to them is the openness of the signal flow. Signal goes in, it passes through modules — oscillators, filters, envelopes, LFOs — and what comes out depends entirely on how you configure those modules. Everything is modulation. Everything is signal.

When I started designing Ferment, I tried to balance two things: the freedom of a modular synth and the practicality of a fixed signal chain for live performance. The result is a defined processing order — your signal flows through a specific sequence of modules — but within each module, the parameters give you enormous freedom. You are not locked into any particular sound. The chain keeps things manageable on stage, but the parameter ranges are wide enough that you can get to places most effects never reach.

On top of the fixed chain, three modulation sources — two LFOs and an envelope follower — run in parallel and can target 17 destinations across the modules. That is where the synthesizer thinking comes in: you are not just setting static values, you are creating movement, response, life in the sound.

The modules in Ferment

The signal chain runs through these stages, in this order. Each module can be bypassed individually — if you do not need it, it is simply not there.

• Pre-EQ — cleans up your input. Phase invert, high-pass to cut rumble, low-pass to tame harshness. The foundation before anything else touches the signal.
• Shape — waveshaper that morphs from soft saturation through hard clipping to wave folding and wrapping. With the envelope follower driving the drive parameter, it responds to how hard you play.
• Dynamics — compressor with lookahead, sidechain filtering, and program-dependent release. Evens out your dynamic range for live performance or adds punch and sustain as a creative tool.
• Multiply — this is the chorus, but not a typical one. Up to 8 voices, each with independent timing, detuning, and stereo placement. The Humanize parameter adds random drift in timing and level, so the voices behave more like real players than identical copies. The result is not just a wider sound — it is the feeling of multiple instruments, each with its own character.
• Octaver — pitch shifting up to ±12 semitones. Signal-adaptive, with accurate pitch tracking. Adds depth below or brightness above your natural range.
• Temporal — delay and echo. From short slap-back to long atmospheric trails, with tone shaping on the feedback loop and stereo spread between repeats.
• Spatial — reverb built on an 8-line feedback delay network. Room size, decay, damping, pre-delay, stereo width, and mid/side balance. This is where your instrument gets a sense of space — a room, a hall, an infinite cavern.
• Character — harmonic enhancement and correction. Body thickens or thins the low end, Tone shifts between dark and bright, and Harm applies pitch-relative harmonic shaping from your Material profile. This is where the instrument's tonal identity gets refined.
• Tube Blend — asymmetric soft-clipping that adds even harmonics. Like running the signal through a warm tube preamp. Subtle at low settings, full saturation at the top.
• Crackle — vinyl and tape noise texture. Adds character and grit when you want something that sounds less digital and more lived-in.
• Analog — console emulation. Subtle stereo decorrelation, noise floor, and soft saturation. The difference between a clean digital signal and something that has been through real hardware.
• Post-EQ — 7-band graphic equalizer for final tone shaping. Sub, Low, Low-Mid, Mid, Hi-Mid, Presence, Air. The last sculpting before the output.

Running alongside the chain: two LFOs and an Envelope Follower that can modulate 17 destinations across the modules. Want the reverb width to respond to how hard you play? Route the envelope follower to Spatial Width. Want the chorus depth to pulse slowly? Route LFO 1 to Multiply Mix. This is where the synthesizer thinking lives — movement and response, not just static settings.

A closer look at Multiply (the chorus)

Standard chorus takes your signal, detunes a copy slightly, and mixes it back in. The result is wider, but it still sounds like one instrument — just thicker. Multiply does something different. It splits your sound into up to 8 independent voices, each with its own timing offset, detune amount, and stereo position. Each voice gets slightly different EQ characteristics through the spread algorithm, so they do not all have the same tonal color. The Humanize control adds random drift — small variations in timing and level that make the voices behave less like copies and more like individual players in a section.

For strings, this is particularly powerful. A single violin through Multiply with 4-6 voices, moderate spread, and humanize turned up sounds remarkably like a small string section. Not because it is faking anything — but because it is doing what a section actually does: multiple instruments playing the same line with natural, human differences between them.

You do not have to touch all 92

This is the part that matters most. When I build a sound, I typically use three to five modules. The rest stay at their neutral settings — they are not doing anything, not adding anything, not getting in the way. They are simply waiting.

A warm ambient sound might use reverb, a touch of chorus, and some EQ shaping. That is maybe 20 parameters out of 92. The other 72 are irrelevant for that particular sound. But next week, when I want something completely different — maybe a tight, percussive tone with compression and saturation — those other modules are right there, ready to go.

I think of it like having a full palette of colors. You do not use all of them in every painting. Most of the time you work with three or four. But you would never want a palette with only three colors on it, because the day you need that specific shade of blue-green, you need it to be there. Having options does not create complexity. It creates freedom.

Learning by seeing

One thing I have found genuinely valuable is being able to see which parameters were changed when a preset is generated. When AI creates a sound from a text description — say, "warm reverb with subtle chorus for fingerpicked acoustic" — it does not just produce a result. It shows you exactly which modules were activated and which parameters were moved away from their defaults.

This connects to how AI helps musicians find their sound — not by replacing them, but by showing the way. This turns every generated preset into a small lesson. You can look at the result and think: "So that is what a warm reverb looks like in terms of decay time and high-frequency damping. And that is how much chorus depth actually creates that shimmer I was hearing." Next time, you know what to reach for. You can do it yourself, faster, with more confidence.

I think this is the most honest way AI can work in a creative tool. Not as a black box that does things for you, but as a transparent process that teaches you how sound works while it helps you get there. AI as a teacher, not a crutch.

Instrument profiles guide you

One thing that helps a lot with 92 parameters is that Ferment knows what instrument you are playing. When you create a Material profile — recording a short sample of your bass, your violin, your guitar through your specific signal chain — the plugin analyzes the spectral characteristics and adjusts how the parameters behave.

In Machine mode, you will see green arcs below each knob showing the recommended range for your instrument — the sweet spot where things sound musical. You can go outside that range (the arc turns orange), and sometimes that is exactly where the interesting sounds are. But the guidance is there when you want it, and it changes depending on your instrument and playing style.

Where Ferment fits in

This is the tool I built around these ideas. It is not the only way to approach sound design, but it is the one that made sense to me as a musician who wanted both freedom and speed.

Magic mode lets you describe a sound in plain language and generates a starting point across the 14 modules. It shows you every parameter it touched, so you learn as you go. It is the fast sketchpad — great for exploring ideas you cannot quite articulate technically yet.

Machine mode gives you full manual access to all 92 parameters. Every module, every knob, every detail. When you know what you want and want to sculpt it precisely, this is where you work. No limits, no simplification.

Both modes show you exactly what is happening. Nothing is hidden. Whether AI suggested the starting point or you built it from scratch, the parameters are right there, transparent and editable. That transparency is something I care about deeply — your sounds should always feel like yours.

Related articles

• How AI Helps Musicians Find Their Sound
• Building a Live Effects Rig with a Laptop
• Expression Pedal for Acoustic Instruments