Drive and compression: energy and density
Two effects that get confused because both make a sound feel bigger. Drive adds harmonics to the waveform; compression evens out loudness across time.
Drive and compression both make a sound feel bigger, which is why beginners mix them up. They work on different axes. Drive reshapes the waveform inside each cycle, which adds harmonics and changes tone. Compression reshapes loudness across time, which evens out dynamics and changes density. Both can raise perceived loudness without raising peak level by much, which is part of why they get confused: drive adds energy into upper harmonics, and compression pushes the average level closer to the peaks. Learning to hear which axis moved is the fastest way to understand a mix you did not make.
Drive: clipping as a harmonics factory
Push a waveform's peaks against a ceiling and they flatten. A flattened peak is a changed shape, and lesson 2 already told you what a changed shape means: new harmonics. The toy's drive stage works by tanh saturation, a smooth curve that bends a signal's peaks toward a ceiling instead of chopping them off flat. Because the curve is symmetric — it treats positive and negative peaks the same way — it only ever adds odd harmonics: the 3rd, 5th, 7th, and on up, the same family lesson 2 assigned to the square wave. Enough drive and a sine approaches a square, which is exactly the shape a symmetric saturator is squeezing it toward.
The drive slider runs from 1 to 25, but the bend is not even across that range: by the default setting of 6 the curve has already bent most of the way to a hard step, and pushing on toward 25 mostly narrows the rounded corners left near the zero crossing rather than reshaping the curve wholesale. Most of the audible change happens low on the dial. Notice, too, what does not happen as drive increases: the sound does not get louder. The curve is normalized so its output always reaches the same ceiling no matter how hard the input drives into it, the same principle behind the makeup gain on a dedicated saturator plugin. What changes is how quickly the signal reaches that ceiling and how much of each cycle spends its time pinned there — which is why more drive reads as brighter and more aggressive rather than louder. Bit reduction is drive's digital cousin: instead of bending the curve, it removes amplitude steps, so a smooth wave turns into a staircase. The staircase's sharp corners spray their own wide, inharmonic buzz, which is why bit reduction sounds colder and grainier than saturation even at a similar intensity.
A clean sine (left) through the toy's tanh curve at its default drive of 6 (centre), and what comes out the other side (right). The dashed diagonal is the curve at zero drive; the solid curve shows how far it has already bent by the default setting, flattening the sine's peaks while leaving its zero crossings steep.
Compression: an automatic volume hand
- Threshold — the level, in dB, where the compressor starts working. Signal below it passes untouched. The toy's threshold runs from -60 dB up to 0 dB, with a default of -24 dB.
- Ratio — how hard it pulls down whatever crosses the threshold. At 4:1, every 4 dB a signal goes over the threshold comes out as only 1 dB over. The toy's ratio runs from 1:1 (no effect) up to 20:1 (a hard limit); real hardware offers that same 4:1 as one of eight fixed ratio choices.
- Attack — how quickly the compressor clamps down once a signal crosses the threshold. The toy fixes this at 3 ms, fast enough to catch a kick drum's transient almost from its first millisecond.
- Release — how quickly it lets go once the signal drops back below the threshold. The toy fixes this at 250 ms, on the order of half a beat at the groove's 112 BPM tempo: fast enough to recover before the next kick but slow enough to still be releasing when it lands, which is what produces the pumping on the reduction meter.
Set threshold to -18 dB and ratio to 4:1 and you can work out exactly what the compressor does to a loud hit. A kick that peaks at -6 dB is 12 dB over the threshold. At 4:1, only a quarter of that overage survives — 3 dB — so the kick comes out at -18 + 3 = -15 dB, a 9 dB reduction. The reduction meter reads exactly that gap: how far below the uncompressed signal the compressor is currently pulling the output. Watch it on the kick-and-hat groove and it should duck hard on every kick and recover before the next one; if it never fully recovers, the ratio or threshold is set too hard for the material. Compression by itself only makes a sound quieter. The reason it gets used to make things sound bigger is what happens next: with peaks reliably tamed, the whole track can be turned back up, which brings the quiet parts up along with it. That is compression's real product: it shrinks the gap between the loudest moments and the quietest ones.
Input level against output level, threshold -18 dB and ratio 4:1 — the worked example from the text. Below threshold the line runs 1:1; above it, every 4 dB in becomes 1 dB out. The marked point is the -6 dB kick: it comes out at -15 dB, a 9 dB gain reduction. The rounded corner is the toy's fixed 6 dB knee.
An internal kick-and-hat groove runs through drive, then a compressor. A/B each stage with its own bypass button: drive changes the tone of the kick even at the same loudness; compression changes how the loud and quiet hits relate to each other. Watch the reduction meter duck with every kick — that dip is the compressor working, whether or not you can currently hear it in bypass.
Digitone II and Digitakt II carry the harmonics factory per track on the FX page (BR, OVER, SRR) and a pattern-wide compressor on the COMP page, where THR, RAT, ATK, and REL are all four knobs under your fingers instead of the toy's fixed pair. Syntakt's drive is a real analog circuit, the only parameter on its FX track's SYN page. Tonverk turns both into FX machines of their own — DEGRADER and COMPRESSOR — with the same eight ratio choices, 1.5:1 up to 20:1, as the Digi-series COMP page: