Compressor (Native)

Parameters

Threshold — Level (in dB) where compression starts. Signal louder than this gets gain reduction; quieter signal passes through untouched. Lower threshold = more of the signal gets compressed (denser, more squashed). Set this by ear: pull it down until you see the gain reduction meter (GR) responding to the loudest parts but not chasing the quietest ones.

Ratio — How aggressively signal above the threshold is reduced. 4:1 means every 4 dB above threshold becomes 1 dB above; 20:1 acts like a soft limiter. Low ratios (2–4:1) produce gentle leveling, useful on busses and vocals. High ratios (10–20:1) clamp dynamics hard, useful on drum room mics and parallel-compression buses.

Attack — How fast the compressor responds once signal crosses the threshold, in ms. Fast attacks (0.1–5 ms) catch transients and squash peaks — useful for taming snare cracks or aggressive bass picks, but can dull the punch. Slow attacks (10–50 ms) let the initial transient through before clamping down on the body — preserves percussive impact while still controlling sustain.

Release — How fast the compressor lets go after the signal drops below threshold, in ms. Fast releases (10–80 ms) follow the signal envelope closely; can sound pumpy on busy material but lively on percussion. Slow releases (200–500+ ms) keep the gain reduction smooth and unobtrusive — better for vocals and busses where you don’t want the compressor to “breathe” audibly.

Knee — Width in dB of the soft-knee transition around the threshold. Hard knee (0 dB) — compression flips on instantly when threshold is crossed; can sound abrupt. Soft knee (6–12 dB) gradually engages compression near threshold for a smoother, more transparent character. Use 0–2 dB for aggressive pumping, 6+ dB for gentle bus glue.

Makeup — Post-compression gain in dB. Compression reduces peak level, so this brings the overall output back up to match (or exceed) the input. The right amount depends on how much gain reduction you’re applying — a typical rule is to add roughly the average GR amount as makeup, then fine-tune by ear. Pair with auto_makeup to let the node estimate this automatically.

Sidechain Hpf — Highpass filter on the detection signal (not the audio output), 20–500 Hz. The compressor will ignore low frequencies below this cutoff when deciding when to clamp down — the audio still passes through unaffected. Essential for bass-heavy material: setting SC HPF around 80–150 Hz prevents the kick or bass from triggering compression on every hit, which would otherwise pump the whole mix.

Dry/Wet — Equal-power blend between the dry signal (0) and the compressed signal (1). At 1 you hear pure compression; at 0.5 you hear classic parallel compression — the dry signal keeps its transients clean while a heavily-compressed copy adds density underneath. Default is 0.5 because parallel compression is the most useful starting point for K2K’s spectral workflows. Node-specific — the FX-slot version handles wet/dry at the mixer.

Additional controls

Detection Mode — How the compressor measures input level. Peak reacts to instantaneous peaks — fast and accurate, suited to controlling transients. RMS averages level over a short window — slower and more “ear-like,” follows perceived loudness rather than raw peaks. Peak is the default choice for surgical control; RMS for natural-sounding leveling on vocals, busses, and full mixes.

Character — Internal envelope and curve model. Clean — neutral digital response, transparent. VCA — fast, punchy, surgical (think SSL bus comp). Opto — slow, smooth, program-dependent (think LA-2A on vocals). FET — aggressive, colored, fast attack with character (think 1176 on drums). The animatable knob parameters work the same across modes; the character changes the feel of the gain reduction more than the math.

Auto Makeup — When on, the compressor estimates a sensible makeup gain based on threshold and ratio, so reducing threshold doesn’t make the signal quieter. Off lets you set makeup manually. Useful for set-and-forget tracking; turn off when fine-tuning a specific tonal balance.

Stereo Link — How the two channels’ detection is combined on stereo material. Dual — each channel detects independently (can pull the image around if one side is louder). Link — channels are summed for detection so both sides get the same gain reduction (preserves stereo image — most common choice). Max — both sides follow whichever is louder (most aggressive image preservation; can over-compress one side).

About Compressor (Native)

Native Compressor is K2K’s full-featured dynamics processor — threshold/ratio/attack/release/knee/makeup with sidechain HPF, peak/RMS detection, and four character models that emulate the classic hardware boxes: VCA (think SSL bus comp — fast, punchy, surgical, the “glue” choice), Opto (think LA-2A — slow, smooth, program-dependent, vocal-friendly), FET (think 1176 — fast attack, colored, aggressive on drums), Clean (transparent digital, no character flavoring). The character mode changes the feel of gain reduction (envelope shape, knee curve nuances) more than the underlying math — same threshold and ratio give different perceived character on the same source. sidechain_hpf is essential on bass-heavy material to stop the kick from triggering compression on every hit and pumping the whole mix. stereo_link preserves stereo image (Link, the standard) vs. independent per-channel detection (Dual, can pull image around) vs. max-follower (Max, most aggressive image preservation). Use as bus glue, vocal leveler, drum room compressor, or in parallel for pumping density. Compare with Native Gate for the inverse problem (silencing below threshold instead of catching above).

Generated 2026-05-05 from K2K_Dev@96730bdc by scripts/gen_lexique.py. Edit _intros/ or _overrides/, not this file.