Detector modes
> The Dynamic EQ on every band can use one of four detector algorithms. Each watches the signal differently and lands the gain reduction with a different character.
The Dynamic EQ on every band can use one of four detector algorithms. Each watches the signal differently and lands the gain reduction with a different character.
Modes
Peak
Feed-forward branching one-pole detector. Reacts to the instantaneous peak of the band-passed signal and lets the gain change happen fast. Surgical — best for de-essing, harsh-transient taming, mouth-noise control. Switch to Peak when you want the band to nail one sound and not the surrounding energy.
RMS
Windowed mean-square with classic compressor ballistics. Tracks the energy of the band rather than the peaks, so it ignores brief transients and reacts to sustained tone. Musical — best for drum-bus glue, vocal-bus compression, smoothing the body of a mix without ducking the snare.
Adaptive
Bipolar A/B-flipped adaptive-chase detector. Asymmetric and program-dependent: the chase speed itself reacts to the input, so the detector "leans into" loud passages and lets quiet passages bloom. Characterful — best for material that benefits from a hardware-like, slightly unpredictable hand on the gain. Try it on busses where you want movement, not control.
Spectral
Routes the band through the shared STFT engine for Soothe2-style per-bin resonance suppression. The time-domain detector is skipped entirely — the band's gain map is built in the frequency domain, and the spectral parameters (Selectivity, Ratio, Knee, Density, Mask) take over. Frequency-selective — best for resonance suppression that adapts to the program (resonances move, the detector follows them). See Spectral dynamics for the dedicated parameters.
Mode switching
Switching detector mode is always glitch-free: a 20 ms equal-gain crossfade in the dB domain bridges the old and new detector output, so you can A/B detectors live on a playing track without any clicks or pumping artefacts.
Tips
- If you don't know which mode to pick, leave it on Peak — every legacy preset starts there.
- Adaptive mode pairs well with the per-band character stages (Op-Amp, Density). It sounds at home in a colored chain.
- Spectral is more expensive than the time-domain modes; the STFT engine only runs while at least one band is set to Spectral. Stack as many Spectral bands as you like — they all share the same engine.
See also
Dynamic EQ
> Every band in Outboard-EQ can react to the audio passing through it. Set a threshold, and the band's gain bends only when the signal crosses it — a static cut that wakes up just for the loud notes,
Spectral dynamics
> When a band's detector mode is set to **Spectral**, the band stops behaving like a single gain reduction and starts working bin-by-bin in the frequency domain. Resonances get pulled down without the