Filter & Amp Sections

Filter Section

Within each Machine that uses the standard signal chain, the sound source is always routed through a similar Filter section, available on Tab 3. Each Filter section then goes to an Amp section before reaching the Mixer.

The Filter tab has 3 pages of controls. When the tab header shows small bar icons, click the corresponding button below to jump between pages.

Main screen of the Filter section

Filter section main screen

Page 1
Filter 1 FrequencyFilter 1 Resonance(Filter Morph) -or- (Filter Gain)Filter 1 Type
Cutoff frequency of the filter.Resonance amount of the filter.Morph between the filter types, from Low-pass to Notch to High-pass. Only available when Filter 1 Type is set to SVF 12. -or- EQ gain. Only available when Filter 1 Type is set to BELL EQ.Selects a filter model and slope. See the Filter types reference below for details. This parameter cannot be modulated.
Page 2

Same as page 1, but for Filter 2. Available only when Routing on page 3 is set to anything other than Single.

Page 3 — routing

Filter routing page

RoutingBalance F1/F2--
Filter routing configuration: - Single — only Filter 1 is enabled. - Serial — Filter 1's output feeds Filter 2. - Para — both filters run in parallel on the same input. - Split — the Machine's two sources are sent independently into Filter 1 and Filter 2 (see Input reference below).Mix between the two filter outputs. Fully clockwise = only Filter 2; fully counter-clockwise = only Filter 1.--

Filter routing diagrams

Single filter
flowchart LR
    I1["INPUT 1"]
    I2["INPUT 2"]
    SUM(("+"))
    F1["FILTER 1"]
    OUT["TO DRIVE & AMP"]

    I1 --> SUM
    I2 --> SUM
    SUM --> F1
    F1 --> OUT

    classDef in fill:#06b6d4,stroke:#22d3ee,color:#fff,stroke-width:2px
    classDef sum fill:#1f2937,stroke:#9ca3af,color:#fff
    classDef filter fill:#65a30d,stroke:#a3e635,color:#fff,stroke-width:2px
    classDef mixer fill:#9333ea,stroke:#c084fc,color:#fff,stroke-width:2px

    class I1,I2 in
    class SUM sum
    class F1 filter
    class OUT mixer
Serial filters
flowchart LR
    I1["INPUT 1"]
    I2["INPUT 2"]
    SUM(("+"))
    F1["FILTER 1"]
    F2["FILTER 2"]
    OUT["TO DRIVE & AMP"]

    I1 --> SUM
    I2 --> SUM
    SUM --> F1
    F1 --> F2
    F2 --> OUT

    classDef in fill:#06b6d4,stroke:#22d3ee,color:#fff,stroke-width:2px
    classDef sum fill:#1f2937,stroke:#9ca3af,color:#fff
    classDef filter fill:#65a30d,stroke:#a3e635,color:#fff,stroke-width:2px
    classDef mixer fill:#9333ea,stroke:#c084fc,color:#fff,stroke-width:2px

    class I1,I2 in
    class SUM sum
    class F1,F2 filter
    class OUT mixer
Parallel filters
flowchart LR
    I1["INPUT 1"]
    I2["INPUT 2"]
    SUM(("+"))
    F1["FILTER 1"]
    F2["FILTER 2"]
    MIX(("MIX"))
    OUT["TO DRIVE & AMP"]

    I1 --> SUM
    I2 --> SUM
    SUM --> F1
    SUM --> F2
    F1 --> MIX
    F2 --> MIX
    MIX --> OUT

    classDef in fill:#06b6d4,stroke:#22d3ee,color:#fff,stroke-width:2px
    classDef sum fill:#1f2937,stroke:#9ca3af,color:#fff
    classDef filter fill:#65a30d,stroke:#a3e635,color:#fff,stroke-width:2px
    classDef mixer fill:#9333ea,stroke:#c084fc,color:#fff,stroke-width:2px

    class I1,I2 in
    class SUM,MIX sum
    class F1,F2 filter
    class OUT mixer
Split filters
flowchart LR
    I1["INPUT 1"]
    I2["INPUT 2"]
    F1["FILTER 1"]
    F2["FILTER 2"]
    MIX(("MIX"))
    OUT["TO DRIVE & AMP"]

    I1 --> F1
    I2 --> F2
    F1 --> MIX
    F2 --> MIX
    MIX --> OUT

    classDef in fill:#06b6d4,stroke:#22d3ee,color:#fff,stroke-width:2px
    classDef sum fill:#1f2937,stroke:#9ca3af,color:#fff
    classDef filter fill:#65a30d,stroke:#a3e635,color:#fff,stroke-width:2px
    classDef mixer fill:#9333ea,stroke:#c084fc,color:#fff,stroke-width:2px

    class I1,I2 in
    class MIX sum
    class F1,F2 filter
    class OUT mixer
Input reference (per machine type, in Split routing)
MachineInput 1Input 2
SynthesizerOscillator 1Oscillator 2
Sample PlayerSample
GranularGranular cloud

The Resonator, Simple Oscillator and Drum Synth Machines do not use this unified Filter section. The Resonator and Drum Synth each have their own filter stage built into the Machine; the Simple Oscillator skips every downstream stage and goes straight to the Mixer.

Filter types reference

The filter slot offers 9 filter models organised into 5 families. Each model can come in several variants (slope, type) — pick the one that matches the slope and character you need.

Filter families
FamilyInspired byCharacter
SVFState Variable Filter (Oberheim SEM)Smooth, neutral, musical. The morphing variant SVF 12 continuously sweeps from LP to Notch to HP via Knob 3.
TLD (Transistor Ladder)Moog ladderWarm, fat, classic analog low-end. The widest selection of slopes (6, 12, 18, 24 dB/oct) and types (LP, BP, HP, Notch).
K35 (Korg 35)Korg MS-10 / MS-20 (early)Aggressive, slightly nasal, well-known for cutting leads.
OTAKorg MS-20 (later, OTA-based)Distinctive bite, vocal resonance peak, edges into self-oscillation.
STP (Steiner-Parker)Steiner-Parker SynthaconPunchy and articulate, with a distinct response curve compared to ladder filters.
DLD (Diode Ladder)Roland TB-303Squelchy, acidic 24 dB/oct low-pass with strong resonance.
COMBComb filter for hollowed-out, metallic and "whoosh" textures. + and variants invert the feedback polarity.
FORMANTVocal tractVowel filter — Knob 1 morphs through A → E → I → O → U.
BELL EQParametric EQSingle-band bell. Knob 2 sets the bandwidth, Knob 3 sets the gain.
Available variants
FamilyVariants
OffFilter is bypassed.
SVFSVF 12 (morphing LP/Notch/HP, 12 dB/oct), SVF BP12 (fixed band-pass, 12 dB/oct).
TLD — low-passTLD LP6, TLD LP12, TLD LP18, TLD LP24.
TLD — band-passTLD BP12, TLD BP24.
TLD — high-passTLD HP6, TLD HP12, TLD HP18, TLD HP24.
TLD — notchTLD N12, TLD N24.
K35K35 LP12, K35 HP6.
OTAOTA LP12, OTA LP24, OTA HP12, OTA HP24.
STPSTP LP12, STP BP12, STP HP12.
DLDDLD LP24.
COMBCOMB+, COMB−.
FORMANTFORMANT.
BELL EQBELL EQ.

Tip: when picking a slope, remember that lower slopes (6 dB/oct) leave more harmonics through and sound airier, while steeper slopes (24 dB/oct) cut more aggressively. The TLD family is the only one that exposes every slope from 6 to 24 dB/oct in LP and HP, which makes it the most flexible workhorse of the bunch.