Pure
Precise translation and clean recursive relationships.
TRANSLATE / RECUR / FOLD
Ghost Frequency Shifter
Move every frequency by the same number of hertz.
Use tiny translations for beating and animation, or larger shifts for metallic spectra. Feedback repeats the translation; Cascade, Gravity and Fold shape where those generations travel.
Frequency shifting is not pitch shifting. Adding 100 Hz changes harmonic spacing; it does not transpose everything by a musical interval.
START HERE / ABOUT 90 SECONDS
From subtle beating to obvious translation.
Use held keys, guitar or a drum loop. Choose Init and monitor quietly.
START FROM INIT
Shift 0 Hz · Direction 0% · Feedback 0% · Cascade 0% · Gravity 0% · Fold 0% · Motion 0% · Spread 0% · Character Pure · Mix 100% · Output 0 dB
0 of 4 steps checked — reading progress only.
WHAT TO LISTEN FOR
The same additive-Hz process spans slow movement and unfamiliar metallic relationships. Recursion can continue after the input ends.
Reduce Mix for a parallel layer. Keep Feedback modest while learning Cascade/Fold and use Panic for sustained regeneration.
LEARNING ILLUSTRATION — NOT CONNECTED TO YOUR PLUGIN
A ladder can converge or expand.
A qualitative explanation with nonzero Feedback, not a calculated spectrum or plugin connection.
Negative Cascade reduces translation across later generations, encouraging convergence.At zero, each recursive generation receives the same additive-Hz shift.Positive Cascade expands later shifts; Fold can return material that meets a spectral boundary.A qualitative explanation, not a calculated response or audio preview. Use the plugin to hear the result.
PARAMETER REFERENCE
A small set of big gestures.
Learn Shift with Feedback at zero. Add recursion gradually, then change the geometry of those generations.
Shift-20000–20000 Hz · Default 0 Hz
Additive frequency translation, with a strong exact-zero center and very fine resolution near zero. Positive and negative move opposite directions.
Try this: Compare +2 Hz with +100 Hz at Feedback 0%.
Direction-100–100% · Default 0%
Zero is clean single-sideband translation according to Shift's sign. Positive adds the opposite sideband, reaching symmetric dual-sideband at +100%; negative adds it inverted.
Try this: Compare 0% and +40% at a clearly audible Shift.
Feedback-110–110% · Default 0%
Re-translates recursive passes. Opposite signs change reinforcement/cancellation; near±100% energy lasts longer and beyond it controlled regeneration is possible.
Try this: Begin at 25–35%, not above unity. Keep Panic available.
Cascade-100–100% · Default 0%
Changes translation across successive generations: zero keeps equal-Hz steps, positive expands them and negative converges them. It needs recursive energy to reveal its role.
Try this: With moderate Feedback, compare±30%.
Gravity-100–100% · Default 0%
Positive attracts translated structures toward harmonic relationships; negative repels them. It preserves the frequency-shifting mechanism, not hard scale quantization.
Try this: Compare±25% on a recursive source.
Fold0–100% · Default 0%
Returns translated energy by reflection at usable spectral boundaries. Zero lets out-of-range components disappear; this is not ordinary amplitude wavefolding.
Try this: Use a larger Shift and gradually raise Fold if energy is leaving the audible range.
Motion-100–100% · Default 0%
Signed deterministic evolution around Shift/Cascade/Direction relationships. Negative complements/reverses positive evolution without resetting it. It is not an exposed internal LFO.
Try this: Try +20%, then -20%, over a sustained source.
Spread-100–100% · Default 0%
Positive creates opposing L/R translation trajectories; negative makes one follow the other. Neutral/unavailable for mono output.
Try this: Try +20% in stereo and compare at zero.
CharacterPure / Analog / Glass / Metal / Fracture / Void · Default Pure
Selects six translation/feedback laws. Their differences often become clearer when recursive energy exists; they are not post-EQ presets.
Try this: Compare Pure and Glass with moderate Feedback.
Mix0–100% · Default 100%
Blends aligned dry audio with the complete translated result. It is not a feedback or sideband-direction control.
Try this: Try 40% for a metallic layer under a recognizable source.
Output-12–12 dB · Default 0 dB
Final gain trim. Regenerative examples need conservative monitoring; trim does not reset the recursive structure.
Try this: Level-match after changing Feedback or Character.
CHARACTER REFERENCE
Choose a character.
Analog
Controlled quadrature imperfections, softened bandwidth and nonlinear feedback.
Glass
Clear upper structures and crystalline recursive trails.
Metal
Dense inharmonic recursive interaction for clangorous textures.
Fracture
Less uniform, branching recursive generations.
Void
Weak generations disappear faster while stronger spectral objects persist.
KNOW YOUR INSTRUMENT
Find your way around.
The shared graph shows measured source material, projected translated branches and real I/O waveform context. A projected branch spectrum is not a separate final-output FFT.
Select a numbered control
Choose a marker, or use the complete keyboard-accessible control reference above.
Horizontal/vertical drags change Shift/Gravity; Shift-horizontal/vertical changes Cascade/Feedback. Pulling a generation maps to existing parameters, not a new per-generation editor.
At nonzero shifts up to 10 Hz, Micro focuses around a measured source region. Its local linear-Hz axis magnifies translation; interpolation is not a claim of extra analyzer resolution.
Full-range spectral frequency/level and waveform time/amplitude share one canvas with different meanings. Tab/arrows/Enter/Escape provide field keyboard control; ordinary knobs remain available.
HEAR THE IDEA
Put it into practice.
An additive 120 Hz move
Actual Pure Frequency Shifter at +120 Hz and Feedback 25%, otherwise neutral. The source stops early to expose remaining recursive energy. This is not a musical pitch-shift interval.
Source
Source WAVProcessed · actual Ghost DSP
Processed WAVExample settings and provenance
{"shift": 120, "feedback": 25, "mix": 100}. Actual native processor, stereo48k/127/double; source latency-aligned. Whole-file wet RMS gain 0.968392; common trim 1.0. Identical final20ms file fade; no listening approval inferred.
Reproduction recordA slowly moving parallel texture
Use held keys or a pad.
Shift 1.5 Hz · Feedback 20% · Spread 20% · Mix 45%
- Hear the small Shift with Feedback zero first.
- Add a little recursion, then stereo Spread.
- Compare over several seconds; do not expect a tempo-locked cycle.
Listen for: Slow beating and related L/R motion from small frequency differences.
Converging crystalline generations
Use short pitched phrases with gaps.
Character Glass · Shift 90 Hz · Feedback 55% · Cascade -35% · Gravity 30%
- Start quietly and hear each translation before adding Cascade.
- Move Cascade negative, then introduce positive Gravity.
- Stop input to inspect the trail and use Panic if you enter sustained territory.
Listen for: A related recursive structure converging and reorganizing, not a conventional pitched delay.
GO DEEPER
Beyond the first move.
Recursion can regenerate
Beyond±100% Feedback, seeded energy may deliberately sustain or regenerate under bounded protection. Normal bypass stops new feed and releases tails; hard bypass/Panic end energy safely. A large Output reduction is not the same as clearing the state.
Latency and memory
The processor reports framing latency and aligns dry/Rack paths. Full same-rate project recall restores recursive energy and phases. Portable sounds intentionally load without an old live tail; rate changes rebuild incompatible history safely.
In your session
Ten continuous controls support internal modulation and MIDI CC learn. Character is discrete host automation only. No note trigger, external audio sidechain, Free/Sync, Rate or Seed control is provided.
Stereo trajectories require stereo output. Boundary behavior is finite-window processing, not a guarantee of ideal mathematical rejection at every edge frequency.
High-rate multi-instance stress is a documented capacity limitation, not a promise of unlimited simultaneous instances. No lighter Character or extra latency mode replaces the approved sound.
The preset browser offers All, Factory, User and Favorites, category/search filters and relevant Character filtering. A row's heart changes your personal favorites without loading or saving the sound. Save As creates a distinct User sound; explicit Save updates only the selected User sound. Factory content is protected. Name, category and tags support organization; deleting a User category keeps its presets as Uncategorized. Manage supports rename, duplicate, confirmed delete and individual/bulk/category import/export. Personal favorites stay on this installation rather than travelling in exported sounds.
Standalone and the matching Ghost Rack module share the same User sounds, categories and favorites. A module preset changes portable sound settings, not Rack placement, Rack wet/dry, bypass or modulation assignments. DAW recall embeds the sound settings and does not depend on a preset file remaining on disk. Rack chain presets are a different library and affect the whole chain.
Drag a knob normally, hold Shift for fine adjustment or Option for extra-fine. Double-click restores its default. Enter/context menus provide exact values, reset and value copy/paste; eligible controls also offer MIDI Learn and modulation. Tab and arrow keys provide keyboard access. A/B compares configurations; the adjacent copy button copies the active configuration to the other. Undo/Redo groups a continuous gesture into one edit. Settings provides UI scale, Reduced Motion, source configuration, portable settings and Panic. UI size/motion preferences do not change the sound.
WHEN SOMETHING ISN’T CLEAR
Find your way back.
Why is Init unchanged?
Shift is exactly 0 Hz with neutral geometry/feedback. Request a translation to begin.
Why does it sound out of tune?
An additive-Hz move intentionally changes harmonic spacing. Use Pitch Shifter for musical interval transposition.
Why does Cascade do little?
It governs successive recursive translations. Introduce a modest nonzero Feedback amount first.
Why does the small-shift axis look different?
Micro mode uses a labelled local linear-Hz region to reveal small motion, not a different audio algorithm.