Misty — Live Granular Audio Processor
(TODO: screenshot)
License
MIT License
(c) 2026 Andy Jenkinson (uglifruit) — with ClaudeCode.
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Overview
Misty is a live granular audio processor for the Phazerville/Hemisphere audio applet framework. It continuously records incoming audio into a 1-second PSRAM circular buffer and plays it back as a cloud of overlapping grains.
Inspired by Mutable Instruments Clouds, Misty extends the simpler Mist applet with:
- Continuous grain window morphing (Texture)
- Density centred at silence, with stochastic or periodic modes
- Grain feedback — the granular output feeds back into the recording buffer
- Pitch spread — each grain is randomly pitched within a configurable range around the base pitch
Available as a MONO applet.
Signal Flow
Audio In ──► AudioEffectClouds (grain cloud) ──► wet ──────────────────┐
Audio In ──────────────────────────────────────► dry ──────────────────┤──► Output
The wet (granular) and dry signals are mixed by the Mix parameter.
Parameters
Page 1
| Param | Range | Description |
|---|---|---|
| Pos | 0–100% | Read position in the buffer. 0% = current (live), 100% = 1 second ago. |
| Den | −20 to +20 Hz | Grain spawn rate, centred at silence. Negative = regular periodic; positive = stochastic (random timing). |
| Sz | 10–500 ms | Grain duration. Short = glitchy fragments; long = smooth smear. |
| Spr | 0–100% | Position scatter. Randomises each grain’s start point around Pos. |
| PSp | 0–100% | Pitch spread. Each grain’s pitch is offset randomly by up to ±12 semitones (quadratic curve — 50% ≈ ±6 st). 0% = all grains at the same pitch. |
Page 2
| Param | Range | Description |
|---|---|---|
| Pt | −12 to +12 st | Pitch shift in semitones. Unity = 0. |
| Fdb | 0–100% | Feedback. Amount of grain output fed back into the recording buffer. |
| Tex | 0–100% | Grain window shape. 0% = rectangular (harsh/clicky), 50% = triangle, 100% = Hann (smooth). |
| Mix | 0–100% | Wet/dry balance. 0% = fully dry, 100% = fully wet. |
| Frz | gate / latch | Freeze. Stops the write pointer so grains replay a fixed snapshot. |
All parameters except Frz accept CV modulation via assignable CV inputs.
Controls
| Control | Action |
|---|---|
| Encoder (browse) | Navigate cursor through parameters |
| Encoder (edit) | Adjust parameter value |
| Button | Enter/exit edit mode; assign CV input when cursor is on a CV slot |
| Aux button | Latch/unlatch manual freeze — for live performance without a gate cable |
CV / Gate Inputs
Each parameter (except Frz) has an assignable CV input slot. Frz accepts a gate/digital input. Assign inputs using the in-applet CV input editor (press button when cursor is on a CV label).
Density
Density is centred at silence (displayed as 0). This differs from Mist’s always-positive density:
- Positive values (+1 to +20 Hz): Stochastic — grains are spawned at randomised intervals, creating an irregular, diffuse cloud.
- Negative values (−1 to −20 Hz): Periodic — grains are spawned at a steady, regular rate.
- Zero: No grains. Silence on the wet output.
Texture
Texture continuously morphs the grain envelope shape:
| Value | Shape | Character |
|---|---|---|
| 0% | Rectangular | Hard edges — harsh, clicky |
| 50% | Triangle | Smooth fade in/out |
| 100% | Hann (sin²) | Softest, most musical |
Freeze
Freeze halts the write pointer. Grains continue to spawn and play from the frozen buffer, creating a looping granular texture from the captured moment. Two ways to freeze:
- Gate input on the Frz parameter (hardware gate cable)
- Aux button latch (no cable needed — great for live use)
When latched, the Frz row label is shown inverted (white on black) on screen.
Feedback
The Fdb parameter feeds a portion of the grain output back into the audio recording buffer. At low amounts this adds warmth and density. At high amounts it creates self-reinforcing feedback textures. With Freeze active, the feedback circulates endlessly within the frozen buffer. Default is 0% — no feedback.
If you want reverb on the granular output, add a Bungverb applet downstream in the processor chain.
Activity Display
A row of pixels at the top of the display (y=7) shows active grain count in real time — one pixel per active grain, up to 12 maximum.
Technical Notes
- Buffer: ~1 second stored in PSRAM (EXTMEM). Falls back to half-size in OCRAM if no PSRAM. Displays “No PSRAM” if unavailable.
- Grains: up to 12 simultaneous grains. Output scaled by 0.25 for headroom.
- Interpolation: Hermite cubic interpolation on grain read pointer.
- Window morph: implemented via a 256-entry Q15 Hann LUT in flash — no trigonometric functions in the audio hot loop.
Files
| File | Purpose |
|---|---|
audio_applets/MistierApplet.h | UI and control layer (HemisphereAudioApplet) |
Audio/AudioEffectClouds.h | DSP engine (AudioStream subclass) |
audio_applets/Mistier_CLAUDE.md | Internal notes for Claude Code |