basilica-audio · yves-vogl · Jul 16, 2026 · Jul 16, 2026
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -5,6 +5,31 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog 1.1.0](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.2.0] - 2026-07-16
+
+### Changed (headline: 2-band → 3-band topology rebuild)
+
+- **Restructured the signal path from a 2-band (low/high) split to a genuine 3-band (low/mid/high) split**, matching the reference class's own defining architecture ("bass, mids, and high frequencies are processed separately" - see `docs/design-brief.md`/`docs/research-notes.md` for the full sourcing). Two cascaded 4th-order Linkwitz-Riley crossovers replace v0.1.x's single split: `Split Low` (60–400 Hz, default 120 Hz, was `Crossover Frequency`) peels off the Low band; `Split High` (300–2000 Hz, default 600 Hz, NEW) further splits the remainder into Mid and High. `Split High` is always clamped at least 1/3 octave above `Split Low` (`src/dsp/SplitGap.h`) to prevent a degenerate near-zero-width Mid band.
+- **Added a new Mid band** (`src/dsp/MidBand.h/.cpp`): staged/cascaded drive-only saturation (0-100%, default 30%) plus an independent output Level, no filter/tone/blend - matching the reference class's own Mid band exactly. Runs its own 4x-oversampled shaping stage.
+- **Promoted the High band's pre-drive highpass ("Tight") from a Razor-only fixed 200 Hz internal constant to a first-class, voicing-independent control** (20-500 Hz, default 100 Hz), applied ahead of all three voicings (Gnaw/Wool/Razor) - closing the gap identified in the design brief's "Why v1 falls short" analysis.
+- **Re-sourced the low-band parallel compressor's ballistics defaults** to the reference class's own fixed, documented "glue" bus-compressor values: ratio 4:1 → **2:1**, attack 10 ms → **3 ms**, release 120 ms → **6 ms** (release range floor lowered from 10 ms to 5 ms, a breaking pre-1.0 change, so the sourced default is reachable). Retires the v0.1.x manual's "New York style" framing, which characterized the wrong sub-genre of parallel compression for what the reference class actually implements - see `docs/research-notes.md` §3-4.
+- **Relocated the IR loader (cab-sim convolution)** to process only the Mid+High post-sum signal, never the Low band - matching the reference class's "low band bypasses the cabsim" architecture. Structurally enforced (not just conventional) and directly tested (`tests/LowBandIsolationTests.cpp`): the Low band's own isolated output is bit-exact identical whether the IR loader is on or off, or which IR is loaded.
+- **Re-anchored the post-sum 4-band EQ's default corner frequencies** to a sourced bass-tone-stack frequency set from the same design lineage as the reference class: Low Shelf 100 → **80 Hz**, Peak 2 2500 → **2800 Hz**, High Shelf 8000 → **5000 Hz** (Peak 1's existing 500 Hz default already matched the sourced anchor). Dormant-until-engaged (EQ ships off by default), so not an audible v0.1.x → v0.2.0 change unless a user or preset turns the EQ on.
+- **Added a phase-alignment allpass filter for the Low band** (`src/dsp/PhaseAlignFilter.h`), required to make the new cascaded (not parallel) crossover topology actually flat-sum - discovered and fixed during this rebuild: a naive cascade of two independent LR4 crossovers does *not* flat-sum on its own (deviations up to −10 dB were measured at close `Split Low`/`Split High` ratios before this fix). The fix is proven algebraically, not just empirically - see `docs/architecture.md`'s "Cascaded 3-band flat-sum and phase alignment" section for the full derivation. This is a genuine engineering necessity this rebuild surfaced, not something anticipated by the original design brief.
+- **Lossy state migration for v0.1.x sessions**: the old single `Crossover Frequency` (`crossoverFreq`) parameter is migrated to the new `Split High` parameter on load, clamped into its 300–2000 Hz range. v0.1.x's shipped default (250 Hz) sits below that floor, so the single most common migration path - an untouched v0.1.x session - lands exactly at the 300 Hz floor (dedicated regression test in `tests/StateMigrationTests.cpp`). `Split Low` and every new Mid-band/Tight parameter fall back to their v0.2.0 defaults; any low-band compressor values a user had explicitly changed away from v0.1.x's old defaults are preserved as-is.
+- Docs rewritten to match: `docs/manual.md`, `docs/architecture.md`, `README.md` (signal-flow diagrams, parameter tables, feature list). `docs/design-brief.md` and `docs/research-notes.md` added (the binding brief and its sourcing for this rebuild).
+
+### Added (M2 preset system)
+
+- Suite-wide M2 preset system (`.scaffold/specs/preset-system-m2.md`), ported from `basilica-audio/nave`'s pilot implementation: `src/presets/PresetManager.{h,cpp}` (factory + user presets, save/save-as/delete/rename, default resolution, import/export of single presets and zip banks, dirty-state tracking) and `src/presets/PresetBar.{h,cpp}` (a horizontal preset strip docked at the top of the editor).
+- Nine factory presets (`presets/factory/*.json`, category `Init`/`Bass`) - see `docs/presets.md` for what each demonstrates: **Default**, **Glue & Grind**, **Sub Lock**, **Throat**, **Fuzz Wall**, **Cut Through**, **Definition Only**, **Clean Low, Loud Top**, **Cab-Colored Grind**.
+- German localisation of the preset UI frame (`resources/i18n/de.txt`), auto-selected via `SystemStats::getUserLanguage()`. Core/DSP terminology (parameter names, units) is never translated.
+- User presets stored at `~/Library/Audio/Presets/Yves Vogl/Crypta/` (macOS) / `%APPDATA%\Yves Vogl\Crypta\Presets\` (Windows).
+
+### Fixed
+
+- Discovered and fixed during the topology rebuild: a naive cascade of two independent LR4 crossovers does not flat-sum (see the phase-alignment entry above) - this was never an issue in v0.1.x's single-crossover topology, so it is not a regression from any prior release, but a new correctness requirement introduced by the 3-band rebuild itself.
+
 ## [0.1.1] - 2026-07-16
 
 ### Changed

diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -11,7 +11,7 @@ set(CMAKE_OSX_DEPLOYMENT_TARGET "11.0" CACHE STRING "Minimum macOS deployment ta
 # here at the top level avoids it being enabled implicitly deeper inside
 # JUCE's CMake helpers, which has been observed to break Ninja's generate
 # step (missing CMAKE_C_COMPILE_OBJECT rule variable).
-project(Crypta VERSION 0.1.1 LANGUAGES C CXX)
+project(Crypta VERSION 0.2.0 LANGUAGES C CXX)
 
 set(CMAKE_CXX_STANDARD 20)
 set(CMAKE_CXX_STANDARD_REQUIRED ON)
@@ -56,6 +56,30 @@ juce_add_plugin(Crypta
     ICON_BIG "docs/assets/icon.png"
 )
 
+# ==============================================================================
+# Embedded assets: M2 preset system factory presets (presets/factory/*.json)
+# and the M2 i18n frame's German translation (resources/i18n/de.txt). See
+# src/presets/PresetManager.h and src/presets/Localisation.h - both consume
+# these purely via BinaryData:: symbols passed in from PluginProcessor.cpp/
+# PluginEditor.cpp, never by including BinaryData.h themselves, so they stay
+# portable to sibling plugins (see docs/preset-system-notes.md in
+# basilica-audio/nave, the M2 pilot this suite-wide pattern was replicated
+# from).
+# ==============================================================================
+
+juce_add_binary_data(CryptaBinaryData SOURCES
+    presets/factory/default.json
+    presets/factory/glueAndGrind.json
+    presets/factory/subLock.json
+    presets/factory/throat.json
+    presets/factory/fuzzWall.json
+    presets/factory/cutThrough.json
+    presets/factory/definitionOnly.json
+    presets/factory/cleanLowLoudTop.json
+    presets/factory/cabColoredGrind.json
+    resources/i18n/de.txt
+)
+
 # ==============================================================================
 # SharedCode: an INTERFACE target carrying the plugin's sources, definitions,
 # and library links so that both the plugin targets (AU/VST3/Standalone) and
@@ -86,9 +110,11 @@ target_compile_definitions(SharedCode INTERFACE
 target_link_libraries(SharedCode INTERFACE
     juce::juce_audio_utils
     juce::juce_dsp
+    juce::juce_gui_basics
     juce::juce_recommended_config_flags
     juce::juce_recommended_lto_flags
     juce::juce_recommended_warning_flags
+    CryptaBinaryData
 )
 
 target_link_libraries(Crypta PRIVATE SharedCode)

diff --git a/README.md b/README.md
@@ -6,56 +6,60 @@
 
 > Formerly **Twist Your Guts**, renamed to Crypta as part of the suite's move to Basilica Audio naming (the crypt: the basilica's low-end foundation). If you have a v0.1.0-era session referencing the old plugin identity (`com.yvesvogl.twistyourguts`, plugin code `Tygt`), see the [Unreleased] entry in [`CHANGELOG.md`](CHANGELOG.md) — the new bundle ID and plugin code (`com.yvesvogl.crypta`, `Cryp`) mean DAWs treat this as a new plugin, so existing sessions will need to be re-pointed at the new plugin.
 
-[![CI](https://github.com/metal-up-your-ass/Crypta/actions/workflows/ci.yml/badge.svg)](https://github.com/metal-up-your-ass/Crypta/actions/workflows/ci.yml)
+[![CI](https://github.com/basilica-audio/Crypta/actions/workflows/ci.yml/badge.svg)](https://github.com/basilica-audio/Crypta/actions/workflows/ci.yml)
 [![License: AGPL v3](https://img.shields.io/badge/License-AGPL%20v3-blue.svg)](https://www.gnu.org/licenses/agpl-3.0)
 
-> **Work in progress.** Crypta is pre-1.0 and under active development (v0.1.0). Binaries for macOS and Windows are available from the [Releases](../../releases) page (macOS builds are signed & notarized); building from source works too. Expect breaking changes until v1.0.0 ships (see [Roadmap](#roadmap)).
+> **Work in progress.** Crypta is pre-1.0 and under active development (v0.2.0). Binaries for macOS and Windows are available from the [Releases](../../releases) page (macOS builds are signed & notarized); building from source works too. Expect breaking changes until v1.0.0 ships (see [Roadmap](#roadmap)).
 
 ## What it is
 
-Crypta is a Parallax-style bass plugin built on JUCE 8. It splits your bass signal into low and high bands with a linear-phase-adjacent Linkwitz-Riley crossover, compresses the low band in parallel, and runs the high band through a choice of three distortion voicings before summing everything back together through a 4-band EQ and an impulse-response (cab sim) loader. See [`docs/manual.md`](docs/manual.md) for the full parameter reference and usage tips.
+Crypta is a Parallax-style bass plugin built on JUCE 8. As of v0.2.0 it splits your bass signal into **three** bands — low, mid, and high — with two cascaded 4th-order Linkwitz-Riley crossovers, compresses the low band in parallel, drives the mid band with staged saturation, and runs the high band through a choice of three distortion voicings before summing everything back together through a 4-band EQ and an impulse-response (cab sim) loader. v0.2.0 also adds a full preset system (factory + user presets, import/export, German localisation). See [`docs/manual.md`](docs/manual.md) for the full parameter reference and usage tips, and [`docs/design-brief.md`](docs/design-brief.md)/[`docs/research-notes.md`](docs/research-notes.md) for the research behind the v0.2.0 topology rebuild.
 
 ## Features
 
-- **Noise gate** — full-band, ahead of the crossover split
-- **LR4 crossover band-split** — 4th-order Linkwitz-Riley split, adjustable 60 Hz – 1000 Hz (default 250 Hz)
-- **Low band**: parallel ("New York style") compressor with makeup gain, wet/dry mix, and output level
-- **High band**: three distortion voicings, each 4x oversampled to keep aliasing under control, with independent drive/tone
+- **Noise gate** — full-band, ahead of both crossover splits
+- **Two cascaded LR4 crossover splits** — Split Low (60–400 Hz, default 120 Hz) and Split High (300–2000 Hz, default 600 Hz), building a genuine 3-band (low/mid/high) topology, replacing v0.1.x's 2-band split
+- **Low band**: parallel "glue" compressor (re-sourced fast/gentle ballistics, ratio 2:1 / attack 3 ms / release 6 ms) with makeup gain, wet/dry mix, and output level
+- **Mid band** (NEW): staged/cascaded drive-only saturation, no filter/tone/blend — a distinct "throatier" character separate from the high band
+- **High band**: three distortion voicings, each 4x oversampled to keep aliasing under control, now with a shared, voicing-independent **Tight** pre-drive highpass (was Razor-only in v0.1.x)
   - **Gnaw** — op-amp-style hard clip
   - **Wool** — cascaded soft-clip fuzz with a mid scoop
-  - **Razor** — tight overdrive: pre-clip highpass, soft clip, mid hump
-  - Clean/distorted blend control per voicing, plus output level
-- **4-band EQ** post-sum (LowShelf / Peak / Peak / HighShelf)
-- **IR loader** (cabinet simulation) on the output stage — convolution engine is live; bundled factory IRs and a GUI file browser land in a later milestone
-- **Delay-compensated signal path** — the high band's oversampling latency is reported to the host and the low band is time-aligned to match
-- **Presets** with full state save/recall *(planned — a dedicated preset manager/versioning scheme is a later milestone; APVTS state save/load already round-trips today)*
+  - **Razor** — tight overdrive: soft clip, mid hump
+  - Clean/distorted blend control, plus drive/tone/output level
+- **4-band EQ** post-sum (LowShelf / Peak / Peak / HighShelf), re-anchored default frequencies (80/500/2800/5000 Hz, sourced from the same design lineage's hardware tone stack)
+- **IR loader** (cabinet simulation), relocated in v0.2.0 to process only the Mid+High post-sum signal — the low band never passes through it, matching the reference class's own architecture. Convolution engine is live; bundled factory IRs and a GUI file browser land in a later milestone
+- **Delay-compensated, phase-aligned signal path** — the Mid+High branch's shared oversampling latency is reported to the host, the low band is time-aligned to match, and a phase-alignment allpass filter keeps the cascaded three-way sum flat-magnitude
+- **Presets** — factory + user presets, save/save-as/delete, import/export (single files and zip banks), German localisation of the preset UI frame
+- **State migration** — a v0.1.x session's single crossover frequency is migrated to the new Split High parameter on load
 - **Metering** throughout the signal chain *(planned, alongside the custom GUI)*
 
 ## Signal flow
 
 ```
-Input Trim → Gate → LR4 Split (60–1000 Hz, default 250 Hz)
+Input Trim → Gate → LR4 Split Low (60–400 Hz, default 120 Hz)
                       │
-        ┌─────────────┴─────────────┐
-        │                           │
-     Low band                   High band
-  Parallel Comp        Voicing → Drive → Tone → Blend
-  → Makeup → Mix                    │
-        │→ Level              → Level
-        └─────────────┬─────────────┘
-                       │
-              Sum (delay-compensated)
-                       │
-                  4-band EQ
-                       │
-                  IR loader
-                       │
-              Safety Clip (optional)
-                       │
-                 Output Trim
+        ┌─────────────┴───────────────────────────────┐
+        │                                              │
+     Low band                              Remainder → LR4 Split High (300–2000 Hz, default 600 Hz)
+  Parallel Comp → Level                                  │
+        │                          ┌───────────────────┴───────────────────┐
+        │                       Mid band                              High band
+        │                    Drive → Level          Tight → Voicing → Drive → Tone → Blend → Level
+        │                          └───────────────────┬───────────────────┘
+        │                                          Mid+High sum → IR loader (cab sim)
+        │                                                │
+        └──────────── Phase-align + delay ───────────────┘
+                                 │
+                       Sum (delay-compensated)
+                                 │
+                            4-band EQ
+                                 │
+                       Safety Clip (optional)
+                                 │
+                            Output Trim
 ```
 
-The high band runs 4x oversampled for the distortion stage; the low band is delay-compensated to stay time-aligned with it before the sum. See [`docs/architecture.md`](docs/architecture.md) for the full breakdown, including the latency-compensation strategy, and [`docs/manual.md`](docs/manual.md) for the full parameter reference.
+The Mid and High bands each run 4x oversampled (identically configured, so their latencies match exactly); the low band is delay-compensated and phase-aligned to stay both time- and magnitude-flat with them before the sum. See [`docs/architecture.md`](docs/architecture.md) for the full breakdown, including the latency-compensation strategy and the phase-alignment proof, and [`docs/manual.md`](docs/manual.md) for the full parameter reference.
 
 ## Parameters
 
@@ -65,9 +69,10 @@ See [`docs/manual.md`](docs/manual.md) for the complete, musically-annotated par
 |---|---|
 | IO / Global | Input Gain, Output Gain, Bypass, Safety Clip |
 | Noise Gate | Enable, Threshold, Ratio, Attack, Release |
-| Crossover | Frequency (60–1000 Hz) |
+| Crossover | Split Low (60–400 Hz), Split High (300–2000 Hz) |
 | Low band | Comp Threshold/Ratio/Attack/Release/Makeup/Mix, Level |
-| High band | Voicing (Gnaw/Wool/Razor), Drive, Tone, Blend, Level |
+| Mid band | Drive, Level |
+| High band | Tight, Voicing (Gnaw/Wool/Razor), Drive, Tone, Blend, Level |
 | EQ | Enable, Low Shelf Freq/Gain, Peak 1 Freq/Gain/Q, Peak 2 Freq/Gain/Q, High Shelf Freq/Gain |
 | IR loader | Enable, Mix |
 
@@ -111,7 +116,7 @@ ctest --test-dir build --output-on-failure
 |---|---|---|
 | M0 | Bootstrap — project skeleton, CI, docs | Done |
 | M1 | DSP completion & test coverage — gate, crossover, parallel compressor, 3 voicings (oversampled), 4-band EQ, IR loader, latency compensation, broadened test suite | Done (v0.1.0) |
-| M2 | Presets & state recall — preset manager, factory presets, versioned state | Planned |
+| M2 | Deep-dive topology rebuild (2-band → 3-band) + presets & state recall — preset manager, 9 factory presets, state migration, German localisation | Done (v0.2.0) |
 | M3 | GUI & accessibility — custom LookAndFeel, metering UI, accessibility pass | Planned |
 | M4 | Release: signing, notarization, v1.0.0 — installers, tagged release | Planned |
 
@@ -132,4 +137,4 @@ Tagged releases (`v*`) are built and published automatically by [`.github/workfl
 - **macOS** — AU (`.component`), VST3 (`.vst3`), and Standalone, Universal Binary (arm64 + x86_64), signed with a Developer ID Application certificate (org-level secrets, shared across the Basilica Audio suite), notarized, and stapled. Installs and opens without a Gatekeeper warning.
 - **Windows** — VST3 and Standalone, **unsigned**. On first run, Windows SmartScreen may show a "Windows protected your PC" warning; choose **More info → Run anyway** to proceed. A signed Windows build is a documented future improvement, not yet available.
 
-See [`v0.1.0`](https://github.com/metal-up-your-ass/Crypta/releases/tag/v0.1.0) for the most recent published release (under the plugin's prior identity as Twist Your Guts); the next tagged release under the Crypta identity is planned as `v0.2.0`.
+See [`v0.1.1`](https://github.com/basilica-audio/Crypta/releases/tag/v0.1.1) for the most recent published release; `v0.2.0` is the next tagged release, shipping the 2-band → 3-band topology rebuild and the M2 preset system.