🎛️ SPANDAN (स्पंदन)

An Advanced Real-Time Polyphonic Synthesizer & C++ DSP Engine

📖 Read the Official SPANDAN Engineering Textbook & DSP Blog

We are building in public. Curious about the core telecommunications mathematics and C++ implementations powering this engine? Dive into our chapter-by-chapter technical documentation, starting with the heart of the engine: the Numerically Controlled Oscillator (NCO).

📌 Overview

SPANDAN is a strictly deterministic, low-latency Digital Signal Processing (DSP) audio engine built natively in C++ using the latest JUCE 9 hardware abstraction framework.

Unlike standard commercial tools that rely heavily on CPU-intensive sample playback, SPANDAN bypasses conventional architecture by generating baseband signals mathematically via discrete-time phase accumulation. It is engineered to translate core telecommunications mathematics into a highly optimized, real-time software environment suitable for baseline CPU architectures.

⚙️ Core DSP Architecture

The engine executes its entire mathematical pipeline within the high-priority audio callback thread (at a 44.1 kHz sample rate) with zero dynamic memory allocation (new/malloc), guaranteeing microsecond-level execution without buffer underruns.

• Numerically Controlled Oscillators (NCO): Mathematical phase accumulators for baseband signal generation.
• Frequency Modulation (FM) Synthesis: Advanced modulation matrix where $f_m$ (modulator) alters the phase of $f_c$ (carrier) to generate complex harmonic sidebands.
• Algorithmic Voice Stealing & Polyphony: A thread-safe dynamic voice allocation system that manages concurrent DSP instances and intelligently reassigns hardware computing threads when polyphony limits are exceeded.
• Infinite Impulse Response (IIR) Filters: Custom discrete digital state-variable filter cascades for frequency spectrum shaping.
• Real-Time Spectral Analytics: Integrated Radix-2 Fast Fourier Transform (FFT) algorithms featuring Hann windowing to prevent spectral leakage, rendering visual diagnostics at 60 FPS on the UI thread.

🛠️ Tech Stack & Target Platforms

• Language: C++23 / C++26 (Deterministic O(1) algorithmic execution, leveraging modern constexpr and vectorization).
• DSP Framework: JUCE 9 (juce_audio_processors, juce_dsp).
• Primary IDE & Compiler: MSVC (Visual Studio 2026) / LLVM.
• Target Plugin Formats:
  • VST3 (Cross-platform digital audio workstation plugin)
  • AU (Audio Unit) (macOS native audio processing standard)
  • Standalone (Executable binary without host DAW requirement)

🚀 Build Instructions

To compile the DSP engine locally from the source code:

1. Clone the repository: git clone https://github.com/cryparc/Spandan.git
2. Open Spandan.jucer using the Projucer 9 application.
3. Ensure the global paths for JUCE 9 modules are correctly set in your Projucer environment.
4. Select your target exporter (Visual Studio 2026 for Windows VST3/Standalone or Xcode for macOS AU/VST3 compilation).
5. Click "Save and Open in IDE".
6. Build the solution in Release mode for optimal DSP CPU SIMD optimizations, or Debug mode for real-time thread profiling.

(Note: The compiled .vst3, .component (AU), and Standalone binaries are explicitly ignored in this repository to preserve source integrity. You must compile the binaries locally.)

🤝 Contributing & Git Cheat Sheet

Whether you are a beginner looking to compile the C++ engine locally, or a contributor ready to submit DSP patches, here are the exact commands you need. You can copy and paste these directly into your terminal.

1. First-Time Setup (Clone the Repo)

To download the entire project to your local machine, open your terminal (or VS Code) and run:

# Clone the repository
git clone [https://github.com/cryparc/Spandan.git](https://github.com/cryparc/Spandan.git)

# Navigate into the project folder
cd Spandan

2. Daily Workflow (Syncing & Pushing)

If you are actively modifying the code, use this standard workflow to keep everything synced:

# Step 1: Always download the latest updates before starting work
git pull origin main

# Step 2: Stage all your modified files
git add .

# Step 3: Commit your changes with a descriptive message
git commit -m "Brief description of what you changed"

# Step 4: Push the updates back to GitHub
git push origin main

3. Collaboration (Branches & Pull Requests)

If you are adding a new feature (like a new filter or GUI element) or fixing a bug, please do it on a separate branch so it can be reviewed cleanly.

Creating and pushing a new feature branch:

# Create and switch to a new branch (name it something relevant)
git checkout -b feature-new-filter

# Stage and commit your changes (same as Step 2 & 3 above)
git add .
git commit -m "Added a new state-variable filter"

# Push the new branch to GitHub for the first time
git push -u origin feature-new-filter

Creating a Pull Request (PR): Once you push your new branch to GitHub using the command above, navigate to the Spandan GitHub Repository in your web browser. You will see a green "Compare & pull request" button at the top of the page. Click it to submit your code for review!

Reporting a Bug or Requesting a Feature: If you find a bug, experience a buffer underrun, or want to suggest a new DSP feature, please Open an Issue. Provide as much detail as possible, including your OS and DAW host.

📄 License

This project is licensed under the GNU General Public License v3.0. See the LICENSE file for more details. Engineered by Prashant Singh.