Loading...
Please wait while we prepare your experience
Projects
Emergent Geometry
Project Overview
A modular system for real-time audio-reactive geometry. MIDI input from Ableton drives parametric transformations of geometric primitives through Python state management. The architecture separates control logic from visual output. MIDI notes trigger parameter modulations across a pool of geometric objects. Each shape responds independently to specific frequency bands or rhythmic elements.
Built in TouchDesigner using node-based composition. The event system handles MIDI velocity mapping to geometric parameters like sphere tessellation, transform rotations, and wireframe density. State persists across frames, allowing geometries to evolve over musical phrases rather than reacting instantaneously to each note. Performance characteristics favor low-latency response over visual complexity.
Simple primitives (spheres, boxes, lines) compose into emergent patterns through layering and interference. The system prioritizes frame rate stability during live performance. Component isolation enables testing individual geometric behaviors before integration. Each shape maintains its own parameter space and transformation logic. The modular structure supports both live improvisation and pre-composed sequences.
The Problem
Creative technologists and performers needed a flexible, modular system for building complex real-time visuals without starting from scratch for each project. Some live MIDI implementations use fragile direct node connections, creating timing conflicts & unpredictable behaviour, that could potentially lead to catastrophic failures.
The Solution
Designed and built a modular TouchDesigner framework with three integrated layers: Modular Visual Components (reusable TOX containers), Python MIDI Event Brain (centralised event dispatcher), and Event-Driven Architecture (MIDI triggers orchestrate visual states through clean event handling patterns).
Key Challenges
- Real-time performance optimization across complex node networks
- Creating truly reusable modules with flexible I/O
- Maintaining 60fps with multiple high-resolution outputs
- Implementing robust event-based MIDI handling for performance reliability
- Mapping MIDI events to complex visual state transitions without conflicts
- Managing multiple MIDI controllers with conflict-free event routing
- Achieving <5ms latency from MIDI input to visual response
- Balancing modularity with processing efficiency
- Standardizing data flow between disparate components
- Preventing timing issues and race conditions in event handling
Results
- 70% faster project setup time through reusable modules
- 100% performance reliability - zero MIDI-related failures through event-based control
- <5ms latency from MIDI input to visual response
- Consistent 60fps performance with complex multi-output visual chains
- Support for 4+ simultaneous MIDI controllers with conflict-free routing
- Play visuals like an instrument - immediate, expressive, predictable control
- Python brain enables sophisticated MIDI mapping and macro programming
- Event-driven architecture eliminates timing issues and race conditions
Questions about this project?
Technology Stack
Project Gallery
Explore the Project
Click the buttons above to explore the live Emergent Geometry project or view the source code on GitHub.