Epic Artificial Intelligence History Film

# Generative Cinematic Documentary: AI History

You are an expert code-based generative artist specializing in algorithmic audio-visual synthesis. Your task is to create a fully procedural 2-minute documentary-style video with synchronized soundtrack—**every pixel and every sound wave generated purely through code**.

## Project Specifications

**Title**: {{video-title}}

**Theme & Tone**: Epic documentary trailer aesthetic—sweeping cinematography, emotional crescendos, historical gravitas. Think BBC Earth meets Interstellar. Build narrative tension through visual metaphors: neural networks blooming like organisms, timelines materializing as light trails, pivotal moments punctuated by dramatic visual beats.

**Technical Requirements**:
- Resolution: 1920×1080, 60 FPS
- Duration: Exactly 120 seconds
- Output: MP4 with embedded audio track
- Save location: {{output-path}}

## Audio Synthesis (Fully Procedural)

Compose a three-act soundtrack using pure synthesis—no samples, libraries, or external audio files:

1. **Act I (0-40s)**: Mysterious origins—sparse synth pads, low-frequency drones, occasional crystalline tones suggesting early computational thought
2. **Act II (40-90s)**: Rising complexity—layered arpeggios, building percussion, tempo acceleration reflecting exponential progress
3. **Act III (90-120s)**: Triumphant/uncertain climax—full orchestral-style synthesis with brass-like leads, cinematic drums, unresolved final chord suggesting open future

Implement using:
- Additive/subtractive synthesis for tonal elements
- Noise generators with envelope shaping for percussion
- LFO modulation for movement and texture
- Reverb/delay via convolution or feedback networks
- Mix to -14 LUFS integrated loudness

## Visual Generation (Fully Procedural)

Every frame rendered algorithmically—no image files, no pre-rendered assets. Suggested techniques:

**Visual Motifs**:
- **Neural network visualizations**: Animated graph structures with pulsing nodes, flowing activation signals
- **Historical timeline**: Particle systems forming dates (1950s, 1980s, 2010s) with era-appropriate visual styles
- **Symbolic imagery**: Geometric brain forms, cascading binary, evolving algorithmic patterns (cellular automata → deep networks)
- **Cinematic camera moves**: Slow zooms, orbital rotations, depth-of-field effects via layered rendering

**Technical Approach**:
- Shader-based rendering (GLSL/compute shaders) for performance
- Procedural geometry (signed distance functions, mesh generation)
- Particle systems for organic motion
- Post-processing: film grain, vignette, color grading for cinematic feel
- **Sync visuals to audio**: Extract beat/amplitude data from your synthesized soundtrack; trigger visual events on transients, modulate parameters with envelope followers

## Implementation Strategy

1. **Audio-first workflow**: Generate complete soundtrack, analyze for tempo/dynamics
2. **Scene planning**: Divide 120s into 8-12 scenes, assign visual concepts
3. **Render pipeline**: Frame-by-frame generation with progress logging
4. **Video encoding**: Mux audio + video frames into MP4 (H.264 video, AAC audio)
5. **Validation**: Verify sync, duration, output quality

## Deliverables

- Complete executable code with clear dependencies listed
- Inline comments explaining synthesis/rendering algorithms
- Final rendered MP4 at specified path
- Estimated render time and hardware requirements

Push the boundaries of procedural generation—this should feel like a professional trailer crafted entirely from mathematical expressions. Make every second count.