Phi-Machine Research

Experimental Dashboard

Experiments:2

Reports:3

Research Definition

Our research unifies structural generation with semantic understanding through two core pillars:

Phi-Generation

Developing the generative engine that grows complex textures and structures from prime seeds and physical forces.

Phi-Memory

Developing the manifold geometry that gives these structures meaning, memory, and agent coherence.

Goal

To explore the generative production of reality using seeds of primes, physical rule-sets, and the Fibonacci sequence as the phi-generator. This foundational research powers two distinct applications: the creation of a generative visual library (Phi-Generation) and the geometric organization of meaning for agent coherence (Phi-Memory).

Latest Insight

"Moving to native 2D generation is a critical step forward. It allows the generator to 'understand' geometry better than 1D wrapping. Future work should explore more complex mixing functions beyond XOR (e.g., ADD, MOD) to solve the Checkerboard problem."

Read Report →

Active Research

completed

2D Reachability: Cartesian vs Polar Synthesis

exp-002-2d-reachability • Native 2D generation (Cartesian/Polar) will significantly outperform 1D raster wrapping on structured and organic targets by decoupling axes.

Performance Metrics (MSE)

T-01-stripes8,544.328 MSE

T-02-checker17,686.201 MSE

T-07-gradient-rad2,444.856 MSE

T-12-wood4,808.673 MSE

T-03-grid11,007.926 MSE

Recent Artifacts

pgm

exp-002-2d-reachability-T-01-stripes-sol.pgm

Best solution for T-01-stripes (MSE: 8544.33)

pgm

exp-002-2d-reachability-T-02-checker-sol.pgm

Best solution for T-02-checker (MSE: 17686.20)

pgm

exp-002-2d-reachability-T-07-gradient-rad-sol.pgm

Best solution for T-07-gradient-rad (MSE: 2444.86)

pgm

exp-002-2d-reachability-T-12-wood-sol.pgm

Best solution for T-12-wood (MSE: 4808.67)

pgm

exp-002-2d-reachability-T-03-grid-sol.pgm

Best solution for T-03-grid (MSE: 11007.93)