🌌 Linear Algebra Visual Engine

Project 01 — 2D Linear Transformation Animator

"Watch space bend, rotate, shear and collapse in real time."

---

🎬 Phase 2 — Blender bpy Cinematic Renders

1920×1080 · 60fps · EEVEE · Emission Shaders · Geometry Nodes · Shape Key Animation

Rotation 90°	Shear	Scale 2x
det = 1 · area preserved	det = 1 · grid slides	det = 4 · area ×4

det = 0 Collapse	Reflection x	det < 0 Flip	Projection x
Space folds to a line	det = -1 · mirrors	det = -1 · flips	det = 0 · collapses to axis

---

🎬 Phase 1 — Python + Matplotlib Preview

Rotation 90°	Shear	Scale 2x

det = 0 Collapse	Identity

---

🧠 What This Proves

This is not a tutorial project. Every line of code here visually proves a concept from linear algebra — the same concepts that power every AI model, graphics engine, and physics simulation on the planet.

Concept	What You See
Basis vectors î ĵ	Red + teal arrows morphing live with the grid
Matrix × vector	Every point in 2D space multiplied by M
Linear transformation	Grid morphing smoothly from identity → M
det(M) as area scale	Grid stretches/shrinks by exactly det(M)
det = 0 collapse	Space folds into a line — no inverse exists
det < 0 flip	Orientation inverts — space turns inside out
Composition M₂M₁	Two transforms chained, proves det(M₂M₁) = det(M₂)·det(M₁)

---

🏗️ Build System — 2 Phases

Phase 1 → Python + NumPy + Matplotlib    ✅ Complete
   Prove the math works. Pure logic. Runs in terminal.

Phase 2 → Blender 5 + bpy               ✅ Complete
   Same math. GPU-rendered cinematic dark void.
   Shape Keys + Geometry Nodes + EEVEE + 1080p 60fps.

---

📁 Project Structure

LA-Visual-Engine/
│
├── 📄 README.md
├── 📄 requirements.txt
├── 📄 LICENSE
│
├── 🐍 Phase_1_Logic/
│   ├── matrices.py           ← All 8 transformation matrices + presets dict
│   ├── math_utils.py         ← lerp, compose, det_status, verify_composition
│   └── preview.py            ← Matplotlib CLI animator — run this
│
├── 🎨 Phase_2_Blender/
│   ├── scenes/
│   │   └── transform.py      ← Entry point — change PRESET_NAME and run
│   └── utils/
│       ├── materials.py      ← Emission shader system
│       ├── scene_builder.py  ← Grid, arrows, camera, Geometry Nodes thickness
│       └── animator.py       ← Shape Key animation + Blender 5.0 interpolation
│
└── 📸 docs/assets/
    ├── phase2_renders/       ← 7 Blender cinematic renders (GIF)
    └── *.gif                 ← Phase 1 Matplotlib previews

---

🚀 Quick Start

Phase 1 — Python Logic

pip3 install -r requirements.txt
cd Phase_1_Logic
python3 preview.py

Phase 2 — Blender Cinematic

1. Open Blender 5.x
2. Scripting tab → Open → Phase_2_Blender/scenes/transform.py
3. Set PRESET_NAME = "shear"  (or any of the 8 presets)
4. ▶ Run Script
5. Render → Render Animation  (Ctrl+F12)

8 presets available:

identity       → No change          det =  1
rotation_90    → Rotate 90°         det =  1  (area preserved)
shear          → Horizontal shear   det =  1  (area preserved)
scale2x        → Scale 2x           det =  4  (area ×4)
reflection_x   → Reflect over x     det = -1  (orientation flips)
projection_x   → Project to x-axis  det =  0  (collapse!)
det_0_collapse → Space collapses    det =  0
det_neg_flip   → Orientation flip   det = -1

---

✅ Verification

A successful Phase 1 run should:

• print the available preset list in the terminal
• open a Matplotlib window after you choose a preset like shear or rotation_90
• animate the grid and basis vectors from identity to the selected matrix
• update the chart title with determinant status as the animation progresses

---

🔬 The Linear Algebra Explained

What is a Linear Transformation?

[a  b] [x]   [ax + by]
[c  d] [y] = [cx + dy]

Every point in the grid moves — lines stay parallel, origin stays fixed.

Why does det(M) matter?

det = 1   →  Area unchanged    (rotation, reflection)
det = 4   →  Area ×4           (scale 2x in both directions)
det = 0   →  Area = 0          (space collapsed — no inverse!)
det < 0   →  Orientation flip  (space turns inside out)

The composition rule

M_total = M₂ · M₁     (order matters — NOT commutative)
det(M₂ · M₁) = det(M₂) × det(M₁)

---

📐 Part of the Simulation Architect Path — 10 Projects Total

#	Project	Core Concept	Status
01	2D Linear Transform Animator ← here	Vectors, Det, Basis	✅ Both phases done
02	Coordinate System Translator	Change of Basis	✅ Both phases done
03	Geometric Linear System Solver	Cramer's Rule	✅ Both phases done
04	Eigenvector Explorer	Eigenvalues, Stable Axes	✅ Both phases done
05A	Solar System Simulator	Real Physics + NASA JPL API	⏳
05B	PROJECT VOID	Non-Uniform Gravity + Custom A*	⏳
06	Neural Network Visual Simulator	Backprop, PyTorch	⏳
07	Optical Fiber & Internet Simulator	Wave Physics	⏳
08	VOID AI — RL Navigator	Gymnasium + SB3	⏳
09	Omniverse Digital Twin	OpenUSD + NVIDIA Omniverse	⏳

---

🧰 Tech Stack

Tool	Purpose
Python 3.12+	Core language
NumPy	Matrix operations, vector math, LA engine
Matplotlib	Phase 1 — 2D visualization
Blender 5 + bpy	Phase 2 — 3D GPU-rendered cinematic
Geometry Nodes	Procedural 3D tube generation from math lines
Shape Keys	Crash-proof vertex animation in Blender
Taichi Lang	GPU physics (from Project 05A onwards)
PyTorch	Neural networks + RL (Projects 06, 08)
NVIDIA Omniverse	Endgame simulation platform (Project 09)

---

👨‍💻 Author

Divyansh Ailani — Simulation Architect in progress

BCA Student · Kanpur, India → The World

"Mathematics is the language of the universe. I am learning to read it."

---

Part of the Simulation Architect Path — from linear algebra to NVIDIA Omniverse. 🌌