Unlike a standard chess interface, this project is a self-learning ecosystem. It uses Stockfish to simulate Computer vs. Computer (CvC) matches.
The core idea is the Hive Mind Memory:
- Every move calculated at Depth 32 is not forgotten.
- The result is stored in
brain.js. - Over time, the simulation stops "thinking" and starts "remembering," making the bots effectively invincible for all cached positions.
- 🤖 Automated CvC Evolution: Leave it running in the background. The bots will play infinitely, constantly expanding the
brain.jsdatabase. - 🎮 Human "Chaos" Factor: At any point, you can pause the simulation, take control of one side, and make a manual move. See if the bot can adapt to your strategy or if it already has a "perfect" response in its memory.
- ⚡ Zero-Latency Recall: If a position is in the "brain", the move is executed instantly without any CPU load.
To get the Hive Mind running on your machine, follow these steps:
-
Clone the Repository:
git clone [https://github.com/AmigoSmuCha/chess.git](https://github.com/AmigoSmuCha/chess.git) cd chess/cviceni/chess_project -
Add the Engine (Crucial): The Stockfish engine is too heavy for Git, so you need to provide it manually:
- Download Stockfish from stockfishchess.org.
- Place the executable into the
cviceni/chess_project/folder. - Rename it to
stockfish(Linux/Mac) orstockfish.exe(Windows).
-
Launch: Open
index.htmlin your browser (Chrome or Edge recommended for best JS performance).
Running Stockfish at Depth 32 in a browser is heavy. To build the "brain" efficiently, we recommend:
| Hardware | Requirement |
|---|---|
| CPU | High Priority: 8-Core+ (Ryzen 7 / i7) with AVX2 support. |
| Memory | 8 GB RAM minimum for large hash tables. |
| Storage | SSD is highly recommended for frequent JSON I/O operations. |
Do you have a brain.js file with thousands of Depth 32 moves? I'd love to merge it! Feel free to open a Pull Request with your learned data so we can build a truly global chess memory.
Created by AmigoSmuCha