MazeCruncher
A cool, fast maze generator and solver written in Rust
Install / Use
/learn @ihawn/MazeCruncherREADME
MazeCruncher
Welcome to maze cruncher!
<a href="https://github.com/ihawn/MazeCruncher/releases/tag/MazeCruncher">Download Here</a>
Usage
- To get started, just run the standalone .exe or compile and run the source code yourself.
- Both the solved and unsolved mazes are saved as a png in the directory that the .exe is run from.
- A solution animation can be displayed too but it will slow the algorithm considerably, though it looks very interesting.
- The algorithm considers the maze solved when it has gone from the upper left corner to the lower right corner.
Settings
-
Generation Algorithm: Current available algorithms are Growing Tree, Kruskal, Prim, Hunt and Kill, and Binary Tree.
-
Solution Algorithm: Current available algorithms are Depth First Search, Breadth First Search, Dead-end Filling, A*, double A*, Dijkstra, and Tremaux.
-
Maze Decimation Probability: Setting this to zero will ensure the maze has only one solution. Setting this to anything greater than zero will increase the probability of more paths.
-
Save Maze: Responding yes will save the solved and unsolved maze to the same directory as the executable.
-
Maze Size: The size of the maze. Large values (>10,000) will take awhile and really large values (>50,000) can use more than 64 GB of memory. Also keep in mind that the solved and unsolved maze image will be saved with dimension n x n.
-
Show Animation: Whether or not to show the animation of the algorithm solving the maze. It looks cool but takes way longer than just letting it run. Automatically disabled for mazes with size >2048.
-
Animation Scale: The scale factor that the maze animation will display at. 1 = one pixel per maze cell. Make sure animation_scale*maze_size < your display resolution height.
-
Animation Speed: The speed scale for the maze animation. 1 = n cell traversals per second where n = your monitor's refresh rate.
A small maze generated with Hunt and Kill and solved with Dijkstra's

A maze solved using Depth First Search with size = 1024:

Related Skills
node-connect
350.8kDiagnose OpenClaw node connection and pairing failures for Android, iOS, and macOS companion apps
frontend-design
110.4kCreate distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
openai-whisper-api
350.8kTranscribe audio via OpenAI Audio Transcriptions API (Whisper).
qqbot-media
350.8kQQBot 富媒体收发能力。使用 <qqmedia> 标签,系统根据文件扩展名自动识别类型(图片/语音/视频/文件)。
