Maniac Mansion 3D Reconstruction
github.com/s-macke/maniac-mansion-3d- Category
- Entertainment
- Rank
- No. 209Apps board
Previous survey · No. 193 ·
- Type
- TOOL
- GitHub
- 6 stars
- Date
About
This project reconstructs the rooms of the 1987 DOS EGA game Maniac Mansion in 3D by feeding its original fixed-viewpoint screens into an AI model (GPT-6 Astra) and Blender, then assembling the results into a single walkable scene of the mansion viewable in a browser. It extends an earlier experiment that reconstructed a real chair from photos the same way, and is explicitly not a playable remake, just the original 2D screens turned into a space you can walk through.
At a glance
Maniac Mansion in 3D turns the original EGA backgrounds into a connected first-person house. It is worth opening if you want to see how a painted room becomes a space you can walk through.
Good fit
Fans of the original art and game designers interested in reconstructing spaces from fixed camera views. Press Start offers a broader tour of game history.
Consider something else when
Skip it if you want to play the original adventure, solve puzzles or verify a canonical floor plan.
Start here
Open the hosted living-room link from the README and use Original to compare the 3D room with its EGA background. Then try a doorway to see how the reconstruction connects rooms.
Cost and requirements
The creator provides a public browser walkthrough and documents keyboard and touch controls. The local build requires Blender tooling and can take tens of minutes according to the README.
Limits to know
The creator says unseen walls, room dimensions and some connections are interpretations. The original artwork belongs to its rights holders. Room navigation and rebuild instructions were reviewed from source.
Related work
- Press Start (context)
Sources checked 2026-11-05. Source-based assessment.
Why it made the leaderboard
You can walk through every room of the original Maniac Mansion mansion in 3D in your browser, rebuilt purely from its old EGA screenshots using an AI reconstruction pipeline.
Tags
Tech Stack
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.