Machine For Pigs Redux

This is a massive overhaul of the public source code release of the base engine. There is a an enhanced standard backend and an option to play the game with full raytracing. 

Graphical Improvements

• Path-traced indirect diffuse and specular lighting, replacing the earlier surfel-based global illumination system.
• NRD denoising for indirect lighting, direct lighting, and ray-traced water reflections.
• AMD FSR temporal upscaling on Windows and Linux, plus Intel XeSS on Windows, with quality presets and render-scale controls.
• Updated campaign lighting, lamp and lantern assets, and shadow softness through map and entity patches.
• Improved texture filtering and mipmap handling, along with updates to transparency, particles, fog, and water.

Editors and tools

The project also includes the Level Editor, Model Editor, Material Editor, and Particle Editor. The Level Editor has a scene outliner, in-place editing of entity contents, and an MCP server for external tools and automation. Asset reload support helps changes appear across editors, and map/entity delta tools allow patches to be distributed without replacing entire source maps.

The editors and tools are still under development.

Latest Releases

GitHub releases and prerelease builds

Installation

1. Download and extract the archive for Windows or Linux.
2. Copy all of its contents into your Amnesia game directory, alongside the existing game assets. Keep the archive's directory structure, including `compiled_shaders/`, bundled libraries, and any resource or map/entity patch folders.
3. Launch `Amnesia.exe` on Windows or `Amnesia` on Linux from that directory.

Compatibility and feedback

Testing has focused on the base game. Custom stories and other mods may have compatibility issues, particularly where they overlap with the renderer or patched resources.

Please report crashes, rendering artifacts, broken gameplay, or mod compatibility problems on the issue tracker. Include your operating system, GPU, build version, and steps to reproduce the problem when possible.

Ideas and ongoing work are also tracked there. Contributions are welcome, whether you want to fix something I missed, improve the tools, or work on a new feature. Feel free to submit a pull request.