GoKawiilDeveloper compiles neural networks into DOOM levels using voodoo dolls for arithmetic
A new open-source tool called Doomformer converts integer neural networks into ZDoom maps that run inside the ViZDoom engine, using the game's 'voodoo doll' player copies to perform calculations and sector light levels and heights to store values. The project includes a demo where a player draws a digit on a 3x3 grid of floor cells, which the level then reads and classifies using 1,355 voodoo dolls, with outputs verified bit-for-bit against a NumPy reference. It is available via pip install and includes a diagnostic tool, 'doomformer doctor', to check dependencies like ViZDoom and Freedoom.
GoKawiil's interpretation of the reporting above, not reported fact.
The project is a novelty demonstration of Turing-completeness and esoteric computation rather than a practical machine learning tool, given it runs roughly 23,000 to 65,000 times slower than NumPy. It could be seen as part of a long tradition of hobbyist projects that repurpose old game engines for unconventional computing tasks, showcasing creative engineering rather than performance gains.
- Doomformer compiles integer neural networks into ZDoom maps executed within the ViZDoom engine.
- The system uses voodoo dolls for arithmetic and sector properties like light levels to store numeric values, verified exactly against NumPy.
- Performance is drastically slower than conventional computing, with a digit classification taking 8 to 163 seconds depending on settings.
Logitech G DOOM Eternal Gaming Keyboard — If you're diving deep into DOOM's engine internals like this wild Transformer-in-ZDoom project, you'll want solid gear for marathon coding and testing sessions. A mechanical gaming keyboard gives tactile precision for scripting ACS probes and navigating ViZDoom test runs. Plus it's just fun to have DOOM-themed hardware while you're hacking neural nets into sector light levels.
See Logitech G DOOM Eternal Gaming Keyboard on Amazon → Affiliate link — we may earn a commission on purchases, at no extra cost to you. Product picked by AI based on this article; it is not a tested recommendation.Source: github.com, 2026-10-29
Published there as: “Show HN: I Turned DOOM into a Transformer”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.