Reverse-engineer a 1997 PC game and make it run on modern hardware
Claude Code got an old DirectX game running on modern hardware, decompiled much of it, built a headless parallel runner, and now writes full level guides with screenshots.
Gemini-powered SIMA 2 navigates Goat Simulator 3, and a human player can direct it to improve its performance.
Gemini-powered SIMA 2 navigates Goat Simulator 3, and a human player can direct it to improve its performance.
Google DeepMind built SIMA 2, a Gemini-powered agent that moves around the world of Goat Simulator 3. It is meant to learn and adapt in open-ended, complicated games. A human player can direct it and help improve its performance. The article gives no detail on the mechanics.
The source states no results or numbers. It only reports that the agent is navigating Goat Simulator 3 and that human direction can improve it.
The article is a brief secondhand report of MIT Technology Review's coverage and has no setup details, so check that original piece before trying to copy this.
I have [game name] running on this desktop. Navigate the world using the keyboard and mouse to reach [destination or goal], and I'll give you extra directions as you go, such as [example instruction]. Check in with me whenever you get stuck, and stop when you've reached the goal or after [number] minutes.
Claude Code got an old DirectX game running on modern hardware, decompiled much of it, built a headless parallel runner, and now writes full level guides with screenshots.
Make pulls contracts from Gmail, Gemini extracts the term and notice period, and a Sheets register emails the owner 60, 30 and 7 days before each deadline.
A self-hosted n8n job reads your calendar, unread email and Notion tasks at 7am and sends one short brief of what matters today.