{"slug": "ancient-roman-board-game", "title": "Ancient Roman Board Game", "summary": "Researchers used artificial intelligence to reconstruct the rules of Ludus Coriovalli, an ancient Roman board game lost for over 1,800 years, by analyzing wear patterns on a carved stone found in Heerlen, the Netherlands. The study, published in Antiquity, used AI simulations to match the stone's geometry to a four-versus-two blocking game, now playable online.", "body_md": "# Ludus Coriovalli — Ancient Roman Board Game Reconstructed by AI\n\nLudus Coriovalli is an asymmetric strategy board game from ancient Rome, in which four Hounds attempt to block two Hares from moving. It was lost to history for over 1,800 years, until a team of researchers used artificial intelligence to reconstruct its rules from the physical traces left on a mysterious carved stone.\n\nIn 2025, a landmark study published in the journal *Antiquity* demonstrated that the geometric pattern incised on a piece of limestone found in Heerlen, the Netherlands, corresponds to a playable board game. By combining microscopic use-wear analysis with AI-driven simulations, researchers identified the wear patterns left by centuries of play and matched them to a blocking game.\n\n## The Discovery\n\nThe story begins with a mysterious stone held in the collection of Het Romeins Museum in Heerlen, the Netherlands. Catalogued as Object 04433, it is a worked piece of white Jurassic limestone (212 × 145 × 71 mm, 3.38 kg) quarried from Norroy in north-eastern France. Its top surface bears an incised geometric pattern, and microscopic examination revealed that the surface along one diagonal line was noticeably smoother than the rest — consistent with repeated abrasion by game pieces during play.\n\n## Coriovallum — A Roman Town\n\nCoriovallum was a Roman town in the province of Germania Inferior, founded under Emperor Augustus (27 BC – AD 14) and inhabited until AD 476. Its remains lie beneath present-day Heerlen. At its peak, the town covered over 48 hectares and was a major centre for Roman pottery production.\n\n## AI-Driven Rule Reconstruction\n\nNo written rules survived. Researchers used the Ludii General Game System with Alpha-Beta search agents, running 1,000 simulated rounds per ruleset at one second per move. Nine game configurations matched the physical wear patterns — all were blocking games, with the four-versus-two format being most frequently matched. This site faithfully reproduces one of these AI-validated configurations.\n\n## How to Play\n\nThe game is played on a graph of 19 nodes connected by lines. Four Hounds start at the top and try to surround the two Hares so they cannot move. Hares start at the middle-right and try to survive for 150 turns or achieve threefold repetition. The Hounds have numerical superiority but must coordinate carefully, while the Hares are outnumbered but more agile.\n\n## Features\n\n- Play against AI (3 difficulty levels using Alpha-Beta search)\n- Play against another person (local PvP)\n- Multiple languages supported\n- No registration or downloads required — play instantly in your browser\n\n## Academic Reference\n\nCrist, W., Piette, É., Jeneson, K., Soemers, D., Stephenson, M., van Goor, L., & Browne, C. (2025). Ludus Coriovalli: using artificial intelligence-driven simulations to identify rules for an ancient board game. *Antiquity*, 100(409). [doi:10.15184/aqy.2025.10264](https://doi.org/10.15184/aqy.2025.10264)\n\n## Privacy Policy\n\nThis site uses Google AdSense for advertising. Google and other third-party vendors use cookies to serve ads based on your prior visits. You may opt out of personalized advertising at [Google Ad Settings](https://adssettings.google.com/) or [aboutads.info](https://www.aboutads.info/choices/).\n\nThis site does not collect personally identifiable information. Game data is stored locally in your browser.", "url": "https://wpnews.pro/news/ancient-roman-board-game", "canonical_source": "https://ludus-coriovalli.web.app/", "published_at": "2026-07-09 20:47:45+00:00", "updated_at": "2026-07-09 21:07:12.925140+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-research"], "entities": ["Het Romeins Museum", "Heerlen", "Ludii General Game System", "Antiquity", "Coriovallum", "Germania Inferior", "Google AdSense"], "alternates": {"html": "https://wpnews.pro/news/ancient-roman-board-game", "markdown": "https://wpnews.pro/news/ancient-roman-board-game.md", "text": "https://wpnews.pro/news/ancient-roman-board-game.txt", "jsonld": "https://wpnews.pro/news/ancient-roman-board-game.jsonld"}}