Banjo-Kazooie PC port developer blasts AI vibe coding as retro gaming fills with low-quality ports
Developer Dario, a major figure in the emerging PC fan port community, recently published a lengthy, harsh critique of the emerging practice of using LLMs to decompile, recompile, and create new ports for retro video games. His thread on Bluesky provides a detailed explanation of the practice and its pitfalls. Read Entire Article
Recently, a developer in the retro gaming community has voiced strong opposition to the use of AI coding tools in PC ports of retro console games. Dario, a prominent figure in the community, has published a scathing critique of the practice, which involves using large language models (LLMs) such as Claude to decompile, recompile, and create new ports for retro video games. In his thread on Bluesky, Dario details the pitfalls of this emerging trend, which has divided retro gaming enthusiasts.
The growing number of retro enthusiasts has decompiled the source code of classic games to study their inner workings and reverse-engineer them. Developers then used this knowledge to port the games to modern platforms, often by recompiling the reverse-engineered code. This process allows for higher resolutions, higher frame rates, mods, and other features not possible with emulation.
However, many of these projects rely on AI tools. HarbourMasters, the team behind popular ports of Super Mario 64, Mario Kart 64, Star Fox 64, The Legend of Zelda: Ocarina of Time, and Majora's Mask, has been using LLMs for years. Besides ethical concerns and the environmental impact of AI, developers argue that LLM-based coding produces verbose and indecipherable code.
Dario, who led a Banjo-Kazooie port earlier this year, also believes that many vibe-coded projects come from amateur developers with limited knowledge of the games they are porting. He disputes the claim that using AI saves time, stating that the damage done by AI in his port is far greater than anticipated. He finds the entire organization and development of AI-based ports to be a "unsalvageable train wreck."
Despite the controversy, some projects have adopted varying policies and disclaimers regarding the use of LLMs. Some require developers to explain their code, while others allow AI use only for non-coding tasks.
Written by urgent.news from TechSpot's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.