Retro Context
The first Grand Theft Auto began with a driving problem: its cars did not feel fun to control. Rockstar veteran Obbe Vermeij has credited programmer Pat Kerr with helping turn that around. Kerr was not even on the development team, but he had a knack for physics and supplied driving code that Vermeij recalls as fixing the cars. It is a small but telling piece of game history: an idea from outside the team helped shape how a major series first moved through its world.
Kerr says the starting point was a 2D physics simulation he wrote over a weekend in late August 1996. That prototype later became the basis for Grand Theft Auto’s vehicle system. At its heart was a classical simulation of rigid bodies moving in two dimensions. Kerr says this method was not yet common in games, though there were notable exceptions; he contrasts it with the simpler “point physics” approach he says many game vehicles used at the time. The distinction helps explain why a modest prototype could matter: its underlying model gave the cars a different foundation.
The driving feel did not come from rigid-body physics alone. Kerr added a vehicle layer built around a simplified approximation of tyre behavior. He knew the model was not technically or physically accurate, describing it as a semi-educated guess. Still, it seemed to produce a useful result. That mix of a clear physical core and practical approximation is part of the prototype’s story: game systems often need to feel convincing to play, even when they are not precise simulations of the real world.
Why It Matters
Kerr’s new playable demo gives players a way to experience those old ideas rather than only read about them. In a video shared on October 7, 2026, he presented it as a “remaster” of the roughly 30-year-old physics prototype. Vermeij has also remarked on how much the new version feels like the original driving. For curious players, that makes the demo a hands-on look at how an early technical choice can become part of a game’s identity.
There is a useful caveat around the word “remaster.” Kerr describes the project as his own independent work and says it is not associated with Rockstar, Take-Two, or Grand Theft Auto. The demo revisits the prototype and its ideas; it is not presented as an official rerelease of the original game. Keeping that distinction in mind helps set expectations, while still leaving room to enjoy the connection to GTA’s development history.
The story may also interest people who care about how games are built, not just how they look. Kerr’s account shows a practical chain: an experiment made in 1996 fed into a vehicle system, and decades later the underlying approach has been made playable again. The available details do not explain every step of the original development or how the demo compares feature by feature with GTA, so it is best treated as a focused glimpse into one part of the game’s creation.
Z-retro View
Kerr’s revival offers a neat bridge between a foundational game and the ideas that helped it work. Its appeal is less about recreating all of Grand Theft Auto and more about seeing how one programmer’s physics experiment contributed to its driving. The demo also gives context to Vermeij’s recollection that Kerr’s code made the cars fun. It is a compact example of how a technical contribution can outlive its original moment and remain interesting to players years later.
For retro readers, the most rewarding approach is to take the demo as both a playable curiosity and a development artifact. Kerr is open about the tyre model being approximate, and that honesty is part of what makes the story useful: a system can be deliberately imperfect and still serve its purpose in a game. The independent status matters too, since it clarifies who made this new version and how it relates to the franchise. Together, those details make the project a thoughtful look back at an early piece of GTA’s driving history.



