Google executive uses Claude to port Command & Conquer Generals: Zero Hour to iPhone and Mac natively

A Google executive used AI-assisted tooling to port the 2003 real-time strategy game Command & Conquer Generals: Zero Hour to iPhone, iPad, and Mac as a native ARM64 build, not an emulator. The project demonstrates how AI can help bridge legacy PC engines to modern mobile hardware, but it comes with licensing and deployment caveats that AI builders should understand.

What happened

Google lead product and design executive Ammar Reshi used Claude Fable 5 to port Command & Conquer Generals: Zero Hour to iPhone, iPad, and Mac. The actual 2003 game engine was compiled natively for ARM64, meaning it runs directly on Apple Silicon and iOS devices without emulation or cloud streaming.

The port relies on EA's GPL source release and builds on existing community work. The original engine expected a writable PC-style file system, but iOS apps live inside a locked-down, code-signed bundle. Reshi redirected save files, cache paths, and configuration writes to the app bundle.

The graphics pipeline required significant work. Zero Hour was built around DirectX 8, while Apple devices use Metal. The project routes DirectX 8 through DXVK (DirectX to Vulkan) and then MoltenVK (Vulkan to Metal). Touch controls were added for the RTS gameplay: tap selection, drag-box selection, two-finger scrolling, pinch zoom, and long-press actions. Campaign, skirmish, and Generals Challenge all work with these controls.

Why AI builders should care

This case illustrates how AI-assisted development can enable non-trivial cross-architecture work. Porting a 2003 PC game to mobile natively is a complex task involving graphics API translation, input system redesign, and sandbox workarounds. Claude Fable 5 helped make this feasible for a single developer.

For AI builders, this shows that AI coding assistants can handle more than boilerplate or simple CRUD apps. They can assist with legacy codebases, platform-specific APIs, and performance-critical translation layers. The project is open-sourced on GitHub, providing a real-world example of AI-assisted porting.

Practical implications

The port demonstrates a pattern for bringing classic PC games to mobile: use GPL source releases, community modernization work, and open-source graphics translation layers. The iOS sandbox was handled by redirecting file system writes, which is a common pattern for any app that needs to write data on iOS.

For game preservation, this approach is more sustainable than emulation because it runs natively with full performance. The project also shows that AI can accelerate the porting process, potentially making it economical to bring more legacy titles to modern platforms.

Caveats

There is no official confirmation from EA or Google about this port. The project uses EA's GPL source release, which means it is open-source and may have licensing restrictions for commercial distribution. The port is not available on the App Store; users would need to build it from source or sideload it, which requires developer accounts and code-signing.

The port depends on community work and may not be maintained long-term. Graphics translation layers like DXVK and MoltenVK add complexity and may not support all game features perfectly. Multiplayer functionality is not mentioned in the available sources.