MacOS on PC becomes possible again thanks to Apple Silicon translator for x86-64

Science & Technology

Stela translation layer lets standard PCs boot macOS on Apple Silicon

Stela, a Linux-based UEFI app by Tyler Smith, boots macOS 27 on x86-64 PCs by emulating an M2 Mac Mini and translating Arm and Metal calls, though it is 'slow and not very stable.'

By
Tom Whitfield
Filed
Channel
Science & Technology
Read
3 min read

A new open-source project called Stela lets an ordinary x86-64 PC boot macOS 27 by presenting itself to the operating system as a virtual M2 Mac Mini. The tool translates Apple Silicon's Arm instructions and Metal graphics calls on the fly, reviving the Hackintosh practice that largely died when Apple dropped Intel processors in 2020.

Apple began moving its Mac line to its own Arm-based designs in 2020. The company's system-on-chip designs offered strong performance per watt, which suited its laptops well, and Apple shifted away from Intel-based Macs as the silicon improved. That transition ended the era of enthusiasts running macOS on standard PC hardware. Stela, built by developer Tyler Smith, reopens that door.

How does the translation work?

Stela is a Linux-based UEFI application. It loads macOS 27 and handles two layers of translation: CPU instructions and graphics calls. For the processor, it converts AArch64 instructions into x86-64 using a combination of ahead-of-time and just-in-time (JIT) compilation. For graphics, it translates Apple's Metal API calls into NIR instructions used by Mesa, the open-source graphics stack.

The macOS image itself remains untouched. Smith explained the point in the project's Discord server: "Stela provides the translation and hardware behavior underneath that lets macOS go through the exact same restore process you might do on a real Mac." In plain terms, the operating system believes it is installing on genuine Apple hardware.

The project is young. Smith shipped three releases in barely half a day after the initial 0.1 version, and he warns that the software "is slow and not very stable." That caution has not slowed adoption. The Discord server has drawn a large influx of users, many posting screenshots of their PCs getting partway through a macOS boot, and most asking whether their specific hardware is supported.

What hardware does it need?

The requirements are modest by modern standards:

  • An x86-64-v2 CPU — roughly a 2008 Intel Core chip or a 2011 AMD FX processor at minimum; AVX2 and AVX-512 "are used when present"
  • 8 GB of RAM
  • UEFI firmware support
  • A GPT partition or an entire drive
  • An Nvidia GPU from the 16xx series through current 5000-series models

Networking support is partial. There is no Wi-Fi yet, and Bluetooth support is limited. The list of working Ethernet controllers is broad, spanning Intel, Realtek, Broadcom, Aquantia, Atheros and even USB Ethernet adapters. Smith says he intends Stela to eventually support every piece of hardware the Linux kernel supports.

Getting started requires downloading Stela from its release pages, preparing a bootable USB stick and supplying your own copy of macOS. Stela handles the instruction translation from there.

Could this become practical?

The performance question is structural. A translation layer imposes overhead on every instruction, and some code paths resist translation entirely. Browser JavaScript engines, for example, compile code at runtime through their own JIT systems, so that code never exists in translated form on the disk image. Pre-translating large parts of macOS with AI automation is conceivable, but it faces this runtime-code problem.

There is also a cryptographic barrier. macOS must accept and validate the software it runs, and a translation layer may require deeper modification to pass those checks. Stability, not raw speed, remains the near-term constraint on the project.

For now, Stela is a proof of concept with a fast release cadence and an eager user base. Whether it matures into a practical route to macOS on commodity hardware will depend on how quickly Smith and contributors can close the gap in graphics performance, hardware coverage and boot reliability.

Source: Tom's Hardware

Share this article:

More from Tom Whitfield

Tom Whitfield

Show full bio

Staff writer covering companies, deals and strategy in Semiconductors, chip manufacturing, hardware, AI compute at Chip Dispatch.

919 articles

Related articles

« Previous articleNext article »