Ampere Computing Publishes Guide For Steam Play Games On Their AArch64 Server CPUs

While Ampere Computing's wares with the Altra (Max) and forthcoming AmpereOne families of AArch64 server processors are designed for the data center, if you feel so inclined they have published a guide on being able to run Steam for Linux on these ARM64 processors -- including Steam Play (Proton) for enjoying Windows games on these Linux servers.

This summer Ampere Computing launched a GitHub repository where they are maintaining a guide on how to run Steam for Linux on their AArch64 platforms, including with Steam Play / Proton. The guide is being maintained for use with the Ampere Altra / Altra Max and AmpereOne hardware, but it can apply to other 64-bit Arm platforms as well -- but you'll need a sufficiently powerful processor in order to properly enjoy the gaming experience. Plus with the need for accelerated 3D OpenGL/Vulkan graphics, it really needs to be more of an Ampere workstation system rather than a headless Ampere server.

Ampere's guide is based around using the Ampere Altra Developer platform and running with a NVIDIA RTX A6000 series graphics card that does have AArch64 proprietary drivers.

For handling the emulation Ampere Computing's guide is making use of Box86 and Box64 for running the Steam x86 binaries and other x86/x86_64 games. Phoronix readers will also know there is also the likes of FEX-Emu and Hangover for improving the Linux binary experience on AArch64, but for the purposes today of gaming on Ampere workstations it appears Box86/Box64 is the best supported route given the endorsement in Ampere Computing's guide.

With accelerated AArch64 Linux graphics drivers and setting up Box86/Box64 emulation, it's then possible to go ahead and install Steam for Linux. If enabling Proton within Steam it's then possible to enjoy Windows-only x86/x86_64 games on the Ampere AArch64 workstation/server processors. No word though on how the performance is for such a setup.

Those interested can see the guide via AmpereComputing/Steam-on-Ampere via GitHub.