Unofficial NVIDIA eGPU support is now a reality for Mac users. It’s all thanks to the developers and researchers that congregate over at, a community for eGPU coverage and support across Mac, Windows, and other platforms. While not perfect, the results of my testing look very promising. Let me preface this post by saying that NVIDIA eGPU support for macOS is still not officially supported by Apple, and the workaround script used to provide support is still in its alpha stages. Even so, I’ve been impressed by the script’s ease of use, and the performance that I’ve seen thus far.
It means that macOS users can now enjoy eGPU setups with cards from Nvidia’s Pascal lineup, including the GTX 1070, 1080, and venerable. And it’s not just for Thunderbolt 3 Mac users, either. The script allows Mac owners to work around restrictions that eliminated support for Thunderbolt 2 eGPUs as well, allowing more users to join in on the graphics-accelerated fun. Watch our video for a hands-on look.
Enabling NVIDIA eGPUs on macOS With macOS 10.13.4, yet there was one glaring omission with the release — a lack of NVIDIA support. The super smart folks that post over at eGPU.io, including well-known eGPU community staples like fr34k and goalque, have worked hard on bringing unofficial support to macOS.
Nvidia Mac Os X Drivers
Apr 14, 2018 - All that is to say that the time is ripe for eGPUs to come to the Mac. This is disappointing, as many users deem Nvidia GPUs to be preferable.
Macbook Pro External Gpu Nvidia
The result is a relatively easy-to-use script (one that’s still in the pre-beta stages) that makes it somewhat trivial for users comfortable with interfacing with a Terminal window to enable NVIDIA eGPU support on their Macs. In my hands-on walkthrough below, I show you how simple it is to take a fresh and clean macOS 10.13.4 install, and add unofficial NVIDIA GPU support to the mix. It’s literally just a matter of disabling SIP, and running the provided script in a Terminal window. After answering a few prompts and rebooting, NVIDIA support becomes available.
Video walkthrough With that said, this is an alpha script, and there are potential bugs and glitches. Those that feel uncomfortable navigating around Terminal should probably stay away, at least for now. The developer behind the script notes that running it could potentially cause problems with your Mac, so run at your own risk, and be sure to back up your system before diving in. From my brief testing, it’s immediately evident that Pascal-based cards are a better fit than AMD cards if you require raw OpenGL gaming performance. I have one of the weakest Pascal cards out there, the, and even it handled OpenGL synthetic benchmarks better than my AMD RX 580. It also performed rather admirably when compared to the Vega 64, a card that these days.
Nvidia Gpu For Machine Learning
My 1050 Ti doesn’t even require a power adapter, as it’s powered solely from the PCIe slot, yet even with its power-sipping TDP it still resulted in relatively smooth ultra/1080p performance in both Heaven and Valley Unigine benchmarks. If I’m getting this kind of performance out of the Geforce GTX 1050 Ti, it goes without saying that the GTX 1080 would likely do wonders for a Mac + eGPU gaming setup. OpenCL performance, as you might expect, is a different ballgame, however. The NVIDIA GTX 1050 Ti didn’t perform nearly as well, and lagged well behind the AMD RX 580 and Vega 64.
Higher performance cards like the 1070 and 1080 would yield better performance, but. The eGPU enabler script has many options available to customize eGPU installations, and goes beyond mere NVIDIA support. For example, it even lets users with Thunderbolt 2-based eGPUs in on the fun, something Apple recently put the kibosh on.
It also can enable eGPUs with, which have never been recognized by the Mac without workarounds. Be sure to, which contains many of the flags and settings needed to customize the installation. And if you’re looking for the best eGPU chassis that Apple officially recommends, then look no further than the. Check out our hands-on for more details.
What do you think about unofficial NVIDIA eGPU support on Mac? Is it something you’re potentially interested in?