Apple’s M1 Pro and M1 Max chips were a GPU-shaped warning sign for Nvidia and AMD last year, and now Apple has turned up the heat with its new M1 Ultra. Apple claims it will rival Nvidia’s massive RTX 3090 graphics card, which is the fastest GPU on the market today. But how does Apple think it can beat an RTX 3090? It turns out that last year’s M1 Max contains a secret sauce.
Apple’s new M1 Ultra is a surprising combination of two M1 Max dies, fused into a single powerful chip. The M1 Max has a secret high-speed interface that allows Apple to combine two chips into one. The result is an M1 Ultra chip with dual CPU cores, double the amount of memory, double the memory bandwidth, and most importantly, dual GPU cores.
Apple calls this combination UltraFusion, and it’s basically Apple’s own 2.5D chip packaging implementation. The chip industry has been turning to chiplets to design processors for years, with AMD’s Zen 2 and Zen 3 based Ryzen chips leading the pack for modern chiplet designs and performance until recently. Apple rivals such as Intel, Samsung and Qualcomm are in the early stages of collaborating on a new standard that will allow companies to build processors from Lego-like chiplets. Apple is leading the way with a chip that fuses two separate GPUs together.
Nvidia and AMD have built similar solutions for combining two GPUs in the past, but if AnandTech points out:, it seems that Apple has solved the holy grail of multi-GPU design here. Apple’s UltraFusion technology supports an impressive 2.5 TB/s bandwidth between the two M1 Max chips. That’s a huge jump in bandwidth from what Nvidia offers with NVLink for SLI or AMD with Infinity Fabric, which are used as fast connections between GPUs.
Apple’s high-speed link means that the two separate M1 Max GPUs in macOS appear as a single GPU, making it easy for apps to leverage the combined power. That should mean apps and games don’t have to do anything special to harness the power of the M1 Ultra, whereas in the past games had to natively support Nvidia’s SLI implementation on Windows to see performance improvements.
Nvidia has practically eliminated multi-GPU support with its RTX 30 series, with only the RTX 3090 offering NVLink support. However, combining two RTX 3090s for productivity or gaming rigs has wildly mixed results. Games that natively support SLI will provide performance benefits, while most will offer no improvement and some will even see a performance drop.
The RTX 3090 is currently the fastest GPU on the market — until Nvidia finally delivers its delayed RTX 3090 Ti — and Apple claims the M1 Ultra can beat a single RTX 3090 while using 200 watts less power.
Apple made similar claims about its M1 Max beating the RTX 3080 last year, but the real-world results were mixed. For productivity-oriented loads, the M1 Max performed extremely well compared to the RTX 3080. Some reviewers found the M1 Max a little slower than a comparable RTX 3080 system for Adobe Premiere Pro tasks, but relative performance really depended on the task at hand. used to be.
Apple has squeezed its M1 Ultra into its new Mac Studio, a desktop computer not much larger in volume than Nvidia’s RTX 3090. The Mac Studio is incredibly powerful thanks to the M1 Ultra, and Apple designed it to support the 27-inch iMac and even Mac Pro models for many.
Apple focuses entirely on productivity applications – not gaming – with its M1 Ultra and Mac Studio. Both the M1 Max and M1 Pro suffered from game reviews and performed comparable to an RTX 3060 in many titles. The M1 Ultra won’t magically solve the lack of macOS games, or that most cross-platform games are still x86.
We’ll have to wait for the reviews to see what double the GPU cores do to performance for productivity apps, but Apple’s already impressive start with the M1 Pro and M1 Max seems to be taking a leap forward with the M1 Ultra. It has taken Qualcomm and Microsoft years to deliver laptop performance on ARM-based chips for Windows, and Apple is already delivering workstation-level performance here.
Where it gets really interesting with Apple’s chip design is the Mac Pro. Apple’s existing Mac Pro is powered by Intel Xeon CPUs and AMD’s Radeon PRO W6000X GPUs. Apple ended its event yesterday by teasing an Apple Silicon version of its Mac Pro machine “for another day”. Bloomberg reporter Mark Gurman unveiled last year that the Mac Pro comes with up to 40 CPU cores and 128 GPU cores. If you keep count, that’s double the M1 Ultra.
The M1 Ultra in the Mac Studio gives us an early look at how well Apple’s core count and GPU performance can be scaled up to high-end systems. It’s a taste of what’s to come when Apple delivers the Mac Pro with the best of what its Apple Silicon can deliver.