The 2025 iteration of the Mac workplace marks a new era of successful individual computing for Apple, pushing the boundaries of what can be achieved using what is basically a bigger variation of Intel's now-defunct next unit of computing (NUC).
I pitted its full-fat, top-end version, the M3 Ultra with a 16TB SSD and 512GB of RAM, against the competition to get a smooth idea of how good a bargain it is. Unsurprisingly, it easily competes with any of the world's most successful workstation PCs.
There's a trio of ThreadRipper rigs, seven Xeon workstations (across 3 generations), and, for good measure, NVIDIA's magnificent Digits and 3 of Apple's own PCs, the Mac Mini (M4 Pro), the Mac Pro (M2 Ultra), and a smaller capable version of the Mac Studio (M4 Max).
The full grid is at the end of this article. Note that all cells are blank because one simply doesn't have the information at hand (e.g., size of any workstations, performance of Digits, etc.).
Credit where credit is due, Apple has managed to present something I did not expect it to do anytime soon: 512GB of unified storage in one computer.
This stroke of genius meant moving the examination to Intel Xeon and AMD ThreadRipper Pro systems, far more expensive skilled platforms compared to AMD Ryzen and Intel center ones.
As well as zeroing in on the workstation vertical, it also firmly signals Apple's ambitions to be a major player in the red-hot AI hardware market.
(Worth checking out my nonfiction published in June 2023 where I suggested that the Mac workplace would rejuvenate the Mac Pro and make brand microcomputer vendors nervous.)
I analyzed the prices of these systems (using the most up-to-date data one could find), and there’s no two ways about it.
A $14,000 Bargain if You’re After Memory
The Apple Mac workplace Ultra, even at more than $14,000, is a bargain if you’re looking for as much framework representation as possible, at $28 per GB.
That’s a bit less than Puget Systems and about a third of what you’d pay for a maxed-out Mac Mini (or half the Apple Mac Pro with its limited 192GB RAM).
Most importantly, it is far cheaper than the likes of Dell or HP; it turns out Windows workstations with half a terabyte of RAM don’t come cheap.
Remember that Apple uses unified memory, which has a much higher bandwidth (800GB/s) than DDR5 memory (even across 8 memory modules).
It delivers the kind of speed you’d expect from an integrated GPU; as Servethehome’s Patrick Kennedy puts it, “The cost is advanced in the discourse of a workstation, but it is inexpensive once you compare it to NVIDIA’s cost per GB of memory.”
Remember as well that one can configure this workstation with 16TB of onboard storage; no other major player has access to that kind of retention unless you resort to RAID-0 across multiple smaller SSDs.
There’s also the fact that it comes with 10GbE LAN by default as well as Thunderbolt 5, a data transfer technology not available on any other workstation at the time of writing, which, in due time, will pave the way for innovative external peripherals.
I also looked at three other composite metrics, all examining processor performance (based on the popular Geekbench 6) and, not surprisingly, Apple excelled in all three: GeekBench component per dollar, Geekbench component per core, and Geekbench component per volume.
The M4 devices (Mac Mini and Mac Studio) had the cheapest speed per dollar, at about $0.20, which is about 1/8th of the most expensive one (the HP G4).
It also had the highest speed per core (1623), 5X faster than the slowest processor, an Intel Xeon, and the highest speed per part volume (more than 140x compared to the worst entry).
GPU details remain a mystery
The large chartless remains the authoritative GPU metrics arsenic it is hard to gauge Apple’s graphics speed crossed antithetic systems-on-chip arsenic Apple doesn’t divulge further information beyond center numbers.
For the interest of simplicity, I chose the Nvidia RTX 2000 Ada procreation as the default card - where possible - as it was the cheapest skilled graphics paper that came with 16GB of RAM. Only 1 entry utilized the much expensive RTX 4000 Ada procreation with 20GB of RAM.
Tech publication, TechPowerUp, estimates that the RTX 2000 AG should beryllium somewhat slower than an RTX 3060.
The pome M3 Ultra scores conscionable nether 130,000 marks connected the fashionable OpenCL Geekbench tests, that’s 50% faster than the RTX 2000 AG, so for a much balanced comparison, readers whitethorn privation to substitute a much powerful (and expensive) card similar the NVIDIA RTX 5000 Ada Generation.
* Geekbench 6 entries are sadly not averaged. So one picked what one considered to beryllium the median connected the archetypal section of all CPU. Your mileage whitethorn vary.
.jpg)
