When Apple unveiled the M1 Ultra, the company’s most powerful in-house processor to date and the crown jewel of its brand new Mac Studio, it did so with charts claiming that the Ultra was capable of outperforming Intel’s top processor or Nvidia’s RTX 3090 GPU on its own. In typical Apple fashion, the charts were labeled on the Y-axis with « relative performance, » and Apple doesn’t tell us what precise tests it does to arrive at whatever numbers it then uses to determine « relative performance. »
But now that we have a Mac Studio, we can say that, contrary to Apple’s claims, the M1 Ultra isn’t actually faster than an RTX 3090 in most tests.
According to Apple, the M1 Ultra is a silicon miracle, combining the hardware of two M1 Max processors into a single chipset that is nothing less than the « world’s most powerful CPU for a personal computer. » And based on Apple’s numbers, you might be tempted to believe those assertions.
On this chart, the M1 Ultra outperforms the RTX 3090 system in terms of « relative » GPU performance while drawing significantly less power. It’s a fantastic accomplishment!
But that’s because Apple’s chart has been cropped, for lack of a better term. The business only presents head-to-head comparisons for areas where the M1 Ultra and the RTX 3090 are comparable, and it’s true: in such cases, the M1 Ultra offers more bang for your money than the RTX 3090.
But the graphic doesn’t reveal that, while the M1 Ultra’s line pretty much ends there, the RTX 3090 has a lot more power to rely on – just look at some of the benchmarks from BlueHillco’s review:
As you can see, the M1 Ultra is a powerful piece of silicon that easily outperforms a nearly $14,000 Mac Pro or Apple’s most capable laptop. However, it appears like Apple is just not displaying the entire performance of the competitor it is chasing here – its chart for the 3090 finishes at around 320W, when Nvidia’s card has a TDP of 350W. (which can be pushed even higher by spikes in demand or additional user modifications).
It’s like claiming that while your electric car uses significantly less fuel when traveling at 80 mph than a Lamborghini, it has a better engine – despite the fact that the Lambo can still drive twice as fast.
And, yeah, it is quite impressive that Apple is able to accomplish so much with (relatively) little power. I’m confident Apple’s chart is correct in demonstrating that, in terms of relative power and performance, the M1 Ultra performs somewhat better than the RTX 3090 in that exact comparison. However, it effectively ignores the rest of the chart, where the 3090’s line leaps well past the M1 Ultra (albeit while using far more power, too).
The chart should generally look something like this:
The thing is, Apple didn’t need to do all of this chart tampering: the M1 Ultra is legitimately something to brag about, and the fact that Apple has seamlessly managed to merge two disparate chips into a single unit at this scale is an impressive feat, the fruits of which can be seen in almost every test that my colleague Monica Chin ran for her review.
In benchmarks and performance tests, Apple’s UltraFusion connection technology delivered nearly twice the performance of the M1 Max. Apple duct-taped two M1 Max chips together to gain double the performance of the M1 Max. No other chipmaker has ever achieved this level of success.
That’s fantastic, and far more impressive and interesting for Apple to have spent time showcasing than its best, most bleeding-edge chip beating out aged Intel processors from computers that have sat out the last several generations of chip design, or fudged charts that set the M1 Ultra up for failure in real-world testing.
It’s fine that Apple’s latest chip can’t outperform the world’s most powerful dedicated GPU! The 3090 is nearly the size of an entire Mac Studio and costs nearly a third of the price of Apple’s most powerful system. But I can’t help but hope that Apple would concentrate on honestly displaying to customers the M1 Ultra’s true strengths, benefits, and successes rather than creating charts that have us chasing after benchmarks that Apple must know deep down that it can’t equal. Not yet, at any rate.
Update March 17th, 2:25pm: Added RTX 3090 power specifications for better comparison.
Correction March 17th, 1:55pm: The Shadow of the Tomb Raider chart in this post originally featured a transposed legend for the 1080p and 4K benchmarks. We regret the error.