Why can’t Apple be clear about the M1 Ultra’s performance?

Apple’s new M1 Ultra will be an impressive processor.

There’s no doubt of that since the M1 Ultra is, well, pretty much two impressive M1 Max chips joined at the hip. That makes it a 20-core (16 performance, 4 efficiency) processor, plus a 64-core GPU and a nutty 800GB of memory bandwidth thanks to its unified LPDDR5/6400 memory for all cores.

But even with all those impressive specs and the M1’s known pedigree, it’s yet again maddening to understand why Apple won’t ever be straight about how fast its new computers are.

For the M1 Ultra’s reveal, Apple trotted out another series of its clear-as-mud performance charts, like the one below. Sure, the main take away is the 20-core M1 Ultra outperforms a “16-core PC desktop” while using less power, but the lack of clarity on just how much faster and what it’s actually faster in is exasperating.

Apple

Apple does the same for graphics, saying the 64-core GPU in the M1 Ultra is slightly faster than the “highest-end discrete GPU” while using less than 200 watts less power. For the record, while Apple didn’t say in its presentation what the components are, the company did spell out in the footnotes that it’s taking shots at Intel’s 12th-gen Core i9-12900K and Nvidia’s GeForce RTX 3090.

Apple

What drives me even battier even is that despite these charts being about as generic as legally possible, there are some legit impressive-as-hell numbers Apple cited—but didn’t mention whatsoever in its presentation.

In Apple’s press release about the M1 Ultra, it says the new processor can transcode a 6K ProRes video 5.6x faster than the current Mac Pro’s 28-core Intel Xeon CPU paired with its fancy Afterburner card. Sure, the Mac Pro’s ancient Xeon is old and moldy now (whose fault is that?), but Apple’s Afterburner card features a custom-designed ASIC made just to accelerate ProRes media on Apple’s desktop. The fact the tiny Mac Studio can smoke it by 5.6x is indeed worth chest-beating over. But Apple didn’t during its M1 Ultra reveal, instead focusing on those smoke-filled and undefined charts that makes people doubt the claims more than believe them.

So what should we believe? Obviously, Apple’s performance claims about the ProRes transcode for one (by the way, ProRes is Apple technology). Early jaw-dropping leaks don’t surprise us either: The M1 Ultra will indeed be fast.

For example, in a seemingly on-schedule goof up, someone leaked a Geekbench 5 score of the M1 Ultra putting it at 24,055 for multi-core performance and 1,793 for single-core performance. Geekbench, if you don’t know, uses a couple-dozen performance “loops” modeled on modern workloads such as text compression, cryptography and JPEG decompression. It’s also famous for people not paying attention to the warning signs that results are uploaded to the Internet.

Geekbench

Is the M1 Ultra faster than Intel? Yes (and no).

Obviously, assuming we accept the leaked performance is close to the real M1 Ultra’s performance, we can see two things from the Geebench scores: The 20-core M1 Ultra looks to be about 29 percent faster than the 18,551 score we saw in our original 12th gen Core i9-12900K review here. Given we’re looking at 20-cores versus 16-cores that’s probably in line.

In single-threaded Geekbench 5 performance though it’s flipped. The M1 Ultra, remember, is basically two M1 Max chips with much better cooling. With the 12th gen Core i9-12900K scoring 2,001, that gives the Intel CPU an 11 percent win over the leaked M1 Ultra score.

If you’re wondering how the M1 Ultra compares to the M1 Max, we’ve seen MacWorld’s score for the 10-core M1 Max at 12,671 for Geekbench 5 multi-core which is about what you’d see if you snapped the 20-core M1 Ultra in half. In single-threaded, it’s also essentially within the margin error with a score 1,788.

Seeing the gigantic, honking cooling Apple has put in the Mac Studio for the M1 Max, we expect final scores to be faster but on the face of it, it looks like the M1 Ultra gives you basically, well, 2x the M1 Max.

If we were to play that out even further by simply doubling the performance of the M1 Max, in say, Maxon’s Cinebench R23, that would give the M1 Ultra a score of 24,750 (2 x12,375) in multi-threaded performance and likely the same performance at 1,529 in single-threaded performance. How does that compare to Intel’s Core i9-12900K? For the multi-threaded performance, we recorded a score of 27,275 which would mean the 16-core Core i9, is outperforming the 20-core M1 Ultra by 10 percent. And with the Core i9’s score of 1,984 the Intel CPU would easily sail by the M1 Ultra by almost 30 percent.

We still don’t totally believe that though as it’s hard to believe the giant copper-cooler Apple used in the Mac Studio wouldn’t pay more dividends (we hope). None of this pays the proper respects to the power claims Apple makes but that’s best left for actual performance results from independent reviewers.

Editor’s note: Updated to add video.