A top reviewer is failing to replicate some of Apple’s given performance numbers on its new M2 Ultra and M2 Max-powered devices. No prominent reviewers, at the time of writing, are suggesting that Apple actually faked any of its numbers. Instead, they are being criticized for using misleading, vague, and difficult-to-replicate figures.

Apple is well known for its unique marketing tactics and its refusal to compare its products to its competitors. The idea is thought to be that if Apple presents itself as a world of its own with no competition in sight, consumers will start to see it the same way. However, according to reviewers and testers, Apple’s marketers may have gotten more devious with tricks to mislead prospective consumers.

Are Apple’s M2 Ultra Performance Figures Misleading?

Reviewers and concerned consumers have long complained about Apple’s marketing tactics in the past, namely their performance numbers and direct comparisons with other devices. One of the company’s newest products, the M2 Ultra chip that can be configured in the Mac Pro and Mac Studio computers, is being scrutinized for its advertised performance results.

On the new Mac Studio page that boasts the various features and benefits of a Studio powered by an M2 Ultra chip, it offers extremely vague performance numbers.

M2 Ultra product description
Screenshot from Apple’s M2 Mac Studio product page

Instead of using any informative figures, it just says that the device boasts 3.3x faster CPU, 6.1x faster GPU, and 5.9x faster machine learning performance without telling you what it’s comparing to until you scroll down for a while to find its next box of performance features, displayed below.

mac studio performance graph
Screenshot from Apple’s M2 Mac Studio product page

Once you reach this screen you can finally figure out where those figures are coming from and dive deeper into Apple’s performance claims but there are immediately evident problems. Not only are the graphs completely unlabeled, making it impossible to compare with devices from the competition but it also doesn’t even compare the M2 Max and M2 Ultra figures to a regular M2-powered device.

It compares the Mac Studio M2 Ultra and M2 Max devices with a (nearly) 3-year-old iMac with an Intel CPU and AMD GPU. It’s hard to imagine that the marketers really thought that this was a fair or remotely useful comparison for a professional machine.

What Are Tech Reviewers Saying About the M2 Ultra and M2 Max Performance Figures?

One of the most popular online tech reviewers, Linus Sebastian from the gargantuan YouTube channel LinusTechTips, called the comparisons “comical” and “really misleading” in his review of the M2 Studio.

Sebastian went as far as to say that “[Apple’s marketers] are back to their old tricks with performance claims that would get any other company sued.”

The tricks don’t seem to have stopped with the misleading graphs and comparison points. Sebastian and his company Linus Media Group is one of the leading sources of real, reliable testing information on consumer electronics. He owns a large testing facility designed to test consumer tech called LABS, though it is still in its early days, and his team heavily benchmarks and tests the performance of the products they review.

When his team tried to replicate the figures listed on the M2 Ultra and M2 Max Mac Studio promotional page, they simply failed to reach the same results. Judging by the (unlabeled) graph shown above, the M2 Ultra should be just over 3 times faster than the M1 Ultra in OptaneX Render performance.

The tests they ran told a much different story. While the M2 Ultra-powered Mac Studio certainly posted better results, they weren’t even close to 3 times better. At 4k (3840×2160), the M2 Ultra posted a 125-second render (faster is better) compared to the M1 Ultra’s 158 seconds, a measly 26.4% increase compared to the claimed 3x performance.

octanex mac studio performance graph
Image courtesy of Linus Media Group

Sebastian’s review also cites a post in the OTOY forum that independently tested the same OptaneX performance of the 2 chips. Despite the multiple failures to replicate Apple’s marketing figures, the Linus Media Group team gave Apple the benefit of the doubt and tried to find specific scenarios that would give these results.

It tested a much longer, higher-resolution render to give the chips more time to reach maximum performance. Unfortunately for Apple, this test ended with roughly the same results as seen below.

M2 ultra mac studio optanex graph
Image courtesy of Linus Media Group

The Numbers Aren’t Matching Up – What’s the Problem?

The most likely reason why reviewers like Sebastian and independent testers like the OTOY forum user can’t replicate the results is that they are using a different render. It’s the only remaining obvious variable that could be changed.

Sebastian and the cited tester used the same application, OTOY Octane X Prime 2022.1.1, but Apple never specified what exactly it was rendering in the tests. It merely says that it tested “a scene with complex materials and 35 million triangles.” How Apple managed to build a render that so favors the M2 Ultra chip is still an unsolved (and pressing) question but it is a likely culprit behind the strange numbers.

One theory is that the render was so large that it exceeded the amount of random-access memory (RAM) in the M1 Ultra Mac Studio but not its M2 Ultra counterpart.

Your computer’s RAM is like its workspace. That’s where all the data it is actively using and computing with is stored. When a task exceeds the amount of physical RAM you have, the extra has to be pushed off into other, much slower storage. This is (mostly) why your computer gets so slow and buggy when you open too many Google Chrome tabs.

So if this render was so large that the M1 Ultra Mac Studio had to push some of it to its much slower regular memory and the M2 Ultra Mac Studio didn’t, it could easily explain Apple’s OptaneX performance claims.

If this is the case, which has not been proven or thoughtfully tested at the time of writing, the numbers Apple uses to sell these devices are extraordinarily misleading. If the only time these devices reach the performance gains it advertises is when rendering an absurdly large project, it will be far from representative of the device’s performance in all other use cases.

Despite these (allegedly) shady numbers, it’s important to remember that the M2 Ultra chip itself is an incredible feat of engineering and the Mac Studio that comes with it is still a great device for all sorts of professionals and tech enthusiasts.

What's the Best Crypto to Buy Now?

  • B2C Listed the Top Rated Cryptocurrencies for 2023
  • Get Early Access to Presales & Private Sales
  • KYC Verified & Audited, Public Teams
  • Most Voted for Tokens on CoinSniper
  • Upcoming Listings on Exchanges, NFT Drops