M1 Ultra VS RTX 3090: New Apple chip is being smoked by Nvidia

Latest update:

When Apple unveiled the new M1 Ultra chip at its Peek Performance event, it unveiled Mac Studio, and with the hardware itself judged, critics put to the test Apple’s claims that it will beat top discrete GPUs. The Apple M1 Ultra integrates two M1 Max chips together, a bit like an SLI bridge to combine their performance and enable them to reach workload performance never before seen on an ARM chip. Those claims hold true for many of the M1 Ultra’s, and there’s already something to be said about the prevalence of silicon ARM among Apple’s chips, but it will be a good while before you can catch up on the massive horsepower under the hood that Nvidia’s most powerful GPU ( Up to 3090Ti versions of course.)

What claims has Apple made about the performance of the M1 Ultra?

Apple’s graphs in their product ads are always pretty vague, and you just need to look at the Y-axis on this graph to understand how ridiculous Apple’s claims can turn out. “Relative” performance is like an odd metric to judge your product by, and while the M1 Ultra may actually be able to match the RTX 3090 in terms of GPU performance in certain workloads, reviews come out that completely disprove that claim. The Verge reported that the RTX 3090 outperformed the M1 Ultra by more than 100% in the Geekbench 5 computing test. The M1 Ultra scored 83,121 in their test, while the RTX 3090 managed 215,034.

So where is that relative performance that Apple claimed the M1 Ultra claimed to match? It appears to be largely nonexistent. Claims like these aren’t new to Apple, and while ARM’s processor acceleration across its entire product portfolio has been impressive, and its performance-per-watt gains have also been equally respectable, it looks like it’s going to take a bit more time before Apple can make it. Catch up on pure performance to match the biggest and best graphics cards in the industry.

The market is becoming saturated at the higher end, with Nvidia, AMD and now Intel entering the fray. Despite this, Apple has yet to match in terms of pure GPU performance more than any of these brands, if this test is correct.

Apple M1 Ultra VS RTX 3090

The winner of the Apple M1 Ultra VS RTX 3090 battle is clearly the 3090, it’s not a close fight. Apple’s colossal farce even claiming that the M1 Ultra reaches a performance level close to the RTX 3090 is incorrect, according to reports via critics. We’re not even sure the M1 Ultra will stack up against the RTX 3060, forget the 3090. Geekbench once again reports that the RTX 2060 Super beats the M1 Ultra, so here’s a fun little graph to impress, thanks to Geekbench.

GPU Geekbench points
RTX 3090 215.034
RTX 3080 181,138
RTX 3070 135.061
RTX 3060 93334
RTX 2060 Super 85163
Apple M1 Ultra 83121
Source: Geekbench

As you can see, Apple’s claims were far from being better than a “top-end” discrete GPU that was bettered by a roughly three-year-old mid-range graphics card. When looking at the card’s relative performance against Nvidia’s current lineup of Ampere graphics cards, it doesn’t become a conversation. This is why the “relative performance” of Apple Charts can be so misleading.

Apple may have decided that relative performance means performance per watt, but it’s still a misleading moniker to drive marketing materials. This is an actively anti-consumer thing, but Apple is no stranger to doing that whether it makes their hardware look good. We’re not saying the Apple M1 Ultra chip is bad, it’s far from it, and offers huge advantages for those looking for a lot of power in a small package, but instead, the reason this misleading marketing is so abrasive is that the consumer can analyze these Information as applied to different workloads, the “relative” part of their graph means they can easily get rid of that graph, simply by not specifying why the M1 Ultra is so good. But at this point, the entire industry knows that ARM chips are more efficient when it comes to power usage, so the entire graph could be interpreted as invalid. This terrible use of data is one of the things that is just designed to make Apple look good when it unveils a new product, and doesn’t give consumers any information they want to know about the actual power under the hood.

Other tech companies are no stranger to this either, but that’s a rather scandalous claim that Apple made in advertising the same product, so calm your Cupertino planes, you’re not that strong yet.

3D artists and players must stick to dedicated GPUs

How to Overclock an AMD or Nvidia GPU Easily

Apple’s claims and sentiments in the creative industry may sound appealing to most people, but the truth as evidenced by the data is that the M1 Ultra and the RTX 3090 are just in very different leagues, for those into 3D art or hard gaming, then still Your bets are firmly placed in ensuring that you get a good dedicated GPU for the best performance and, most importantly, to save time while working.

Sure, Geekbench is standard, and Apple has made improvements elsewhere, but when it comes to the raw power of the chip, more traditional solutions are often best. It will take a long time before Apple can catch up to the pure power of other chips, so instead of looking at Mac Studio as a desktop alternative, think of it instead as a viable complement to your existing powerful backbone machine. Thinking of it as anything more than that will end up wasting time and money.

Share this article…