Build it and they will come. Nobody will take advantage of power that does not exist.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
M5 Chip Achieves Impressive Feat in 14-Inch MacBook Pro Speed Test
- Thread starter MacRumors
- Start date
- Sort by reaction score
Build it and they will come. Nobody will take advantage of power that does not exist.
Do it the other way around: 100 * (high/low - 1). 100 * (4263 / 3784 -1) = 12.66% increase
You may have missed this from March, 2025:
Microsoft no longer includes a power supply with Surface Pro 11 in some regions — now you have to pay extra for one
Microsoft is now complying with an EU initiative that allows the Surface Pro 11 to ship without a power supply, instead charging extra for it as an optional accessory.
www.windowscentral.com
That's because Apple killed their own market with Mac Pro. Shrinking the market is the strategy? Wow.
They clearly compared with RTX 3090 and that chart is totally false.
Since Apple cant even make a high-end GPU, how can you even compare with RTX 5090 and beyond such as workstation GPU? Mac Pro dont even support that with multiple GPUs.
Apple's GPUs sucks. Dont tell me It's better than Nvidia while Apple cant even make their own Mac Pro again.
It's like you didn't read a word I wrote.
How is the chart false? Break the chart down for us using the data Apple used to create it and demonstrate how it is incorrect. Look at the numbers Apple used and then demonstrate that the chart is in fact "totally false".
Also, you're assuming that because Apple hasn't made a "high-end GPU" that Apple cannot do so. People were skeptical about Apple Silicon before it was released and look how that turned out. It is proving to be revolutionary for computing.
"Apple cant even make their own Mac Pro again."
What does this mean? Apple literally is unable to update the Mac Pro or hasn't updated it in a while in part because it's not a priority?
Last edited:
M1 Ultra Doesn't Beat Out Nvidia's RTX 3090 GPU Despite Apple's Charts
Despite Apple's claims and charts, the new M1 Ultra chip is not able to outperform Nvidia's RTX 3090 in terms of raw GPU performance,...
www.macrumors.com
Such a false claim.
Actually, in Blender benchmarks the base M5 already pulls ahead of the 5700XT. On Blender OpenData, the M5 scores around 1700 points, while the 5700XT iMac comes in closer to 1000 points. So for rendering workloads (especially with Metal), the M5 isn’t just “catching up” - it’s already significantly faster and more efficient.
Really curious to see what the M5 Max can do once more results show up.
Apple M5 (GPU - 10 cores) - Blender Open Data Benchmark
Blender benchmark for Apple M5 (GPU - 10 cores)
opendata.blender.org
AMD Radeon RX 5700 XT - Blender Open Data Benchmark
Blender benchmark for AMD Radeon RX 5700 XT
opendata.blender.org
this is how these things go, they cherry pick one benchmark with M1 Ultra is roughly on par with 3090 and they make this claim. see: https://gfxbench.com/compare.jsp?be...&hwtype2=dGPU&hwname2=NVIDIA+GeForce+RTX+3090Then explain that chart directly from Apple. They clearly stated that RTX 3090's max performance as equal as M1 Ultra while the power consumption is 200W lower. Guess what? The performance difference is huge.M1 Ultra Doesn't Beat Out Nvidia's RTX 3090 GPU Despite Apple's Charts
Despite Apple's claims and charts, the new M1 Ultra chip is not able to outperform Nvidia's RTX 3090 in terms of raw GPU performance,...
Such a false claim.
M1 Ultra is actually competitive with 3090 in the gfxbench benchmarks, but it is not close in actual games. Is it misleading? yes. is it a false claim? technically no.
To make this a car analogy, think of the chip variants as classes of vehicle: The base chip are the minivans; Pros are the sports cars; Max are the supercars; Ultras are hypercars.
A 2025 Porsche 919 Hybrid Evo did the Nurburgring in 5:19.55
A 2020 Porsche Panamera Turbo S did it in 7:29.81.
The Ford SuperVan record is 6:48.393.
(Times are from Google non-AI results, but have not been vetted for accuracy!)
A van beat a 5 year old Porsche. Sure, not the fastest, but impressive for its class.
An M5 beat a 5 year old M1 Max. Sure, not the fastest, but impressive for its class.
Last edited:
Benchmark is benchmark, nothing more and doesn't really represent actual performance in real life and that's also why they choose specific benchmark to advertise that M1 Ultra is as good as RTX 3090 but consume way less power. I thought this issue had been discussed widely? This is not the first time that Apple used specific benchmarks.
I hadn’t seen this, incredible
That is one of the most beautiful things I have ever viewed.. there are times when it simply looks fake and sped up 2x, but it’s not.
Truly incredible. It’s like watching Michele Mouton hurl an Audi Sport Quattro through the forests of Finland
Sadly, when finding the video, it seems the record was from 2018, not 2025, kinda ruining my class comparison... Thanks Google. Perhaps we can consider this an M(-1) Ultra.
okay, so you said the chart was false. for this benchmark it is not false.
i don't think anybody is disagreeing that this was a misleading chart and claim, not sure what else to say here.
If you think Apple’s recent implementations of this are in any way the same as Intel and Nintendo I’ve got a bridge for sale you might like.
Good point. But you do immediately see that it's around 12.5% without any more typing. Granted, you'll need to be 'good at math' without a calculator.
If you read the post I was responding to, you would see that it's speculation on what Apple is working on, not products they have recently put to market.
Just doing high/low takes you 99% of the way there. 4263 / 3784 ~= 1.1266 I'm sure you can get from 1.1266 to 12.66 mentally without any effort at all.