Apple M4 Performance Leaks

Apple M4 Performance Leaks

The Six Five team discusses Apple M4 Performance Leaks

If you are interested in watching the full episode you can check it out here.

Disclaimer: The Six Five Webcast is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors and we ask that you do not treat us as such.

Transcript:

Patrick Moorhead: So a couple of Russian YouTubers came out with what were supposedly benchmarks from an M4 powered MacBook Pro. Full packaging, all of the packaging looked like older packaging, so there’s some dispute of it, but the figures passed merit. So right now, there’s only CPU benchmarks out there. It does not look like there is anything related to the GPU or the NPU yet. I don’t know if they’re slowly trying to torture people on this to be able to tune in. I guess if you’re a YouTuber, it’s a smart thing to jack up your subs and views. But what it shows, nothing extraordinarily surprising does very well on the CPU with single-threaded scores higher than AMD, Qualcomm, and Intel. By the way, the M3 performed well on single-threaded Geekbench as well as Cinebench.

Where it gets, I guess, interesting here is multi-threaded benchmarks, Geekbench and Cinebench, and Apple got an extraordinary lift here where just to be brutally honest, they were somewhat of a… I mean, they were a complete lagger, right. You had Qualcomm and AMD pretty much running circles around the M3 and the score of 971 for Cinebench, AMD is still in the lead by a sizable margin. Go AMD. It’s very similar to the Qualcomm X Elite and what we know about the Core Ultra 288V, it shines mostly in single threaded, but it gets hit pretty substantially because it’s an eight core solution versus 12 and multi threading with AMD. So nothing surprising. I mean, I think for me, the MT scores were a lot better. And if I dial out what does this mean? I think what it means is that AMD, Qualcomm, Intel are still very competitive. AMD still has superior multi-threaded performance. Apple does put some distance with Intel Core 7 Ultra Lunar Lake, and less distance.

I mean, they’re almost the same on Qualcomm, but it’s what you would expect. Let’s just say that customers are paying a 200, 300, $400 premium for this chip and this design over everything else. It better perform well. But I think it also shows the durability of Qualcomm and AMD designs. And the funny part is Qualcomm is ARM and AMD is x86. But hey, we’ll have to see what the GPU, we’re going to have to see the NPU. But just the fact we’re talking about anybody else being in the zone with Apple, I think is a big freaking deal because Apple has been running circles on the laptop side around everybody on the CPU side. And I think it’s important too. Companies like Signal65, one of our sister companies gets in and does the hardcore testing to get underneath the reality.

Daniel Newman: Yeah, you hit the leaks in the data side. I just want to have a little fun with this, by the way, because the market perception is that Apple has this sort of ability to keep its launches and new products airtight, and over the years it’s gotten harder, and of course in a connected world. I pulled up some data points, Pat, on this. They’re saying that this was potentially the worst product leak since… You remember what happened with the iPhone 4?

Patrick Moorhead: Yeah, yeah. Legal got involved, investigators, police, all that stuff.

Daniel Newman: The prototype was apparently left in a bar in 2010-

Patrick Moorhead: Yeah, exactly.

Daniel Newman: … found that. It said that there’s a rumor also that 200 of these units may have been stolen from a Chinese warehouse. That’s one of the other rumors that’s out there right now and that it’s obviously they were selling them in the black market on Russian classified sites. I just thought that was super interesting how this happens. I mean, Apple has historically been incredibly good at containing this kind of stuff. Now the fact of everything you said is interesting, but I think didn’t it all kind of fall in line? Were there any surprises in these numbers to you?

Patrick Moorhead: I mean, the MT was a little bigger than I thought it would be. And by the way, the latest arm instructions are definitely playing a role in this, although Apple won’t give ARM credit, it should, but you’re seeing a boost just with ARM V92.

Daniel Newman: Yeah, yeah, exactly. No, I mean, the results are positive, and I think you hit it. We’re going to need to, one, see this actually get launched. Two, we’re going to need to put it in the lab and see how it compares. With all these advancements and these new AI PCs, Mac has sort of been on the outside. No one’s really even counting it in, and maybe it needs to be. So of course, when it comes to these ARM based, they were first to do it, and they were first to show. And you and I saw from M1 to M2, the generational improvements, Pat, you had some of the best documentation in the world on how crappy M1 was, but they did figure it out, and they did sort it out quickly. And by this next version, it looks like it’s going to be pretty darn compelling.

Author Information

Daniel is the CEO of The Futurum Group. Living his life at the intersection of people and technology, Daniel works with the world’s largest technology brands exploring Digital Transformation and how it is influencing the enterprise.

From the leading edge of AI to global technology policy, Daniel makes the connections between business, people and tech that are required for companies to benefit most from their technology investments. Daniel is a top 5 globally ranked industry analyst and his ideas are regularly cited or shared in television appearances by CNBC, Bloomberg, Wall Street Journal and hundreds of other sites around the world.

A 7x Best-Selling Author including his most recent book “Human/Machine.” Daniel is also a Forbes and MarketWatch (Dow Jones) contributor.

An MBA and Former Graduate Adjunct Faculty, Daniel is an Austin Texas transplant after 40 years in Chicago. His speaking takes him around the world each year as he shares his vision of the role technology will play in our future.

Related Insights
Will Glean's NVIDIA Nemotron 3 Ultra Integration Shift the Enterprise AI Stack?
June 6, 2026

Will Glean’s NVIDIA Nemotron 3 Ultra Integration Shift the Enterprise AI Stack?

Glean's integration of NVIDIA Nemotron 3 Ultra marks a pivotal moment in enterprise AI, where model flexibility and infrastructure alignment become strategic competitive advantages for buyers seeking cost-effective, high-context solutions....
Marvell’s Teralynx T100 Puts Power Efficiency at the Center of AI Networking
June 5, 2026

Marvell’s Teralynx T100 Puts Power Efficiency at the Center of AI Networking

Tom Hollingsworth, Networking Technology Advisor and Event Lead at Futurum, examines how the Marvell Teralynx T100 addresses AI networking power and latency constraints as hyperscalers build larger AI clusters....
Can NVIDIA Cosmos 3 Make Open Physical AI a Reality, Or Will Fragmentation Stall Progress?
June 5, 2026

Can NVIDIA Cosmos 3 Make Open Physical AI a Reality, Or Will Fragmentation Stall Progress?

NVIDIA Cosmos 3 launches as the first open omni-model for physical AI, targeting robotics and embodied AI with an open-source approach that challenges proprietary models from OpenAI, Google, and Amazon,...
HPE Q2 FY 2026: AI Orders Remain Strong as Supply Constraints Persist
June 4, 2026

HPE Q2 FY 2026: AI Orders Remain Strong as Supply Constraints Persist

Futurum Research analyzes HPE Q2 FY 2026 earnings, focusing on AI-driven demand across servers and networking, supply constraints affecting conversion, and what updated FY 2026 and FY 2027 guidance implies...
Intel’s COMPUTEX Keynote Reframes an Iconic Company as a Silicon-to-Systems AI Lab
June 4, 2026

Intel’s COMPUTEX Keynote Reframes an Iconic Company as a Silicon-to-Systems AI Lab

Brendan Burke, Research Director at Futurum, examines the Intel agentic AI pivot at COMPUTEX 2026, where Xeon 6+ on 18A, Rackscale Blueprints, and a Perplexity hybrid demo reframe the CPU...
Agentic AI
June 3, 2026

Salesforce Bets on Agentic Marketing: Will Unified AI Agents Redefine Martech ROI?

Keith Kirkpatrick, Vice President & Research Director, Enterprise Software & Di at Futurum, analyzes how Salesforce's agentic AI marketing platform leverages collaborative agents for campaign creation, lead qualification, and customer...

Book a Demo

Newsletter Sign-up Form

Get important insights straight to your inbox, receive first looks at eBooks, exclusive event invitations, custom content, and more. We promise not to spam you or sell your name to anyone. You can always unsubscribe at any time.

All fields are required






Thank you, we received your request, a member of our team will be in contact with you.