News Apple M4 benchmarks suggest it is the new single-core performance champ, beating Intel's Core i9-14900KS — incredible results of 3,800+ posted

[helper800]

Are they all complaining like you??

So refuting your appeal to the majority is a complaint, got it. I was pointing out the illusion that apple users are in some meaningful majority or even minority of users. The implication of your claim is that millions of apple users cannot be wrong compared to the complaints of the few in this forum because there are so many more of them not complaining here.

 

[Alvar "Miles" Udell]

It's not unexpected, to borrow a chart or two from the TomsGuide review of the M3 Macbook Air. Apple's M chip means it is faster in some things than Intel and slower in others, but it's -far- more power efficient.

	MacBook Air 13-inch M3	MacBook Air 15-inch M2	Dell XPS 14 (2024)
Geekbench 6 (single-core)	3,082	2,613	2,398
Geekbench 6 (multi-core)	12,087	9,993	12,939
Handbrake	6:32	7:46	6:20
Photoshop	8,550 / 12.4 mins	7,465 / 13.9 mins	6,045 / 21.6 mins
Premiere Pro	3,603 / 7.8 mins	3,170 / 8.7 mins	4,918 / 7.3 mins

Battery life results	Time (min:secs)
MacBook Air 13-inch M3	15:13
Acer Swift Go 14	9:50
Dell XPS 14	6:26

 

[gruffi]

Am I reading this correctly? A tablet handily beats Intels flagship desktop CPU in single threaded?

"But muh multicore" THIS IS A FREAKING TABLET VS A DESKTOP CPU THAT CONSUMES OVER 300 WATTS

It doesn't consume 300W single threaded. And it easily beats the M4 multi-threaded. But that comparison is pointless. First of all, the 14900K was made for high performance. And as the last weeks showed, Intel pushed the limits way too much since 13th gen. Apples's M-series is optimized for efficiency. Second, Intel isn't the benchmark anymore for quite a long time. Speaking of x86, AMD has the better and more efficient solutions for some years now. And third, the M4 will compete with AMD's Zen 5 and Intel's Arrow Lake.

 

[JTWrenn]

Reviewers and tech bloggers need to stop this nonsense. Until we see real world scores we know nothing about the performance. These leaked early benchmarks are always dialed in and almost always best case scenarios. Geekbench especially. Stop it. Just stop.

 

[gruffi]

Highest result for an intel cpu under windows is a single score of 4580 with an i9-14900KS, now I'm no mathematician but I think that is higher than 3800 points.

It's either an invalid result or no stock setting. The i9 usually scores something like 3.3k 1T.

 

[hushnecampus]

How come? Seems like you're severely hindering yourself by opting to use an Apple computer.

lol, OK

 

[razvanics]

Don't get me wrong , but the whole article is kinda click bait. Why would i care if in a benchmark M4 scores better than a 14900K or my 7950X single core ... It's like comparing a Kart to a Truck , they do diffrent things . What will people do on that tablet ? play mario , browser will open 0.05 sec faster ? , youtube , word ? wow , the power of M4 unleashed ... ARM sucks , it has been more than 3 years since M1 and Rosetta still sucks, ARM support is 0 and the numbers mean nothing if you can't do 99% of the stuff Intel or AMD does on X86 CPU's.

 

[Notton]

Let's keep in mind, Geekbench doesn't run longer tests where thermal throttling would kick in.
This is highly advantageous to smartphones and tablets with a very limited thermal envelope.

It's inherent bias is not as bad as AnTuTu, but the results from Geekbench have always been questionable every time it comes up.

 

[TerryLaze]

Let's keep in mind, Geekbench doesn't run longer tests where thermal throttling would kick in.
This is highly advantageous to smartphones and tablets with a very limited thermal envelope.

Eh, that also helps desktop CPUs that can run much higher power for short durations than they can for longer durations.

 

[Notton]

Eh, that also helps desktop CPUs that can run much higher power for short durations than they can for longer durations.

Yes, and yet Geekbench is widely considered to be useless when comparing across different architectures and platforms.

 

[bp_968]

These comparisons pop up with every new M-chip release but always feel pointless. Who are the comparisons for? Who is going to switch from a intel/amd based solution to a Mac that isn't already there?

Unless apple suddenly decided to make their CPU/GPU setup compatible with Linux/MS *and* was willing to dive big time into server development or make the chip available to systems builders it's just not relevant (and LOL, we will have fusion power and FTL before THAT happens!)

I also honestly believe that if Apple didn't control the entire stack top to bottom that it's scores wouldn't look nearly as impressive on their arm solutions. Being able to optimize everything for yourself isn't exactly a small advantage.

 

[bp_968]

Ho so? I've always hated Apple. I hate their hypocritical preaching on "sustainability". I hate their smugness. I hate the walled garden, but the only thing that can't be done on a macOS device that can be done better in Windows is gaming. (Also, Microsoft has turned just as preachy & hypocritical. Windows literally tells me to turn down my screen brightness to save on "carbon emissions".)

What else would you use a computer for then gaming? Lol. I know for sure I can't find many people who are willing to pay 1k+ for a computer other then for gaming at least.

That said, we are very close to not needing Microsoft for gaming anymore. If Valve would hurry up and release a desktop version of steamOS I'd seriously consider switching my desktop at this point. I ran my wifes gaming PC on Linux for a year a while back and she hardly noticed a difference.

I wish valve would really push the OS to other handheld makers and get a desktop version rolling. Steam machines were underbaked the first time around but with proton that issue should mostly be resolved now.

 

[M0rtis]

According to me these comparisons are pointless. No doubt the M series is more efficient when it comes to power but I would like to see benchmarks of processors on a universal OS using a universal Benchmark that is not optimized for any particular processor. Lets see the difference in that scenario.

 

[vijosef]

At equal price, Intel still destroys anything Apple has to offer, by a huge margin.

 

[Sam Bridges]

While I don’t put much stock into Geekbench scores I’m here for people just dismissing this because it’s an Apple product.

 

[ahmedfarazch]

Hi!

The article clearly mentions that some aspects of the benchmark are now accelerated by the extensions added to the M4. Such extensions are also expected to be added to X86 architecture as well on similar nodes. With new process nodes (that are arbitrarily named ... 3nm means nothing) performance should be similar for similar power budget. The difference should come from accelerators/dedicated silicon and how the software uses such additions. AV1 decode and encode is a good example.

 

[skiwi44]

Everybody gawks at the performance.

Not one of you wants to be trapped in that walled garden.

Says the guy who has been stuck in the Intel/AMD walled garden for over a decade….

;-)

 

[skiwi44]

They missed two huge advantages of the Intel 14900 processor that always gets overlooked - as Geekbench doesn’t really mention it.

1) The ability to heat a room, and
2) Helping those with tinnitus by providing a stronger background noise…

 

[ezst036]

Says the guy who has been stuck in the Intel/AMD walled garden for over a decade….

;-)

That is just a processor, that is not a whole ecosystem.

 

[TheHerald]

Am I reading this correctly? A tablet handily beats Intels flagship desktop CPU in single threaded?

"But muh multicore" THIS IS A FREAKING TABLET VS A DESKTOP CPU THAT CONSUMES OVER 300 WATTS

Ιt does not consume OVER 300 WATTS at single thread workloads. Heck, it doesn't even consume that much in the MT portion of GB6. My 12900k does GB6 at around 70 watts. It's not that MT intensive.

 

[TheHerald]

It only has 4 performance cores and 6 e cores. The fact that its performance is 69% of a processor that has 4 more performance cores and 12 more e cores is pretty impressive. If the M4 architecture can scale, an M4 Pro or Max could beat a 14900K in both single and multi thread benchmarks while using way less power to do so.

Completely and utterly wrong. Number of cores is irrelevant. The M4 is a bigger chip than their 14900k. It's made of 28 billion transistors vs 25 billion of the 14900k. It should be beating the crap out of the 14900k in both ST and MT performance (since each core is much bigger in transistor size) but it doesn't.

 

[TerryLaze]

They missed two huge advantages of the Intel 14900 processor that always gets overlooked - as Geekbench doesn’t really mention it.

1) The ability to heat a room, and
2) Helping those with tinnitus by providing a stronger background noise…

Man, you didn't even make the effort to type the k ,the non-k 14900 is a 65W base power CPU, that will suck at warming your room and can be cooled with passive cooling.
A little more effort in your trolling would be much appreciated.

 

[YouFilthyHippo]

Ok, someones gonna hafta give me an ELI5 on this one. A tablet CPU beating a 14900K? There has to be something I'm missing. Is this one unique cherry-picked benchmark, specifically tailored in favour of the M4, or is this some massive revolution on apples part, and where does the 7950X3D fit in? Can someone tell me whats going on here? I am very confused

 

[skiwi44]

Am I supposed to be impressed by Apple beating 10nm+++++++++++?!

Yes you should. A battery-powered tablet beating a 400w desktop processor.

And 10nm?? That says a LOT more about Intel and its inbility to provide any real leadership. Whcih has been the case for, let's see, about the last TEN years.

 

[skiwi44]

I don't think most of Apple's users realize it's a walled garden, most pick Apple because it's convenient.

No, most choose it because it does the job. I can get an all-day battery life with no decrease in performance, even when under a heavy imaging workload on site (and away from power points). I'll take a walled garden anyway, if it can deliver the results with none of the trade-offs I was used to in the Wintel world.

And "walled garden"? If you've done any physiology, you will understand that if you want a garden, it has to be walled. Otherwise your garden will not last long. I will leave the rest of the analogy to the reader....