Geekbench score for the Intel i9-13900K is out and it’s not looking pretty for Apple …

[Joe Dohn]

If Apple wanted to clock their CPUs up to those kinds of clocks (and use those kinds of power budgets), they could blow both AMD and Intel out of the water, so I wouldn't worry too much yet. Apple is playing softball at this point, they still haven't taken all their tricks out of the bag.

Are they, really?
Because it would be logical to go all-out if I had a monster chip that would crush my competitors.
M1 was really impressive, but M2 is showing that things don't scale all that well without thermal throttling. We'll see how things go with their 2022 refreshes, but I am not so sure.

 

[rafark]

Now that Intel will have a chunky performance lead over the M2 later this year, it would only be hypocritical to dismiss benchmarks when they don't favor Apple chips.

I’m no windows user (thankfully!) but this is just so true.

 

[Jumpthesnark]

Oh snap, aren't you a rebel! Good on you man, you sure showed the entire PC industry!

Gosh, I'm really sorry if my personal feelings on the matter offended you this much. /s

 

[sre.live]

When the M1 came out, I was the first one to quote benchmarks and say how ahead Apple was against AMD and Intel's 11th gen chips. For laptops, it was the biggest leap of the last 10 years. Now that Intel will have a chunky performance lead over the M2 later this year, it would only be hypocritical to dismiss benchmarks when they don't favor Apple chips.

Do you use your computer as a benchmark generator or as a tool to do stuff?

One can look at geekbench all day long or you can experience what M1 Pro or higher deliver in real world high load workflow scenarios like video editing, animation, photography.

Theory vs practical application.

 

[Gerdi]

Are they, really?
Because it would be logical to go all-out if I had a monster chip that would crush my competitors.
M1 was really impressive, but M2 is showing that things don't scale all that well without thermal throttling. We'll see how things go with their 2022 refreshes, but I am not so sure.

Because it makes no sense for Apple unless they have a product, where they can put such an chip into - the i9-13900k has a PL2 of 250+W - it can go only into Desktops with very high cooling capabilities. To put things into perspective, the single Intel core achieving barely higher score than the M2 is consuming more power than the whole M2 running on all cores.
Strange, that you are asking this question, while mentioning thermal issues in the next sentence.

 

[uller6]

i9 is the desktop CPU. How does the Mac Studio compare? I think many people on desktop would be ok with higher power draw to get more performance.

Many people are also satisfied with existing performance at very low noise levels and do not want a jet engine sitting on their desk.

 

[uller6]

This is true, but the M-series is not as impressive as it was when it was first released in November 2020.

Apple may have the edge in performance per watt. But, when looking at high-end computers, performance per dollar is even more important.

Apple charges a premium for its computers. It is fine when Apple sells an M1 MacBook Air for $999, which is faster and lighter than any Windows laptop at the same price point.

Now let us get a high-end desktop. Look at these two configurations:

Mac Studio: Apple M1 Ultra 48-core GPU, 64 GB RAM, 1 TB SSD, USD 3,999.00
PC: Intel Core i9-12900K, GeForce RTX 3090 Ti, 64 GB RAM DDR4-3600, 1 TB NVME SSD, USD 3,700

Mac Studio: Apple M1 Ultra 64-core GPU, 128 GB RAM, 8 TB SSD, USD 7,999.00
PC: Intel Core i9-12900K, GeForce RTX 3090 Ti, 128 GB RAM DDR4-4000, 12 TB (3x 4 TB) NVME SSD, USD 6,000

The PC has a far better video card, and yet is cheaper than the Mac Studio in both scenarios. In the second one, the PC is 25% cheaper, even having a better video card and 50% more storage.

If I were a high-end user that needs a desktop with lots of power, I would probably care more about performance per dollar than performance per watt. Perhaps the forthcoming Mac Pro addresses this issue, but we should check the price tag.

Yeah dude! I agree! Performance per dollar is where its at! My Toyota Camry is about the most performance/dollar for any car ever made. What a beast. But, it's certainly not a luxury vehicle by any stretch of the imagination. Nor is it especially performant.

 

[SpotOnT]

I remember when Apple was using water cooled PowerPC G5 processors and looking at Intel and seeing the power they were able to achieve and how much cooler and more efficient they ran. It was so obvious that Apple was just lagging behind the competition.

That is not the situation here though. Apple Silicon and Intel are around the same ballpark in terms of performance. Differe release schedules means one will always be 10-20% faster than the other at any given time. Plus Apple Silicon is still running way cooler, which means you can put the thing in a laptop rather than some giant tower. As far as I can tell, Apple Silicon still has the edge.

I think you are making a mountain out of a molehill here.

 

[uller6]

Well, performance can also rely on software, but hardware plays a very important part in it. I have watched some other videos, and Adobe software, such as Premiere, tends to be faster on Macs indeed. However, other software (most of them) will run faster on a PC with better specs. Overall, the performance will loosely follow the benchmarks, although there may be some variations.

The bottom line is that the custom-built PC is faster than the Mac. In terms of performance-per-watt, the M-series processor in the Mac is faster, as it is more efficient. However, in terms of performance-per-dollar, the PC is better, as PC manufacturers charge less for their faster (although less efficient) components than Apple.

There are, however, specific cases in which the Mac may be faster. But they are not the rule.

You're forgetting to add the cost of electricity over the lifetime of the machine to the cost. PCs definitely do not come out ahead in performance per dollar if you include the cost of power for 3+ years...

 

[Gerdi]

Differe release schedules means one will always be 10-20% faster than the other at any given time.

Excuse my ignorance but are you really comparing a ultra-mobile CPU against a desktop class CPU? The pendant to the M1 is the i7 1255U, with much lower scores, mind you? And whatever Intel's Alderlake successor is, perhaps the i7 1355U - it will still be slower.
I mean, why are we even comparing with a CPU, which, from power perspective, could not possibly go into any Apple products. Or where there Apple products with Intel K processors in the past?

 

[Homy]

Source

The i9-13900K chip will be out later this year and we now have Geekbench results. Single core: 2133 and Multi core: 23701

In comparison, the M2 in the new MacBook Pro scored: 1919 in single core
8929 in multi core.

Sure, Apple is much better at performance per watt than Intel but it’s not a good look to fall behind in single core performance. Most day to day tasks are single core.

Apple upended the chip industry with the M1 but AMD and Intel came back swinging and it seems like Apple now needs to pull another rabbit out of the hat with the M3.

What comparison by the way? You're comparing a low-powered 8-core mobile CPU to a high-powered 24-core desktop CPU? In comparison 20-core M1 Ultra already scores 24147 in multi-core, 1.9% higher than the unreleased i9-13900K. Since M2 is about 13.5% faster than M1 M2 Ultra should do about 27400 in multi-core.

Oh, I see you're a newbie, never mind.

 

[Cognizant.]

You’re comparing the high end Intel chip that hasn’t been released yet to the lowest end chip in the Pro that isn’t even going to be remotely as powerful as what’s going into the next Pro, Max, and Ultra chips. In fact the performance in the M2 MBP pales in comparison to what’s in the current M1 Pro and Max chips. What’s that point of these threads other than displaying your utter lack of ability to compare similar hardware? Also we haven’t even seen what the Mac Pro will do.

 

[Unregistered 4U]

Apple upended the chip industry with the M1 but AMD and Intel came back swinging and it seems like Apple now needs to pull another rabbit out of the hat with the M3.

Should be great for macOS users! This will be included in a Mac, right? It won’t? Ah, well nevermind. Doesn’t matter how fast it is, if it doesn’t run macOS, the majority of folks using macOS have no use for it.

 

[Joniz]

I can’t even comprehend the kind of person who would see that and go, “Gawd, I’m going to ditch fully functional my computer with an entirely different operating system and different apps and get that!”

 

[Cognizant.]

I can’t even comprehend the kind of person who would see that and go, “Gawd, I’m going to ditch fully functional my computer with an entirely different operating system and different apps and get that!”

Indeed. I'm going to eviscerate my entire workflow and productivity cuz of this.

 

[grkm3]

Thise is just an engineering sample and not final stepping. It's ddr5 was also clocked way slower 4400 vs 5600 that the final version will run at. Most likely clocked slower to take away any instability from memory used. Final silicon will break 2300 single core

 

[StoneJack]

So conclusion: unreleased intel top i9 13 gen is already beaten by last year’s m1 😂

 

[Unregistered 4U]

Because it would be logical to go all-out if I had a monster chip that would crush my competitors.

They don’t need to crush their competitors. They just need to be more performant than the last released Mac. Any who wants to do non-Mac stuff should feel free to do so. Apple is focused on that small number of folks that want what they’re selling and not the even tinier number of folks that want a benchmark number to crow about.

In a very real sense, the high end is growing more and more irrelevant as the wider computer buying public are happy with all the mobile products they’re buying. Produce a performant mobile solution and you’ve got something selling to the masses of folks buying those. Produce a performant desktop solution and you’re looking at something that’ll NEVER sell as well.

 

[Unregistered 4U]

Thise is just an engineering sample and not final stepping. It's ddr5 was also clocked way slower 4400 vs 5600 that the final version will run at. Most likely clocked slower to take away any instability from memory used. Final silicon will break 2300 single core

Very likely, this result comes from a processor they flew to Antarctica in order to have enough frigid temperatures and very cold water to keep passing over it. That glacier that cleaved itself from the main ice shelf? That was Intel trying to get this thing to just run so they could post a number in the few minutes it took to melt a section of the ice shelf.

 

[boss.king]

Source

The i9-13900K chip will be out later this year and we now have Geekbench results. Single core: 2133 and Multi core: 23701

In comparison, the M2 in the new MacBook Pro scored: 1919 in single core
8929 in multi core.

Sure, Apple is much better at performance per watt than Intel but it’s not a good look to fall behind in single core performance. Most day to day tasks are single core.

Apple upended the chip industry with the M1 but AMD and Intel came back swinging and it seems like Apple now needs to pull another rabbit out of the hat with the M3.

Where do I get the portable nuclear fusion cells needed to power that chip in a laptop?

 

[Unregistered 4U]

I can’t even comprehend the kind of person who would see that and go, “Gawd, I’m going to ditch fully functional my computer with an entirely different operating system and different apps and get that!”

There are people that post because they want to engage in a conversation. Learn something, help someone, etc. Then, there are those that merely post with the goal of “How can I write this to ensure I get at least 6 pages of comments?” I’m not saying that anyone in THIS thread is doing that, but I do know it happens.

 

[lcubed]

i9 is the desktop CPU. How does the Mac Studio compare? I think many people on desktop would be ok with higher power draw to get more performance.

there seem to be quite few people (at last count, 37) in the mac studio sub-forum complaining that current 'whisper quiet' M1 studio is too loud using the current M1 Ultra power envelope

 

[rgwebb]

Source

The i9-13900K chip will be out later this year and we now have Geekbench results. Single core: 2133 and Multi core: 23701

In comparison, the M2 in the new MacBook Pro scored: 1919 in single core
8929 in multi core.

Sure, Apple is much better at performance per watt than Intel but it’s not a good look to fall behind in single core performance. Most day to day tasks are single core.

Apple upended the chip industry with the M1 but AMD and Intel came back swinging and it seems like Apple now needs to pull another rabbit out of the hat with the M3.

An i9 out performs something that would slot like a i3 in their lineup? No way!

 

[jav6454]

Source

The i9-13900K chip will be out later this year and we now have Geekbench results. Single core: 2133 and Multi core: 23701

In comparison, the M2 in the new MacBook Pro scored: 1919 in single core
8929 in multi core.

Sure, Apple is much better at performance per watt than Intel but it’s not a good look to fall behind in single core performance. Most day to day tasks are single core.

Apple upended the chip industry with the M1 but AMD and Intel came back swinging and it seems like Apple now needs to pull another rabbit out of the hat with the M3.

Well, one is a desktop chip that eats 80W+ and the other one badly hits 25W. Furthermore, an i9 is more M2 Ultra territory than M2 territory.

 

[jav6454]

At 125W it would be an embarrassment if it weren’t ahead.

I laughed too hard at this. Thank you for pointing out this hard truth for Intel fans. Kudos.