Intel 12900HK is faster than M1 Max for both single and multi cores.

[JMacHack]

SKU	Cores P+E/T	E-Core Base	E-Core Turbo	P-Core Base	P-Core Turbo	IGP	Base W	Turbo W	Price
i9-12900K	8+8/24	2400	3900	3200	5200	770	125	241	$589
i9-12900KF	8+8/24	2400	3900	3200	5200	-	125	241	$564
i7-12700K	8+4/20	2700	3800	3600	5000	770	125	190	$409
i7-12700KF	8+4/20	2700	3800	3600	5000	-	125	190	$384
i5-12600K	6+4/16	2800	3600	3700	4900	770	125	150	$289
i5-12600KF	6+4/16	2800	3600	3700	4900	-	125	150	$264

Up to 241 W on the spec sheet for the desktop parts.

Maybe someone more versed in cpu architecture can answer me, but what’s the point in having a turbo clock for the efficiency cores? I’m assuming the turbo clock pushes the core outside it’s “sweet spot” and just increases the tdp for diminishing returns.

If the point of e cores is to offload low power tasks, then a turbo clock seems unnecessary.

 

[Wolff Weber]

Maybe someone more versed in cpu architecture can answer me, but what’s the point in having a turbo clock for the efficiency cores? I’m assuming the turbo clock pushes the core outside it’s “sweet spot” and just increases the tdp for diminishing returns.

If the point of e cores is to offload low power tasks, then a turbo clock seems unnecessary.

For benchmark scores?

 

[ghstmars]

Maybe someone more versed in cpu architecture can answer me, but what’s the point in having a turbo clock for the efficiency cores? I’m assuming the turbo clock pushes the core outside it’s “sweet spot” and just increases the tdp for diminishing returns.

If the point of e cores is to offload low power tasks, then a turbo clock seems unnecessary.

Cmaier …..????!!!!

Maybe someone more versed in cpu architecture can answer me, but what’s the point in having a turbo clock for the efficiency cores? I’m assuming the turbo clock pushes the core outside it’s “sweet spot” and just increases the tdp for diminishing returns.

If the point of e cores is to offload low power tasks, then a turbo clock seems unnecessary.

 

[Romain_H]

Cmaier …..????!!!!

Suspended… he‘s around somewhere else now

 

[leman]

Maybe someone more versed in cpu architecture can answer me, but what’s the point in having a turbo clock for the efficiency cores? I’m assuming the turbo clock pushes the core outside it’s “sweet spot” and just increases the tdp for diminishing returns.

If the point of e cores is to offload low power tasks, then a turbo clock seems unnecessary.

Im definitely not versed in CPU architecture, but if I might be so bold to offer my speculation… its all about a certain performance range that might make sence for the chip. E-cores on Alder Lake do not fulfill the same function as E-cores on ARM chips, neither are they as efficient. Their purpose is power-efficient multithreaded throughput. So having a turbo range makes sense and will allow the chip to better adapt to various workloads (e.g. if you have fewer threads to run). In fact, not using turbo is going to cost you performance.

 

[JimmyjamesEU]

Not sure if this is the correct thread, but couldn't see a better one currently!

https://twitter.com/x/status/1453035362591420419

An interesting improvement for the M1 Pro and Max.

 

[R2FX]

Intel Alder Lake Mobility CPU Benchmarks Leaked: Faster Than The Apple M1 Max, Smokes AMD 5980HX, 11980HK

So we have something very interesting for our readers today. We were able to exclusively get our hands on the first-ever benchmarks for Intel's upcoming Alder Lake mobility processors - which are going to go head to head against the Apple M1 Max processors clawing away market share as well as...

wccftech.com

How much for your vapor ware right now?😆

 

[m1maverick]

Maybe someone more versed in cpu architecture can answer me, but what’s the point in having a turbo clock for the efficiency cores? I’m assuming the turbo clock pushes the core outside it’s “sweet spot” and just increases the tdp for diminishing returns.

If the point of e cores is to offload low power tasks, then a turbo clock seems unnecessary.

Geekbench, which is one of the reasons I dislike GB.

 

[Taz Mangus]

The new Intel cpu is much faster and that’s only because Apple decided to use only 8 performance cores instead of 32.
In other words: It’s Apples fault.

Vapor-ware.

 

[Taz Mangus]

Not sure if this is the correct thread, but couldn't see a better one currently!

https://twitter.com/x/status/1453035362591420419

An interesting improvement for the M1 Pro and Max.

Stomps all over the Ryzen.

 

[Taz Mangus]

Anyone find a benchmark for the Intel 12900HK iGPU. I am surprised we aren’t seeing that bad boy hyped up as well. Now I wonder….. why that could be. Hmmm…

 

[dapa0s]

I love these threads.

Apple destroy completion : “omg! Look at these leaks. So much power! So good!”

Intel perform better than apple : “got to wait for real world tests”

Yeah, the copium is really high in this thread.

 

[Rigby]

Maybe someone more versed in cpu architecture can answer me, but what’s the point in having a turbo clock for the efficiency cores? I’m assuming the turbo clock pushes the core outside it’s “sweet spot” and just increases the tdp for diminishing returns.

If the point of e cores is to offload low power tasks, then a turbo clock seems unnecessary.

During the press briefing, Intel made a point that four E-cores occupy the same space on the die as one P-core, but together have significantly higher performance than one P-core when the task is multi-threaded (i.e. can be parallelized). So it is not just about saving power (which isn't a high priority on desktop CPUs anyway), but also to offer a better balance of single- and multi-thread performance (multi-threading was a bit of a weakness compared to AMD's CPUs in the past).

Besides, Intel's "little" cores aren't really that little. They are about as fast as their Skylake-based cores from a few years ago.

 

[Melbourne Park]

On the Apple tax:
A 8 months into Covid, I called HP service, querying whether their 32" highish end monitor would run two notebooks at the same time, half screen (for my wife working at home, one was a temp work notebook, both HPs, each with a single T3 / USB-C displayport cabability which would also charge the T-3 connected notebook). HP service confirmed it would work, I bought the monitor. My wife's HP is an Elitebook and was still under warranty and also even on-site support (not valid though with Covid). The monitor would not do as promised. It took me around 28 hours of test software test software downloads uploads and 8 hours on the phone over six days. 9th day I requested the monitor be returned. Then later, I had to seek legal redress in order to get HP to repay the money - because they would not answer, would not arrange to refund the money, it was always in someone else's area, responsibility, etc etc etc I spent a couple of hours and 8 phone calls and innumerable texts and emails trying to get the money refunded 6 weeks after the monitor went back.

I bought a 15% more expensive Dell 32" and it worked fine.

I should have paid the Dell tax I guess (the extra 15%) but the issue is: how does one know? With Apple, my experience would have been a lot better.

Further on the tax: My wife got a 14" DEC notebook in about 1993 - it cost $14,000. It was very compact though. She hardly used it. OK she's a crap typer. and had a secretary. But her IT guys loved it.

I paid in 1991 about $20,000 for a 80486 to do a job for a merchant bank. I still can't belief I paid all that for what quickly became a piece of junk.

Don't talk to me about the PC tax thanks. I have a games machine in my home office, it doesn't work, its built by my son, but its crap. Its got faults all over it. Plus the cooling has never worked, so the noisy fans are full speed, it sounds like there's a motor bike in the room.

And you know, if Intel had given Apple powerful integrated CPU/GPUs, who knows whether Apple would have ever jumped - they may have hung and hung on. Like they had being doing for years ... with Intel not giving Apple would they were pleading for.

Now Apple have got what they wanted. And for some of us, it's better for us that way.

 

[TrueBlou]

I've always been of the position that Apple (or ARM) is not magical and that other companies, especially one dedicated to making microprocessors, can be competitive. Intel has had trouble producing processors on anything other than 14nm for quite some time. As the M1 is manufactured on 4nm process technology that alone gives it a big step up on Intel. That said until Alder Lake ships and is in systems which can be benchmarked I am not going to declare it a winner.

Either way competition is a good thing.

I’m sure it was a typo, but M1 is still 5nm.
And it’s anything but magical. Beating the performance of the 1st gen M1 is most definitely not impossible. It will be done, and probably before too long.

Beating it on a performance to watt ratio though, that’s probably going to be a bit of a stretch for the moment.

But the sooner they do, the better (for me, well, everyone I suppose), I’ve been wanting to upgrade one of my Windows servers for a while

 

[Taz Mangus]

I’m sure it was a typo, but M1 is still 5nm.
And it’s anything but magical. Beating the performance of the 1st gen M1 is most definitely not impossible. It will be done, and probably before too long.

Beating it on a performance to watt ratio though, that’s probably going to be a bit of a stretch for the moment.

But the sooner they do, the better (for me, well, everyone I suppose), I’ve been wanting to upgrade one of my Windows servers for a while

We could be seeing the same evolution on the M series that we have seen with the A series. The QualComm SnapDragon, Samsung Exynos, Huawei Kirin all don't seem to be able to compete with Apple's A series processors on efficiency and performance. Apple always seems to be able to do more with less. I think Apple will be able to do the same with the M series chips. Apple has gotten really good at performance and efficiency, and the performance per watt is simply amazing. The trend in the PC market has been high performance and high wattage processors and GPUs. This is Apple's first generation computer silicon and if the A series is any indication, the Apple's M series will follow the same trend, high performance and low wattage. It is speculated that next year Apple will move to using Arm V9 cores. That should give their silicon more performance than they have now and couple that with moving to 4nm which will help with better power efficiency, run cooler.

 

[RamGuy]

Intel's 10nm are more or less equivalent to TSMC's 7nm. https://www.pcgamer.com/chipmaking-process-node-naming-lmc-paper/

Intel's biggest problem has been their massive lead over AMD and their competition with their Intel Core 2 Duo "Allendale" release back in 2007. They've been resting on their laurels ever since.

Their struggles moving beyond 14nm has been causing some additional issues for Intel, but their biggest issue has been their minimum effort approach when it comes to improving their architecture and CPU design due to their lack of competition. Now they are finally back at it and Alder Lake looks really interesting and promising.


This is basically Steve Jobs story about IBM all over again.

 

[TrueBlou]

We could be seeing the same evolution on the M series that we have seen with the A series. The QualComm SnapDragon, Samsung Exynos, Huawei Kirin all don't seem to be able to compete with Apple's A series processors on efficiency and performance. Apple always seems to be able to do more with less. I think Apple will be able to do the same with the M series chips. Apple has gotten really good at performance and efficiency, and the performance per watt is simply amazing. The trend in the PC market has been high performance and high wattage processors and GPUs. This is Apple's first generation computer silicon and if the A series is any indication, the Apple's M series will follow the same trend, high performance and low wattage. It is speculated that next year Apple will move to using Arm V9 cores. That should give their silicon more performance than they have now and couple that with moving to 4nm which will help with better power efficiency, run cooler.

You’re preaching to the choir my friend, I’ve just taken delivery of my 3rd M1 system and I’m currently trying not to drool all over its keyboard

But I’m not naive enough to believe that they are unbeatable on a pure performance level. It will happen.

But as I say, it’s performance per watt where I think Apple will remain dominant for the foreseeable future.
As you mentioned yourself, they have proven time and time again just how much they can eek out of their hardware on iOS/iPadOS, when compared to the rest of the mobile ARM based devices.

Intel/AMD, coupled with Nvidia/AMD again, could thrash the M1 at some point, but they will be doing so with considerably higher power requirements.

But M1, even as a first generation product is amazing. Hell, it has a higher Geekbench score for single core processing than the 2,500 quid Xeon W-3175X.
And my new M1 Pro just craps all over the i9-10900 I use in my Windows server, in Geekbench single and multi-core.

Regardless of what happens with the main chipmakers over the next few years, it’s going to be an interesting journey. And the ones who will benefit most, are us, no matter the platform you use.

 

[leman]

It is speculated that next year Apple will move to using Arm V9 cores. That should give their silicon more performance

Why would Apple use ARM v9 cores? Their current proprietary cores are already faster than ARM‘s upcoming cores. Or do you mean adopting v9? Why would that improve performance?

 

[Taz Mangus]

Why would Apple use ARM v9 cores? Their current proprietary cores are already faster than ARM‘s upcoming cores. Or do you mean adopting v9? Why would that improve performance?

Arm’s Solution to the Future Needs of AI, Security and Specialized Computing is v9

Today, Arm introduced the Armv9 architecture in response to the global demand for ubiquitous specialized processing with increasingly capable security and artificial intelligence (AI).

www.arm.com

 

[pshufd]

How much for your vapor ware right now?😆

Now keep in mind, I have no qualms that Apple is still going to win on a power efficiency metric - they always have since the A11 - but Apple's reign as the fastest mobility chip "period" seems like it is going to be short-lived (we expect ADL-P to land in early 2022). Another important point to note here is that this benchmark was done using Windows 11, which allows Intel's brand new Thread Director technology to run - so some of these gains could be the result of better hardware scheduling as well.

Apple's design goals are PPW so that they can make these thin and light laptops. That was their design goal for the A15 so I don't expect M2 to be that much faster but do expect it to be more power-efficient.

 

[leman]

Arm’s Solution to the Future Needs of AI, Security and Specialized Computing is v9

Today, Arm introduced the Armv9 architecture in response to the global demand for ubiquitous specialized processing with increasingly capable security and artificial intelligence (AI).

www.arm.com

Yeah, I am very well of this. I just don’t understand what it has to do with Apple ir why you say that ARM v9 brings performance improvements. It’s mostly consists of security oriented ISA extensions for servers.

 

[Taz Mangus]

Yeah, I am very well of this. I just don’t understand what it has to do with Apple ir why you say that ARM v9 brings performance improvements. It’s mostly consists of security oriented ISA extensions for servers.

Apple can't just continue to add more cores without diminishing returns on performance. This was the article that caught my eye originally:

Digitalengr.com

 

[leman]

Apple can't just continue to add more cores without diminishing returns on performance. This was the article that caught my eye originally:

Digitalengr.com

That article claims that A15 would support ARM v9. Well, it didn’t. I still don’t understand why you are claiming that ARM v9 improves performance. The only thing in there that has anything to do with performance is SVE (but even that just for few specific applications), and you don’t need to adopt the entire v9 to have SVE (which is an optional v8 extension).

 

[JMacHack]

I’m sure it was a typo, but M1 is still 5nm.
And it’s anything but magical. Beating the performance of the 1st gen M1 is most definitely not impossible. It will be done, and probably before too long.

Beating it on a performance to watt ratio though, that’s probably going to be a bit of a stretch for the moment.

But the sooner they do, the better (for me, well, everyone I suppose), I’ve been wanting to upgrade one of my Windows servers for a while

I don’t think anyone is silly enough to think the M1 series will always be unbeatable.

I’ve said this before, but I think Intels real problem is within the company and not necessarily it’s technology. There have been reports of temp workers and salaried workers clashing, and I think that’s the root of their problems.

Until Intel can quash the infighting, they’re going to have an uphill fight against the competition.

As cmaier said, Apple has an architecture lead, not just a process lead, so Intels fight is more than just getting a process node advantage again, it’s making their architecture better.

So far, since Ryzen happened, Intel’s M.O. for maintaining performance leads has been to underrate TDP and clock the nuts off their processors. Efficiency be damned.

I dunno how long it will take for Intel to dig themselves out of this hole, or if they can, but for the foreseeable future they’ve got a big disadvantage.