Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads

[Tigerick]

Wildcat Lake (WCL) Preliminary Specs

Intel Wildcat Lake (WCL) is upcoming mobile SoC replacing ADL-N. WCL consists of 2 tiles: compute tile and PCD tile. It is true single die consists of CPU, GPU and NPU that is fabbed by 18-A process. Last time I checked, PCD tile is fabbed by TSMC N6 process. They are connected through UCIe, not D2D; a first from Intel. Expecting launching in Q2/Computex 2026. In case people don't remember AlderLake-N, I have created a table below to compare the detail specs of ADL-N and WCL. Just for fun, I am throwing LNL and upcoming Mediatek D9500 SoC.

		Intel Alder Lake - N	Intel Wildcat Lake	Intel Lunar Lake		Mediatek D9500
Launch Date		Q1-2023	Q2-2026 ?	Q3-2024		Q3-2025
Model		Intel N300	?	Core Ultra 7 268V		Dimensity 9500 5G
Dies		2	2	2		1
Node		Intel 7 + ?	Intel 18-A + TSMC N6	TSMC N3B + N6		TSMC N3P
CPU		8 E-cores	2 P-core + 4 LP E-cores	4 P-core + 4 LP E-cores		C1 1+3+4
Threads		8	6	8		8
Max Clock		3.8 GHz	?	5 GHz
L3 Cache		6 MB	?	12 MB
TDP		7 W	Fanless ?	17 W		Fanless
Memory		64-bit LPDDR5-4800	64-bit LPDDR5-6800 ?	128-bit LPDDR5X-8533		64-bit LPDDR5X-10667
Size		16 GB	?	32 GB		24 GB ?
Bandwidth			~ 55 GB/s	136 GB/s		85.6 GB/s
GPU		UHD Graphics		Arc 140V		G1 Ultra
EU / Xe		32 EU	2 Xe	8 Xe		12
Max Clock		1.25 GHz		2 GHz
NPU		NA	18 TOPS	48 TOPS		100 TOPS ?

As Hot Chips 34 starting this week, Intel will unveil technical information of upcoming Meteor Lake (MTL) and Arrow Lake (ARL), new generation platform after Raptor Lake. Both MTL and ARL represent new direction which Intel will move to multiple chiplets and combine as one SoC platform.

MTL also represents new compute tile that based on Intel 4 process which is based on EUV lithography, a first from Intel. Intel expects to ship MTL mobile SoC in 2023.

ARL will come after MTL so Intel should be shipping it in 2024, that is what Intel roadmap is telling us. ARL compute tile will be manufactured by Intel 20A process, a first from Intel to use GAA transistors called RibbonFET.

 

[igor_kavinski]

Lunar Lake is a slam dunk.

That slipped my mind. OK, I'll give him that. But considering everything else he's been responsible for, maybe this was a fluke?

Seriously though, all he needed to make sure was that Arrow Lake delivered. I wouldn't be lambasting him right now if he had done that one thing. It pains me to see Arrow Lake posting great scores in some benchmarks despite having only 24 threads. It means it had real potential that may or may not be realized before the next Core Ultra iteration. For all he cares, he will let the Intel CPU teams move on and make some new mistakes with Nova Lake

Kinda reminds me of how he handled Alchemist. If he had thrown more manpower on that to improve the drivers and especially the backward compatibility, it could've been a real competitive alternative for those of us who have some issue with either AMD or Nvidia or both or just want to try something new. Instead, he just let Alchemist rot. Guy seems not to care about hardware like an enthusiast does. His heart's turned into stone. Money is the only thing that motivates him now and he will say whatever crap he can come up with to keep being the CEO and making promises of a future that may never come. First it was, oh we'll be the leading fab in 2024 or 2025. What was his latest promise? Something like 2027 or 2028. Yeah, just keep moving the goalposts. Someone needs to put a stop to his chicanery.

 

[igor_kavinski]

Making fun of him for being religious is not criticism.

Pretending to be religious by tweeting verses isn't really being religious. Real religion is practicing one's faith and one of the tenets of his faith is honesty. How honest has he been with his investors and his customers? He's wasted tons of money on the promise of fabs with no real progress. Why wasn't Intel 3 good enough for Arrow Lake? That was his promise, wasn't it? To get Intel back to the leading edge of foundry processes. He's bungled way too many things to be given the benefit of the doubt.

 

[igor_kavinski]

You are estimating HT gain of LionCove based on RaptorCove and Lion is noticeably wider core than Raptor. It stands to reason it might have higher HT uplift than Raptor, since more of the core is likely to sit idle, but we will never know.

Thank you.

 

[igor_kavinski]

They're just 2012-era AMD. They may recover.

Did AMD have over 100K employees to feed?

 

[igor_kavinski]

I think he's laying a lot of important ground work success that the next CEO will get credit for.

Oh, so you TOO think Pat's not gonna last long

 

[OneEng2]

That's a lot of layers of obfuscation that eventually points back to MLID and his "sources"

ARL could have issues, but I'd definitely want to see more than his drama of the week video on it.

Fair point.

Both AMD and Intel have cores optimized for ST and MT. The rationale seems to be more for area efficiency rather than energy efficiency as Intel stated when Alder Lake was architecturally released years ago.

AMD's c cores are an area optimized version of its big cores while Intel's area optimized cores are a completely different architecture. But the point is the same, sacrifice ST performance (lower frequency and/or IPC) for more compute in a given area.

AMD has opted to keep SMT so I would call that a "mild" hybrid design. Intel has gone all in. Optimize some P cores for the best throughput possible in terms of IPC and frequency and optimize MT compute/area with E cores.

It took quite a few years for the E cores to go from many called "spam" to Raptor Cove level performance. Part 1 of the promise fulfilled. Part 2, which is the promise of significantly better P core is in doubt currently, I will admit that. I wonder how ARL would have done if it were monolithic?

The 9950X is a beast. I said it when lots of people were down on it at release and I'll say (write) it again now. It's a fabulous design and implimentation. ARL is competitive and one can argue specific use cases where it might be better, but overall, at this point in time I think we have to admit the 9950X is the king of the x86 desktop hill. Overall it wins more benches and is more energy efficient.

Zen5 core architecture is used in its P and E cores for desktop, laptop, and DC processors. Across this entire set of markets, AMD has a single generational architecture that works very well across all the markets. This has huge benefits for AMD in development time, validation, and risk.

If one is setting out with this design goal in mind, leaving out AVX512 and SMT doesn't seem like a good idea. Time will tell I guess.

As for performance/size optimization vs. performance/power optimization, I believe in the ultra-high volume laptop and desktop market, COGS is a HUGE consideration, therefore size optimization makes sense. In the server market, it makes no sense because the socket power delivery will become the bottleneck making performance/power optimization the most important factor.

It is my understanding that SMT is better for both performance/size and performance/power in general. This makes sense to me since getting the most performance out of the design per area and per watt seems the same from an architecture standpoint.

What AMD has done is made their core architecture flexible to different transistor libraries so they can render the same core targeting max single core performance (which takes more space, power, and clocks higher) or max performance per watt which takes less space, power, but clocks lower). They combined a bunch of multi-core CCD's into a single package and get economy of scale for each die. They have also made it so they only have to design and validate a single core design to accomplish all of this.

LibFLAC is not multithreaded. Pretty much all multithreaded audio encoders are more or less experimental. In any case, it makes more sense to just convert multiple tracks parallel. You can basically treat is at as pure single threaded operation in 99% of software out there that people would use.

It makes little sense to me that any audio encoder would be single threaded (although I have never written one). Why would audio processing be any different than video/image processing? You divide the file up into smaller pieces, then process each piece in a thread.

Audio files are so small that the only answer I have is that even single threaded audio encoding is very fast .... so why bother?

 

[SiliconFly]

.

 

[igor_kavinski]

The reality of the situation as explained to us by the stupid Intel engineers, which all of us are smarter than, and the really stupid CEO, who any of us could replace tonight and we could turn the company around by next week, is that anything that hinders ST performance is illogical and a detriment to the ARL design philosophy. You see thermal considerations and transistor budget are a real thing for those dumb engineers. Of course we know better but they are stuck in their old ways I guess.

When they first showed off those slides about getting rid of HT for better ST uplift, we all gained hopes that Lion Cove would be an ST screamer. It isn't. So what the heck were those slides for? To cover up their shame for being unable to get HT ready for prime time. I'm seriously tired of everyone associated with Intel who thinks it's completely OK to lie because they have been doing things this way for so long. Would've been a simpler thing for them to just admit that due to a myriad of complications, HT won't be available this year but they will try their best to sneak it into the refresh next year. I wouldn't be complaining as much then.

 

[controlflow]

Remove the AI bullshit benchmarks in there and its slower in 90% of applications. Why would you remove photoshop as an outlier when it is a legit application?

Remove the "AI bullshit" unless it is the AI benchmark that Zen 5 wins at. Very objective, ok.

By what metric? Intel is just as bad now if not worse off than under any previous CEO. Their future is also worse than ever. Pat is the worst CEO in Intel history at this point since he is driving the company to bankruptcy. Intel stock in down >60% since he took over.


Because the CEO is responsible for the company and its in their job description? Why does anyone need to be an "armchair CEO" to give criticism?

Pat has only been Intel CEO since 2021. The development timescales on the product design side and especially the process node side can be 5+ years long and often require tremendous up front capital expenditure. You can make mostly all good decisions and invest well for the future but it takes years to realize the results all while you are still suffering from poor decisions made in previous years.

It is fair to start grading Pat on the basis of how good PTL, CWF and 18A are and much more so with subsequent 2026 products.

 

[moinmoin]

Lunar Lake is a slam dunk.

Why is it reportedly a one off instead the base of things to come then?

Lip-Bu Tan resigned because Pat was laying off only 15% of the workforce - he wanted much bigger layoffs.

Reports were that he wanted middle management to be cut. Which should be a good idea considering that this is the level where much of the unneeded internal politics tends to go on.

"Tan wanted specific cuts, including middle managers who do not contribute to Intel's engineering efforts."

"Tan has told people he believed Intel was overrun by bureaucratic layers of middle managers who impeded progress at Intel’s server and desktop chips divisions and the cuts should have focused on these people."

https://www.reuters.com/technology/intel-board-member-quit-after-differences-over-chipmakers-revival-plan-2024-08-27/

 

[SiliconFly]

.

 

[igor_kavinski]

even Lisa would have messed it up equally or worse if put under the same situation.

Disagree. She's got a real talent to get people from different teams/companies to work together fruitfully. The PS3 Cell CPU took IBM, Toshiba and Sony working in tandem to create. She made those collaborative meetings happen and her social wizardry resulted in productive progress. She is also good at weeding out weak actors from AMD such as Raja Koduri. She isn't as easily fooled as Pat.

 

[SiliconFly]

.

 

[igor_kavinski]

Oh pls. It took her 3 years to come out with Zen!

From a team who initially thought they couldn't create a world class CPU (as revealed by Jim Keller). I suppose if Pat had been in her place, he would've gone for the kill, "Let's get GloFlo back!".

 

[GTracing]

Pretending to be religious by tweeting verses isn't really being religious. Real religion is practicing one's faith and one of the tenets of his faith is honesty. How honest has he been with his investors and his customers? He's wasted tons of money on the promise of fabs with no real progress. Why wasn't Intel 3 good enough for Arrow Lake? That was his promise, wasn't it? To get Intel back to the leading edge of foundry processes. He's bungled way too many things to be given the benefit of the doubt.

Companies fail to hit their goals all the time. Sometimes you can do your best and still fail. I would not accuse someone of being a bad Christian because of poor performance at work.

 

[igor_kavinski]

It is fair to start grading Pat on the basis of how good PTL, CWF and 18A are and much more so with subsequent 2026 products.

That's a really nice job to have. No wonder he's in such high spirits on stage all the time. No need to show any progress until he's filled his coffers with 5+ years of excellent remuneration and a lavish lifestyle.

 

[SiliconFly]

.

 

[MoistOintment]

Why is it reportedly a one off instead the base of things to come then?

The "one off" aspect of LNL is its MoP. Which, while disappointing, makes sense considering how the Windows OEM business works and how it hurts Intel's margins to have to include the memory at cost.

But the SoC design of LNL is certainly carrying forward into PTL.

 

[igor_kavinski]

Next year we have some major redesign with Panther Lake with memory controller integrated in the core tile. Followed by Nova Lake with Arctic Wolf E cores that might surpass P core in IPC.

Sounds very promising. Though NVL isn't until 2026. Too long a wait.

 

[MoistOintment]

That's a really nice job to have. No wonder he's in such high spirits on stage all the time. No need to show any progress until he's filled his coffers with 5+ years of excellent remuneration and a lavish lifestyle.

More so I believed him when he said, back in 2021, that it would take him 5 years to correct Intel.

i dont expect him to stay CEO until 2030, and in 2 more years he'll be 65. There was an interview a while back when he was first hired when he discussed that he had this plan with the board and told them that it would take him 5 years to course correct, during his interview.

I fully believe that he'll retire sometime in 2026, maybe 2027, and Intel will look a lot healthier by the late 2020s, with the time between then and now being rough.

And Intel's stronger future (after 18A, when 14A is hitting the market, unified core is out, etc.) Will happen under a new CEO, and that CEO will probably get the praise of the press.

 

[desrever]

Yeah. And the previous two CEOs were godsend. Pat is trying hard to fix years of Intel's failure 14++++++, 10+++++. Fixing a decade of failure and turning the ship around isn't a trivial task.

Lets put it this way which you may like better... even Lisa would have messed it up equally or worse if put under the same situation.

He hasn't shown that he has been able to fix anything. Lisa would have made very different decisions than Pat. Whether she would have done better is irrelevant tho since we are looking at Pat's failings.

Making fun of him for being religious is not criticism.

His actions of tweeting Bible verses publicly as Intel's CEO is worthy of criticism. Which successful CEO does this?

 

[Hulk]

Almost 17 million in compensation for Pat last year. Most in INTC so, you know, their/his future at stake with that compensation.

 

[Markfw]

Almost 17 million in compensation for Pat last year. Most in INTC so, you know, their/his future at stake with that compensation.

Maybe I am off my rocker, but to get that kind of money, you should be getting results like Lisa Su !

 

[desrever]

Pat has only been Intel CEO since 2021. The development timescales on the product design side and especially the process node side can be 5+ years long and often require tremendous up front capital expenditure. You can make mostly all good decisions and invest well for the future but it takes years to realize the results all while you are still suffering from poor decisions made in previous years.

It is fair to start grading Pat on the basis of how good PTL, CWF and 18A are and much more so with subsequent 2026 products.

Its fare to judge his other actions while as CEO. Just off the top of my head we have:

1. Failing to execute a proper AI plan
2. Not cutting dividend until the last minute, revenue/profit drop without any sign that Intel saw it coming
3. Over hire during a temp pandemic boost by completely missing the boat on forecasting
4. Firing >25% of workforce due to over hiring and other failures, losing most of their best engineers because of this kind of terrible decisions
5. Betting everything on foundry but still can't seem to secure chip act funding in a timely manner
6. Even bothering to continue development on ARL when it basically can't beat previous gen.
7. Poor QA leading to issues with RPL, he should know it is coming if he has a good grasp over the engineering

Just seems like Intel's is full of yes men at the top and Pat is listening to them because he doesn't have a good understanding of anything. He didn't take the challenge seriosly on day 1 and now, 3 years later there is still no real progress.

 

[moinmoin]

Oh pls. It took her 3 years to come out with Zen!

It took Mark Papermaster 5 years to assemble the new leadership and get AMD to launch Zen. Lisa Su joined on the way and, after becoming CEO, added much needed focus to the business side when AMD was at its lowest point.

More so I believed him when he said, back in 2021, that it would take him 5 years to correct Intel.

This matches above timeline at AMD. But Intel obviously is a much bigger company than AMD at its lowest point so business has to go on in the interim time. I personally think Intel could use more focus, and some of the current issues in execution are almost comically bad. But I'm happily accepting any positive surprises from Intel in 2026.

But the SoC design of LNL is certainly carrying forward into PTL.

I sure hope so, the big difference between ARL and LNL is LLC's latency.

Almost 17 million in compensation for Pat last year. Most in INTC so, you know, their/his future at stake with that compensation.

It's hard to get a good CEO. Intel tried really hard to get a good one after Krzanich, Swan already was a stopgap. Pat apparently could only be lured with that kind of money.