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.

 

[Io Magnesso]

I always thought the problem was it was weak compared to RaptorLake; in gaming

The clock is low, so it can't be helped
Even so, it's only a few percent slower, and it's much less Power consumption than Raptor Lake...

 

[Io Magnesso]

That was almost inevitable, barring some breakthrough the likes of which Intel hasn't had in years. Raptor Lake flew too close to the sun. Arrow Lake isn't as suicidal.

Raptor Lake certainly got too close to the sun,
However, rather than saying that the movement at 6GHz led to deterioration.
I think it's a problem with the voltage control algorithm.
It's a scary place that this problem can happen with any manufacturer's processor.

 

[vanplayer]

Why is there so much stuff coming about Nova Lake, where is Panther Lake is so quiet?

Sometimes people are interested in something that's still on paper rather than something which is already in hand.

 

[dullard]

I always thought the problem was it was weak compared to RaptorLake; in gaming

And I included Raptor Lake in my post. 3% to 5% slower at 25% to 33% less power. Most people wouldn't even notice the difference in 4% fewer FPS. That is why I said they are on par.

 

[OneEng2]

For all our sakes, lets hope ARL and NVL do well and that Intel corrects its cost structure problems.

 

[511]

NVL Cost structure is much more better than ARL for sure they will be able to use Intel Advanced Accounting ™.

 

[OneEng2]

NVL Cost structure is much more better than ARL for sure they will be able to use Intel Advanced Accounting ™.

LOL.

Where will they hide the weeny once the foundry is spun off?

 

[AcrosTinus]

For all our sakes, lets hope ARL and NVL do well and that Intel corrects its cost structure problems.

I really dislike the fake concern about production costs of X-product or process to a company (investors aside). All they need to do is have a competitive product but who am I fooling, meritocracy is a bigger joke than it was in the past.

In no reality will people praise Intel, they just aren't popular with the media crowd which is quite depressing even for the employees. People are basically betting on the downfall or counting the days before Intel turns into IBM.

I can hear the voices "they are faster but at a significantly higher power...; we don't know about platform longevity, this could be the only product on the socket, you will need a new mobo, the difference in only x%-you can get Zen25 on this board, so just wait; they are faster but we are already at very high framerates"

I could go on and on because I have seen it happen with coffee lake and Alder lake, when Intel wins, the media starts mentioning caveats. Ohhh nooo my CPU takes an extra 40W while gaming compared to AMD-Xyz while being next to a 450W GPU, 30W of fans and 25W of RGB.... I cannot....

 

[igor_kavinski]

I could go on and on because I have seen it happen with coffee lake and Alder lake, when Intel wins, the media starts mentioning caveats. Ohhh nooo my CPU takes an extra 40W while gaming compared to AMD-Xyz while being next to a 450W GPU, 30W of fans and 25W of RGB.... I cannot....

It's the job of the free press to report on both sides of the coin. Intel's job was to keep executing which they failed again and again with Raptor, Meteor and Arrow Lake. Sure, Lunar Lake is great but the volume just isn't there. I have not seen any 288V laptops available on my local Amazon. Limiting the RAM capacity to 32GB was also unfortunate when Snapdragon went up to 64GB for only a few hundred dollars more. Intel wants better press? Intel should work harder and more honestly instead of letting their dirty internal politics ruin their overall image in the minds of the average consumer.

You know what's funny? I can think of two fairly high profile AMD firings: Raja Koduri and Robert Hallock. Both ended up at Intel. Coincidence?

 

[Thunder 57]

I really dislike the fake concern about production costs of X-product or process to a company (investors aside). All they need to do is have a competitive product but who am I fooling, meritocracy is a bigger joke than it was in the past.

In no reality will people praise Intel, they just aren't popular with the media crowd which is quite depressing even for the employees. People are basically betting on the downfall or counting the days before Intel turns into IBM.

I can hear the voices "they are faster but at a significantly higher power...; we don't about platform longevity, this could be the only product on the socket, you will need a new mobo, the difference in only x%-you can get Zen25 on this board, so just wait; they are faster but we are already at very high framerates"

I could go on and on because I have seen it happen with coffee lake and Alder lake, when Intel wins, the media starts mentioning caveats. Ohhh nooo my CPU takes an extra 40W while gaming compared to AMD-Xyz while being next to a 450W GPU, 30W of fans and 25W of RGB.... I cannot....

Another one who believes in the "mass media (and techtuber) conspiracy"?

 

[jpiniero]

It's the job of the free press to report on both sides of the coin. Intel's job was to keep executing which they failed again and again with Raptor, Meteor and Arrow Lake. Sure, Lunar Lake is great but the volume just isn't there. I have not seen any 288V laptops available on my local Amazon.

Dell sells it:

Dell 14 Plus Laptop - 14 Inch Laptop | Dell USA

The Dell 14 Plus: Your ideal 14-inch laptop for work and play. Enjoy lightweight portability and reliable performance. Shop Dell.com today!

www.dell.com

It's not cheap though.

 

[AcrosTinus]

It's the job of the free press to report on both sides of the coin. Intel's job was to keep executing which they failed again and again with Raptor, Meteor and Arrow Lake. Sure, Lunar Lake is great but the volume just isn't there. I have not seen any 288V laptops available on my local Amazon. Limiting the RAM capacity to 32GB was also unfortunate when Snapdragon went up to 64GB for only a few hundred dollars more. Intel wants better press? Intel should work harder and more honestly instead of letting their dirty internal politics ruin their overall image in the minds of the average consumer.

You know what's funny? I can think of two fairly high profile AMD firings: Raja Koduri and Robert Hallock. Both ended up at Intel. Coincidence?

I understand what you are saying but you have to be honest, I am from Germany and here people have a soft spot for AMD. AMD was failing for a decade straight and I still had people blaming it on Intel and still buying AMD, basically bla bla bla anti-consumer, monopolistic and anticompetitive...(some claims have validity but the impact is overstated IMHO)


The truth is that their products are quite competitive and the standards they introduce are also quite popular in the industry also nodes are hard business, you can count the manufactures on one hand.

 

[igor_kavinski]

The truth is that their products are quite competitive and the standards they introduce are also quite popular in the industry

Not disagreeing with that but they could've been so much better than AMD if they hadn't made several panicky mis-steps such as hybrid core strategy, insistence on going with chiplets come hell or high water and refusal to introduce P-core only Alder Lake and Raptor Lake with AVX-512 enabled. In my opinion, they deserve the bad press. It's their own doing. I don't deny that their products are good. I have a 245KF that effortlessly does DDR5-8200 but what's the point? The rest of the chip is badly hobbled so that RAM overclock advantage vanishes into thin air. Had this been a monolithic chip with an IMC this good, man, it would've been unbelievable value.

 

[DavidC1]

1) The topic was laptop vs desktop, yet you only quote Intel numbers as if you forget AMD exists.
2) What Intel cares about is not what is being discussed.
3) What Intel revenue ratios might be are not what is being discussed.

Up until recently, and even now, they are the majority of the market, and they break down the data into greater detail which AMD does not do.

Back to the original debate. As for the claim that "Gaming isn't that big a deal to most consumers of x86 chips.",

Since we were talking about whether Novalake is important for Intel or not, and high end chips with high profitability surely does, the answer is a definite "Yes!".

2023 Intel data is also important in that they stated the increased desktop market partially offset the decrease in revenue, which would be due to competitive Alder/Raptorlake chips on the desktop, because laptops based on those chips weren't that very good.

People keep heavily underestimating the enthusiast market and also gaming's impact on driving marketshare. How that even makes sense at all when so many people entered the IT industry by gaming in the first place boggles my mind. Those same people are also the ones that persuade plain Janes and Joes into what to buy(likely away from Apple computers and WARM).

 

[AcrosTinus]

Not disagreeing with that but they could've been so much better than AMD if they hadn't made several panicky mis-steps such as hybrid core strategy, insistence on going with chiplets come hell or high water and refusal to introduce P-core only Alder Lake and Raptor Lake with AVX-512 enabled. In my opinion, they deserve the bad press. It's their own doing. I don't deny that their products are good. I have a 245KF that effortlessly does DDR5-8200 but what's the point? The rest of the chip is badly hobbled so that RAM overclock advantage vanishes into thin air. Had this been a monolithic chip with an IMC this good, man, it would've been unbelievable value.

I urge you to follow some tuning guides, like the PCGH or Skatterbencher if you haven't already. I got my memory latency down to 77ns, D2D is 3,4, NGU 3,3 and Ring at 4. 4,9E and 5,5P and 7200 UDIMM. It feels like Raptor Lake !
All I want to say is that with a memory bandwidth of over 110GB/s this chiplet implementation is flat out superior. They played it safe due to the Raptor Lake fiasco and the cache tiering in Lion Cove is not ideal, I suspect most of the issues can be related to bad caches in the P-Core. The overall fabric and design, after living with it, and tuning it, behaves like a monolithic chip.

Again, you make my point, all the things you mention are a result of hindsight knowledge.

• Hybrid core is not a mis-step but a natural reaction to competition. They scaled their arch to meet the MT scaling of the competition while working in with their constraints. (Ring, power and so on)
• Chiplet arch is not good, see my explanation above, it is more advanced than anything on the market (x86 consumer). Nothing comes close !
• They had Rocket Lake with AVX512, no one cared and even argued that mainstream does not need it. At that time AMD had a very weak AVX2 implementation and people said nothing ! Only when they fixed their AVX2 and later introduced or reintroduced AVX512 to the mainstream, people deem it important ? Again this just makes my point.
  • People(online) don't care about MT performance, but cared when AMD did it. People care about AVX512 but cannot find a use case for the MT performance Intel is pushing ? An oxymoron !

All I want to say is, cut the billion dollar company some slack, at the end it is a company of people building products only few can.

 

[OneEng2]

I really dislike the fake concern about production costs of X-product or process to a company (investors aside). All they need to do is have a competitive product but who am I fooling, meritocracy is a bigger joke than it was in the past.

In no reality will people praise Intel, they just aren't popular with the media crowd which is quite depressing even for the employees. People are basically betting on the downfall or counting the days before Intel turns into IBM.

I can hear the voices "they are faster but at a significantly higher power...; we don't know about platform longevity, this could be the only product on the socket, you will need a new mobo, the difference in only x%-you can get Zen25 on this board, so just wait; they are faster but we are already at very high framerates"

I could go on and on because I have seen it happen with coffee lake and Alder lake, when Intel wins, the media starts mentioning caveats. Ohhh nooo my CPU takes an extra 40W while gaming compared to AMD-Xyz while being next to a 450W GPU, 30W of fans and 25W of RGB.... I cannot....

I am definitely not hoping for Intel to fail. I also do not consider ARL to be a poor architecture.

My concern for Intel is their cost structure and lack of competitiveness in the fastest growing, highest margin x86 market (DC).

While many here believe that if Intel could only create the worlds greatest gaming chip, all would be well ..... somehow. The data does not support this. Intel was right to minimize their focus on gaming. They were wrong to lose their focus on DC. They were also late to realize there was no future in monolithic only products .... especially in DC, but even in desktop.

I understand what you are saying but you have to be honest, I am from Germany and here people have a soft spot for AMD. AMD was failing for a decade straight and I still had people blaming it on Intel and still buying AMD, basically bla bla bla anti-consumer, monopolistic and anticompetitive...(some claims have validity but the impact is overstated IMHO)


The truth is that their products are quite competitive and the standards they introduce are also quite popular in the industry also nodes are hard business, you can count the manufactures on one hand.

There were plenty of people on forums the world over that praised all things Intel in the days of Core 2 clear up to the day AMD released Ryzen (especially Ryzen 3). For many, Intel could do no wrong.

Ironically, many of those people are now ardent AMD supporters due to AMD's unusually strong position in gaming (despite gaming being only a fraction of the 30% desktop x86 sales compared to 70% laptop (and the lions share of that being corporate as well).

Not disagreeing with that but they could've been so much better than AMD if they hadn't made several panicky mis-steps such as hybrid core strategy, insistence on going with chiplets come hell or high water and refusal to introduce P-core only Alder Lake and Raptor Lake with AVX-512 enabled. In my opinion, they deserve the bad press. It's their own doing. I don't deny that their products are good. I have a 245KF that effortlessly does DDR5-8200 but what's the point? The rest of the chip is badly hobbled so that RAM overclock advantage vanishes into thin air. Had this been a monolithic chip with an IMC this good, man, it would've been unbelievable value.

I'll give you the AVX-512 complaint; but Intel is super late to chiplets and that failure to move has cost them the DC market which in turn has starved them of the cash they need to support their foundry and higher volume production on leading edge nodes. Note, I believe that Intel remaining vertically integrated was a strategic mistake starting back with the 18nm +++++++ timeframe.

People keep heavily underestimating the enthusiast market and also gaming's impact on driving marketshare.

Not sure what evidence exists to support this claim. Gaming appears to be responsible for a small fraction of x86 sales.

 

[jpiniero]

Gaming appears to be responsible for a small fraction of x86 sales.

I would say Chrome and Office are the two biggest... but they also benefit mainly from strong single/low thread performance, like games do.

OTOH Gaming is what drives DIY.

 

[Io Magnesso]

It's the job of the free press to report on both sides of the coin. Intel's job was to keep executing which they failed again and again with Raptor, Meteor and Arrow Lake. Sure, Lunar Lake is great but the volume just isn't there. I have not seen any 288V laptops available on my local Amazon. Limiting the RAM capacity to 32GB was also unfortunate when Snapdragon went up to 64GB for only a few hundred dollars more. Intel wants better press? Intel should work harder and more honestly instead of letting their dirty internal politics ruin their overall image in the minds of the average consumer.

You know what's funny? I can think of two fairly high profile AMD firings: Raja Koduri and Robert Hallock. Both ended up at Intel. Coincidence?

It's hard to be 64GB for a LUNAR LAKE that can only have 2 memory chips... With today's technology

 

[DavidC1]

Not sure what evidence exists to support this claim. Gaming appears to be responsible for a small fraction of x86 sales.

Would you recommend a gaming system for a person who doesn't need that much power? No.

Are you still influencing their purchasing decisions? Yes.

 

[511]

I understand what you are saying but you have to be honest, I am from Germany and here people have a soft spot for AMD. AMD was failing for a decade straight and I still had people blaming it on Intel and still buying AMD, basically bla bla bla anti-consumer, monopolistic and anticompetitive...(some claims have validity but the impact is overstated IMHO)


The truth is that their products are quite competitive and the standards they introduce are also quite popular in the industry also nodes are hard business, you can count the manufactures on one hand.

On one finger nowdays 🤣 we only have 3 leading edge manufacturer left. Also what standard did AMD ever introduced besides HBM?

I am definitely not hoping for Intel to fail. I also do not consider ARL to be a poor architecture.

My concern for Intel is their cost structure and lack of competitiveness in the fastest growing, highest margin x86 market (DC).

While many here believe that if Intel could only create the worlds greatest gaming chip, all would be well ..... somehow. The data does not support this. Intel was right to minimize their focus on gaming. They were wrong to lose their focus on DC. They were also late to realize there was no future in monolithic only products .... especially in DC, but even in desktop.

Tbh Intel launched chiplets around same time as AMD with more advanced design called Lakefield. AMD don't have to worry about manufacturing that is a big thing not being to worry about it.

I'll give you the AVX-512 complaint; but Intel is super late to chiplets and that failure to move has cost them the DC market which in turn has starved them of the cash they need to support their foundry and higher volume production on leading edge nodes. Note, I believe that Intel remaining vertically integrated was a strategic mistake starting back with the 18nm +++++++ timeframe.

They fired the SPR Validation team thanks BK and if not for 10nm and 7nm delays they would have been sooner if not for Foundry thingy that has been going on since BK/Swan.

 

[igor_kavinski]

All I want to say is, cut the billion dollar company some slack, at the end it is a company of people building products only few can.

I'm not strictly anti-Intel. I have more Intel hardware (only some FX chip, 2700X, 9950X3D and Zen 2 128 thread Epyc from AMD).

I think one reason people overpromote AMD is because they think (rightfully so) that AMD needs to gain more marketshare which would be great to fuel even more innovation and competition between the two x86 players and bring prices down for everyone. Let's face it. Even if Intel does come out with superior hardware in 2026, their prices will remain higher. The AMD overpromoters are trying to ensure that the general public knows that Intel isn't the only player in town, especially since a VAST majority of people don't even know that AMD exists. When they try to point out the benefits of AMD hardware, I think the underlying motive is more to educate the public than to downplay Intel's strengths. Intel was, after all, the 800 pound gorilla (and arrogantly behaving like one) for a very long time and this pissed a lot of people into going AMD and never returning.

 

[511]

We need competition in foundry space as well if we want cheaper prices

 

[Thunder 57]

On one finger nowdays 🤣 we only have 3 leading edge manufacturer left. Also what standard did AMD ever introduced besides HBM?

Tbh Intel launched chiplets around same time as AMD with more advanced design called Lakefield. AMD don't have to worry about manufacturing that is a big thing not being to worry about it.

They fired the SPR Validation team thanks BK and if not for 10nm and 7nm delays they would have been sooner if not for Foundry thingy that has been going on since BK/Swan.

I'm sure you could find a few if you look around. One you are using right now though would be x86-64, or you know, AMD64 as it was once called . And before you say it was based on an Intel design, think about what Intel was trying to do back then. Force IA64 on the world while leaving x86 to die off. I think it worked out for the best in the end.

 

[Thunder 57]

We need competition in foundry space as well if we want cheaper prices

Anyone who doesn't want more competition is a shill IMHO.

 

[igor_kavinski]

There were plenty of people on forums the world over that praised all things Intel in the days of Core 2 clear up to the day AMD released Ryzen (especially Ryzen 3). For many, Intel could do no wrong.

Ironically, many of those people are now ardent AMD supporters due to AMD's unusually strong position in gaming (despite gaming being only a fraction of the 30% desktop x86 sales compared to 70% laptop (and the lions share of that being corporate as well).

It's not just gaming that powers the AMD slant of the enthusiast crowd. Workloads on Ryzen CPUs are more deterministic due to not having hybrid cores that suddenly slow things to a crawl due to less refined hybrid core scheduling plus eventually a power user will have some important thread ending up on an E-core if doing a lot of multitasking and that may lead to frustration.

People are seeing good acceleration out of using AVX-512 supported applications (there's even Chromium browser AVX-512 builds that show definite and very palpable perf improvement).

Ryzen SMT is more performant than Intel HT so those with especially eight or fewer cores can expect a longer and more satisfying computing experience than people with hybrid Intel cores whose computing experience will get diluted a bit when the extra threads will have no choice but to run on the slower E-cores.

For Intel to win over this crowd in 26H2, their E-cores need to be VERY close in performance to Zen 6 cores and their P-cores need to convincingly beat or essentially tie the average performance of Zen 6 cores over a wide variety of workloads. People will switch back to Intel if given enough sufficient reasons to do so. Unfortunately, Intel (and specifically Pat) gave most users enough reasons to give up on them.