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.

 

[DrMrLordX]

they won’t be soldered onto to the mobo, it’s part of the SoC. It’s better like that, dGPUs are power hungry and the end game for laptops is using NVIDIA IP in an Intel SoC.

Personally I think we'll see both. There's still a market for mdGPUs and NV can sew it up nicely (especially since AMD may not produce any more mdGPUs from this point forward. We'll see).

It wouldn't be hard but it would be bad for power efficiency and obviously no unified memory. Besides, that's what they are doing since MXM lost its popularity.

For your typical gamer laptop, of which there are many, unified memory isn't much of a boon. It'll be hard for NV to replace 5070-tier products (respective to a given future product stack) with something that's lower power and fully integrated into the SoC.

 

[Magio]

View attachment 131885
actually twice as funny given Intel isn't even pretending PTL-H has much of a perf bump over LNL at 28W.

Am I crazy in thinking that's legit good news for 18A, tho?

Near identical CPU cores going from N3B to 18A leads to a significant efficiency gains in ST and MT, meanwhile on the GPU side new arch + wider GPU + going from N3B to N3E is a modest improvement only in terms of efficiency.

 

[Josh128]

No, you don't.

The market OVERWHELMINGLY wants iGPUs not dGPUs. Why are you continuing to beat a dead horse insisting that the "no one wants iGPUs" when that's 1) clearly not true, 2) dGPUs are a niche market, and 3) a shrinking niche at that.

Just had to jump in here. The market wants:

1) Sleek/thin
2) Light
3) Fast
4) Cool
5) Long battery life

Period.

 

[511]

Just had to jump in here. The market wants:

1) Sleek/thin
2) Light
3) Fast
4) Cool
5) Long battery life

Period.

Long battery and cool is only possible without dGPU

 

[poke01]

It depends on the size of the laptop.

13” and below iGPU is needed for efficiency. 14” and 15” users prefer to have an dGPU but a good iGPU like the 12 core would be fine for 1080p low-medium gaming.

16” and up a dGPU is a must, imo.

 

[DavidC1]

It is interesting that even Intel seems to be acknowledging that fact with respect to gaming. In that part of the improved PTL gaming performance comes from changing scheduling priority from the P cores to the E cores when GPU bound regardless. So yeah, for gaming purposes they certainly could offer a lower spec CPU with the 12Xe iGPU... but they won't. Better to tie them together to maximize profits.

Exactly, and I don't care about their profit margins. And I know they can offer a separate part and not lose money.

Just offer a SKU with 3 series and good graphics. And by the way, 12 Xe3 isn't even "good" anyway. It's not halo like Strix Halo. It's regular graphics with different branding.

You offer a 3 series chip with the anemic GPU that costs X dollars. And then you offer another 3 CPU with the top GPU config that's +$100. Call it Core Ultra 3G or whatever. That's how you actually grow, not the whole beancounter logic that's pervasive through the industry. I don't and won't make excuses for them. At the worst they want "software defined" silicon which is a sugarcoated way of saying "We have it in hardware, but we'll disable it and charge you $500 more for something that already exists". They are doing that with cars. They'll gatekeep engine horsepower behind mere electrical signals controlled by keystrokes(software).

You shouldn't care about their profit margin, or revenue. These guys overhire, then fire them at the slightest loss, and a quarter after that they announce "record revenues".

 

[OneEng2]

It's super weird that you're insisting so much on this flex when we know:

- AMD's Medusa range goes so hard on iGPUs that they basically converted their entire consumer RDNA5 dGPU range into chiplets that can go bolt onto APUs to improve iGPU performance.

- Intel's Nova Lake AX has up to 48 Xe3 cores, over twice as large as the B580 dGPU, with a whopping 144MB LLC and 256bit LP5X.

- Nvidia just gave $10B to Intel so they could put their iGPU chiplets into Intel's x86 APUs and be able to enter the big iGPU market.


Nvidia, Intel and AMD are all putting serious R&D money into these larger iGPUs you insist no one wants and no one uses.
Either all these 100billion/trillion companies (who undoubtedly spend millions on market research and user metrics) are collectively delirious at the same time on this specific subject, or you're just completely wrong.

It's not weird, he is just plain wrong and fixated on his own perception. Sales figures don't lie.

Just had to jump in here. The market wants:

1) Sleek/thin
2) Light
3) Fast
4) Cool
5) Long battery life

Period.

.. add inexpensive to the top of the list.

 

[DavidC1]

Near identical CPU cores going from N3B to 18A leads to a significant efficiency gains in ST and MT, meanwhile on the GPU side new arch + wider GPU + going from N3B to N3E is a modest improvement only in terms of efficiency.

It's 22% more efficient than Lunarlake at the same power, plus they are changing the power allocation so the CPU doesn't hog all the power like they did since the Extreme Graphics days.

 

[Josh128]

It's not weird, he is just plain wrong and fixated on his own perception. Sales figures don't lie.

.. add inexpensive to the top of the list.

Not really. If they did, AMD would have a much better position in the laptop market over the past 5 years. People these days apparently spend quite frivalously.

 

[511]

AMDs market position is due to the fact they refused to do OEMs work while Intel does it for them and OEMs are a incompetent bunch.

 

[Josh128]

Long battery and cool is only possible without dGPU

For the record, Im not fan of dGPUs in laptops, but thats simply not true. I have an iGPU thin and light Tiger Lake laptop that gets hot as hell during general office usage.

A proper dGPU can be just as efficient and cool as an iGPU. It just has to be engineered that way. Unfortunately, none really are because iGPUs are included on every CPU today.

 

[Josh128]

AMDs market position is due to the fact they refused to do OEMs work while Intel does it for them and OEMs are a incompetent bunch.

Thats AMDs problem as laptops are the largest TAM in the PC world.

 

[vanplayer]

Am I crazy in thinking that's legit good news for 18A, tho?

Near identical CPU cores going from N3B to 18A leads to a significant efficiency gains in ST and MT, meanwhile on the GPU side new arch + wider GPU + going from N3B to N3E is a modest improvement only in terms of efficiency.

Not crazy but just too optimistic. Don't be fooled by slides. No matter 18A or PTL are just too little too late.

 

[DavidC1]

Not crazy but just too optimistic. Don't be fooled by slides. No matter 18A or PTL are just too little too late.

Lunarlake delivered...

 

[511]

For the record, Im not fan of dGPUs in laptops, but thats simply not true. I have an iGPU thin and light Tiger Lake laptop that gets hot as hell during general office usage.

A proper dGPU can be just as efficient and cool as an iGPU. It just has to be engineered that way. Unfortunately, none really are because iGPUs are included on every CPU today.

OEMs forget to configure it properly properly but the problem with having dGPU Is random small prorgam gets scheduled to it somehow your laptop is hot.

 

[Hulk]

Based on the review from The Tech Report I created a little "cheat sheet" on Intel's release of Panther Lake info. I want to stress that this pulled directly from the excellent Tech Report article. Much of it is word-for-word. If there are errors they are most likely on me. Let me know and I will update.

As many of you know after 30+ years of me being "Intel Inside" I switched to a 9950x coming up on a year ago in December. I had degradation with 2 Raptors and was weary of fooling around with BIOS tweaking. Anyway, while I have lost some faith in the "old Intel," Panther Lake looks promising.

18A compute tile finally coming from Intel.. but where will clocks be
Cougar Lake P core with 5 to 10% IPC increase over Lion Cove.
Darkmont E core with as Intel seems to imply more substantial architecture changes than Cougar Lake. Could this mean 10%+ IPC over Skymont?
Updated fabric with near monolithic latency, again an Intel claim.
Verious memory subsystem improvements to "keeps things moving" among the various core types.
Significant efficiency improvement claims.

Looking at Arrow Lake.. performance is somewhat inconsistent, mainly tripping on itself in games, but for productivity it seems quite robust and (fingers crossed for Intel) reliable.

If Panther Lake has corrected some of the ARL inconsistencies, increased IPC in both cores, increased efficiency, and reduced latency, Panther Lake and the subsequent desktop counterpart could provide serious competition for Zen 6.

Intel is going for a lot all at once. I hope they can pull it off. Their tile based scheme might finally be coming into focus on the third effort. They don't appear to be going quietly, if they are going at all.

 

[DrMrLordX]

Am I crazy in thinking that's legit good news for 18A, tho?

Near identical CPU cores going from N3B to 18A leads to a significant efficiency gains in ST and MT, meanwhile on the GPU side new arch + wider GPU + going from N3B to N3E is a modest improvement only in terms of efficiency.

The underlying problem with Panther Lake on 18A is best encapsulated in one question: how many 5.1 GHz compute tiles can they yield per wafer on 18A? If what we're being told about 18A is true, binning for 5.1 GHz has not been easy. That's going to take its toll on margins at the very least.

 

[Hulk]

The underlying problem with Panther Lake on 18A is best encapsulated in one question: how many 5.1 GHz compute tiles can they yield per wafer on 18A? If what we're being told about 18A is true, binning for 5.1 GHz has not been easy. That's going to take its toll on margins at the very least.

According to Intel yields are "as good as or better than any product in the last 15 years."

I know, I know. That is a somewhat loaded and possibly laughable statement. So yields are better than the first crack at 14nm with Broadwell? Better than Cannon Lake? Not promising.

14nm was a disaster of ++++++++++'s.
Intel 7 was Raptor Lake primarily, Intel 4 Meteor Lake, Intel 3 Xeon.

So the question is out of Intel 3, 4, and 7, which had the best ramp up? That's how 18A is doing according to Intel, right?

 

[MoistOintment]

The underlying problem with Panther Lake on 18A is best encapsulated in one question: how many 5.1 GHz compute tiles can they yield per wafer on 18A? If what we're being told about 18A is true, binning for 5.1 GHz has not been easy. That's going to take its toll on margins at the very least.

If it's already been decided to bin 5.1Ghz as the Ultra 9, then what % of dies need to be good enough to qualify as the 9 SKU? 10%? 20%?

I dont know if Intel has ever publicly released the sales ratio of their 5 to 7 to 9, but I can't imagine 9 has ever exceeded 20% - 25% of laptop volume

 

[DrMrLordX]

If it's already been decided to bin 5.1Ghz as the Ultra 9, then what % of dies need to be good enough to qualify as the 9 SKU? 10%? 20%?

As many as people want to buy. Also don't you think they wanted a 5.3GHz part in that position, or higher? So they could sell 5.1 GHz parts lower down on the stack? But they can't do that now, and the entire product stack suffers for it. The top three SKUs have the exact same core config but different clockspeeds (if I recall correctly, they're all 4 + 8 + 4 + 12Xe3). We're all judging the CPU based on the 5.1 GHz SKU, but that SKU may not represent what the majority of people will be able to buy once it actually hits shelves in volume.

I dont know if Intel has ever publicly released the sales ratio of their 5 to 7 to 9, but I can't imagine 9 has ever exceeded 20% - 25% of laptop volume

They'll have to burn through wafers to keep up with demand, or let the market supply dry up. One or the other.

 

[Josh128]

14nm was a disaster of ++++++++++'s.
Intel 7 was Raptor Lake primarily, Intel 4 Meteor Lake, Intel 3 Xeon.

So the question is out of Intel 3, 4, and 7, which had the best ramp up? That's how 18A is doing according to Intel, right?

Re: Intel 7, you completely skipped Alder Lake, the one that doesnt degrade itself.

 

[HurleyBird]

Just had to jump in here. The market wants:

1) Sleek/thin
2) Light
3) Fast
4) Cool
5) Long battery life

Period.

And 6) inaudible, although this is really underserved. Wide consumer exposure to fanless arm devices makes laptops that audibly ramp up and down even more annoying than they already are.

 

[511]

According to Intel yields are "as good as or better than any product in the last 15 years."

I know, I know. That is a somewhat loaded and possibly laughable statement. So yields are better than the first crack at 14nm with Broadwell? Better than Cannon Lake? Not promising.

14nm was a disaster of ++++++++++'s.
Intel 7 was Raptor Lake primarily, Intel 4 Meteor Lake, Intel 3 Xeon.

So the question is out of Intel 3, 4, and 7, which had the best ramp up? That's how 18A is doing according to Intel, right?

It includes 22nm as well so the D0 might be good the parametric may not be that good

 

[Doug S]

And 6) inaudible, although this is really underserved. Wide consumer exposure to fanless arm devices makes laptops that audibly ramp up and down even more annoying than they already are.

What wide exposure? There's Apple, and that's it - and Macbook Air is the only one that's fanless. AFAIK none of the Qualcomm based laptops are fanless.

There's nothing stopping OEMs from making a fanless Intel based laptop. Stick a U chip in there, and let it throttle as needed to avoid exceeding Tmax. That's what Apple does for Macbook Air. The problem is that when it gets reviewed and benchmarked it'll be savaged because it'll be slower than comparable models that have a fan. Reviewers don't value silence, they are probably always surrounded by the hum of fans and likely wouldn't notice the silence if they didn't know about it in advance.

I've always gone out of my way to build very quiet PCs - quiet not silent mind you - and even that isn't made at all easy. The selection sucks - companies will sell stuff they claim is nearly silent but that's at idle not under load so it is a lot of work to put something together that will stay quiet even when you push it. So I don't buy that the market is underserved - I think I'm just an outlier, because if there was a market for this I wouldn't have to dig so much and have to try and return parts until I get it right. And sometimes have to take a multitool and start cutting metal to make things fit even when using what I thought would be a rather generous mATX sized case (since the mITX case I used previously was an even bigger issue) I'd like to go even smaller, but we're still using outdated standards like ATX for power, and futzing around connecting the motherboard to the front panel with jumper cables on terminal blocks like its 1975, so I don't hold out hope for any modernization on that front since everyone else seems to think that's a perfectly reasonable situation I guess.

 

[poke01]

What wide exposure

isnt that wide exposure though? The MacBook air sells ~10 million units every year.

There's nothing stopping OEMs from making a fanless Intel based laptop. Stick a U chip in there, and let it throttle as needed to avoid exceeding Tmax. That's what Apple does for Macbook Air. The problem is that when it gets reviewed and benchmarked it'll be savaged because it'll be slower than comparable models that have a fan. Reviewers don't value silence, they are probably always surrounded by the hum of fans and likely wouldn't notice the silence if they didn't know about it in advance

For that to happen Intels cores need massive improvements. I remember watching a review where the fanless M4 MacBook Air beat an actively cooled 256V LNL laptop in real world cpu tests.