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.

 

[KompuKare]

This sounds about right:

https://twitter.com/x/status/1844037034094252402

Poor value, but I'm sure that ARL plus DDR5 9500 will win in some things. Probably not a lot of things though.

 

[AcrosTinus]

But why would the average gamer/consumer buy the $590 285K instead of the <$500 14.9KK?

They won't, soo I hope those leaks are false

 

[H433x0n]

But why would the average gamer/consumer buy the $590 285K instead of the <$500 14.9KK?

This just illustrates how dumb the power efficiency argument was for desktop PC.

Realistically, if they perform the same then there’s no point. Nobody is going to spend all of this extra money because ARL performs better in cinememe at 125W.

 

[Hitman928]

This is the point I'm making right here. What is the price difference. 142mm^2 of N4X is more expensive than ~90 mm^2 of N3B. What's the cost difference between N4X and N3B when equalizing for size?

AMD is using N4P, not N4X. I'll say this, the cost difference doesn't make up for the size difference, but it closes the gap significantly. Granted, I don't know the exact contract terms as large customers always get negotiated prices.

And? Intel in the earnings call was speaking of general CCG margins including LNL and ARL together, and specifically highlighted MoP as the biggest issue regarding margins this year.

LNL isn't even high volume. It's going to be higher than they originally planned, but compared to their other offerings, it is not high volume. Again, take the picture as a whole, Intel said margins towards the end of this year into next year will be low, putting the blame largely on LNL and ARL due to MoP for LNL and external sourcing for both. Additionally, they said they don't expect to see healthy margins again until they get back onto their own foundry.

I feel like this is being pedantic. Clearly the CCD's and SoC chiplet are bonded to a package that has the infinity fabric interconnects between the two. That silicon isn't free.

It's not pedantic because there is no silicon there for AMD, that's the point.

And those are? AMD's Q2 earnings show 6% operating margin in client., with embedded and datacenter buoying their overall margins. Intel client operating income in the same quarter was showing over 33%, dragged down by poor margins in DCAI and especially foundry. While these figures don't have ARL/LNL, they do have MTL, which should have similar margins to ARL.

Those are operating margins, not gross margins.

 

[Wolverine2349]

But why would the average gamer/consumer buy the $590 285K instead of the <$500 14.9KK?

Uhm maybe because the 14.9KK has stability and degradation issues and is a bad die/stepping very possibly/likely not fixed by microcode update. Its not trusted by many as such.

 

[gdansk]

In any case I think people are being too negative pre-emptively. We don't have reviews. There may be some cases where it is pretty good and that might make it worthwhile for some people. It might end up being a jack of all trades even if it isn't the master of gaming.

 

[Wolverine2349]

This just illustrates how dumb the power efficiency argument was for desktop PC.

Realistically, if they perform the same then there’s no point. Nobody is going to spend all of this extra money because ARL performs better in cinememe at 125W.

They will if they do not want so much excess heat dumped into their case and to get stability and reliability without the degradation that the 13th and 14th Gen stepping/die has.

As long as it is par or better for 99% of gaming workloads with reasonably priced RAM. That remains to be seen.

 

[gdansk]

Uhm maybe because the 14.9KK has stability and degradation issues and is a bad die/stepping not fixed by microcode update. Its not trusted by many as such.

Yeah, that's one reason. But I would personally suspect all new Intel products if I did care. They shipped a product that was over the limit for a process that they designed and should have known well. What stops them from shipping a part that goes over the limits of a process they didn't design?

Anyway, I got slightly off track

 

[Wolverine2349]

In any case I think people are being too negative pre-emptively. We don't have reviews. There may be some cases where it is pretty good and that might make it worthwhile for some people. It might end up being a jack of all trades even if it isn't the master of gaming.

Yes lets wait for reviews first. Cannot wait. It has to be good at gaming. Not jack of all trades good, but on par with at least well tuned 13th and 14th Gen with reasonably priced/speed RAM.

Can trade blows with 7800X3D or slight behind but not by much but has to slam Zen 4 and 5 vanilla. That way you get more cores on one die and on par gaming performance and more future proof.

 

[Hitman928]

In any case I think people are being too negative pre-emptively. We don't have reviews. There may be some cases where it is pretty good and that might make it worthwhile for some people. It might end up being a jack of all trades even if it isn't the master of gaming.

Just like with Zen 5, this isn’t bulldozer/P4 situation. I’m sure ARL is a solid CPU and I am very glad that the efficiency is much improved, things were getting really out of hand. I actually wish they went even lower with the power limit, but at least we got things down from last gen. The main problem with this audience and the diy crowd in general, is that gaming is the biggest concern, so when the latest gen shows little improvement, or even seemingly a regression for ARL, the pitchforks come out, even if it ends up being strong in other areas.

 

[inquiss]

Why would they weaken AMD by helping intel, who are clear they will reduce capacity in the future. That's not good long term thinking.

 

[Wolverine2349]

Yeah, that's one reason. But I would personally suspect all new Intel products if I did care. They shipped a product that was over the limit for a process that they designed and should have known well. What stops them from shipping a part that goes over the limits of a process they didn't design?

Anyway, I got slightly off track

That is true it is hard to trust Intel products.

Though Raptor Lake was more than just pushed over the limit. There is a reason spec boards chips failed and had stability problems as well.

And Intel admitted (Google it) Raptor Lake was a rushed design so not surprising it has so many problems.

Intel stuff has been pretty reliable prior to Raptor Lake including Alder Lake. I think they will be ok with Arrow Lake. But if choosing to go intel of course Arrow Lake over Raptor Lake given what is known about Raptor Lake.

Otherwise go Intel 12th Gen and earlier or AMD if weary about intel products in general. But avoid 13th and 14th Gen 8 +16 stepping/die like the plague regardless of microcode update.

 

[Doug S]

TSMC N3B is very expensive for Intel. They commented on it multiple times in their latest earnings call. It is one of the key factors they identified for low gross margin forecasts for the end of the year and into 2025.

Of course its expensive. All the profit TSMC is making off Intel would instead accrue to Intel's bottom line if they were making the chips themselves at the same cost basis as TSMC (whether they get there or not is another matter but they'll have to get there eventually if they want their foundry to be competitive outside of the "must be made in the USA" market)

 

[gdansk]

Just like with Zen 5, this isn’t bulldozer/P4 situation. I’m sure ARL is a solid CPU and I am very glad that the efficiency is much improved, things were getting really out of hand. I actually wish they went even lower with the power limit, but at least we got things down from last gen. The main problem with this audience and the diy crowd in general, is that gaming is the biggest concern, so when the latest gen shows little improvement, or even seemingly a regression for ARL, the pitchforks come out, even if it ends up being strong in other areas.

I get the pitchforks to some extent because I'm about 300MHz disappointed with both Zen 5 and ARL based on the tests I have seen so far. And I'd like to see them stay competitive with ARM while it seems they are falling behind.

It's okay to be disappointed but I think people should wait for reviews. Let's see what they were able to improve and get a balanced picture instead of whatever Intel's slides presented.

 

[Wolverine2349]

Just like with Zen 5, this isn’t bulldozer/P4 situation. I’m sure ARL is a solid CPU and I am very glad that the efficiency is much improved, things were getting really out of hand. I actually wish they went even lower with the power limit, but at least we got things down from last gen. The main problem with this audience and the diy crowd in general, is that gaming is the biggest concern, so when the latest gen shows little improvement, or even seemingly a regression for ARL, the pitchforks come out, even if it ends up being strong in other areas.

Yes that is true. However if gaming performance does in fact regress, that is a difference from being on par.versus regressing.

Vanilla Zen 5 did not regress gaming over vanilla Zen 4, it just did not improve it. Itr was on par or 1-2% better.

If Arrow Lake is on par with Raptor Lake gaming and is stable and has lower power usage it iwll be welcome more so than Zen 5.

Zen 5 hardly had any improvement for gaming, and was no more power efficient and was the same as Zen 4 in power efficiency.

Now if gaming regressed especially too much, that will be bad and make people think P4 Bulldozer situation.

 

[Abwx]

Vanilla Zen 5 did not regress gaming over vanilla Zen 4, it just did not improve it. Itr was on par or 1-2% better.

On day 1 review, before the Windows update and new agesa :

Ryzen 9 9950X & 9900X: Gaming-Benchmarks: Spiele-Benchmarks auf zwei verschiedenen Grafikkarten

Ryzen 9 9950X & 9900X im Test: Spiele-Benchmarks auf zwei verschiedenen Grafikkarten / Gaming-Benchmarks mit einer GeForce RTX 4090

www.computerbase.de

 

[Josh128]

I get the pitchforks to some extent because I'm about 300MHz disappointed with both Zen 5 and ARL based on the tests I have seen so far. And I'd like to see them stay competitive with ARM while it seems they are falling behind.

It's okay to be disappointed but I think people should wait for reviews. Let's see what they were able to improve and get a balanced picture instead of whatever Intel's slides presented.

Has anyone been able to measure CPU fps on something like a MBP compared to a 9700X or 14700K or something. I have a feeling the x86 chips would roast the Apple chip in fps in modern games.

 

[H433x0n]

If Arrow Lake is on par with Raptor Lake gaming and is stable and has lower power usage it iwll be welcome more so than Zen 5.

Tech press is gaming focused, the rest really doesn’t matter as far as reviews are concerned.

I’m sure in efficiency and productivity it’ll be fine, but so was Zen 5 and it was excoriated in reviews.

 

[Markfw]

Tech press is gaming focused, the rest really doesn’t matter as far as reviews are concerned.

I’m sure in efficiency and productivity it’ll be fine, but so was Zen 5 and it was excoriated in reviews.

This is why independant sites that we respect are the only ones that I trust.

 

[poke01]

Has anyone been able to measure CPU fps on something like a MBP compared to a 9700X or 14700K or something. I have a feeling the x86 chips would roast the Apple chip in fps in modern games.

For gaming you shouldn’t consider anything ARM. Especially MacBooks.

X86 will always reign supreme there, mainly due to the shear x86 gaming PC and consoles.

 

[poke01]

The LGA 1851 platform has great I/O. Thunderbolt 5 so 80gbps and dual 10G LAN. For content creators and 3D folks this will be a great platform.

 

[jdubs03]

https://twitter.com/x/status/1844018875706531948

The LGA 1851 platform has great I/O. Thunderbolt 5 so 80gbps and dual 10G LAN. For content creators and 3D folks this will be a great platform.

Shame that it’s only useful for Arrow Lake and nothing else.

 

[Hulk]

As I have been lamenting for over a year now, Intel hemmed themself into a corner with Raptor Lake Refresh. Raptor Lake was itself a refresh itself of Alder Lake with increase I cores and frequency. Raptor Refresh brought even more frequency, some of which turned out to be "too much."

It looks like Raptor Lake Refresh was the last of the huge monolithic, high clock, desktop CPU's. Having pushed clocks up to 6GHz and beyond with the KS version along with the latency advantage of the monolithic design, Intel was looking at at tough challenge as they moved to a new process with undoubtedly lower clocks along with a tile penality.

We saw much the same thing with Meteor Lake with Redwood Cove and it's minor modifications supposedly "holding even" IPC against Raptor Cove. "Holding even" is perhaps kind of generous actually. It took the second iteration with Lunar Lake for the P and E cores to make meaningful performance improvements from Raptor and Meteor Mobile.

From what we know so far I think we are looking at a similar "reset" with Arrow Lake. Lion Cove looks to provide some IPC increase of Raptor Cove even after the tile latency and clock deficits. Overall P core performance thought, might now show much if any increase in some scenarios. Good on Intel for at least moving backward, not so good on not really moving foward.

Skymont as many around here have noted look to be the saving grace for Arrow Lake as it is showing more than significant increases over Gracemont and this will surely make Arrow Lake a good MT performer.

At the end of the day I think Arrow Lake will be an upgrade from Raptor Refresh from a performance and efficiency point-of-view. In addition, I think Zen 5 and Arrow Lake will trade blows and each will find homes based on the apps they excel for their respective owners.

Arrow Lake and Zen 5 look to be quite strong products, as well as the amazingly efficient Apples M chips. They perform quite the magic act when it comes to performance and efficiency, no denying that.

I'm excited for ARL's release as I love to study up on the details and performance metrics as I do with each new CPU generation release.

 

[511]

Arguing BoM of Zen 5 vs ARL is pointless without talking specifics. Until you give hard numbers on costs and margins, it's all hearsay.

A 9950X uses 141mm^2 of N4P for its CCDs.
Someone else here is claiming ARL-S 8+16's compute tile is 88mm^2 of N3B.

N4P is what, $20K per wafer?

N3B is not 46% more expensive than N4P, so if these die sizes are true, then we at least know that ARL's compute tile is cheaper than Zen 5's dual CCDs.

Someone else can do the math on packaging and the cost structure of the remaining tiles, but I'm really not seeing any evidence that ARL is "substantially" more expensive to produce.

Let the die size leak and we will figure the pricing out roughly for zen 5 we have the data

 

[adroc_thurston]

N4P is what, $20K per wafer?

Less.