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.

 

[Henry swagger]

Royal core will have 4 or 8 way smt it will be big.. jim keller influenced

 

[DrMrLordX]

you did complain a good bit mate.

Do not mistake irritation or outrage for ennui.

In any case

Don't believe anything about Intel until you see it. Intel has shown off the 6+8 package in mockups and elsewhere. They've made quite a big deal about it. Until we see more on Meteor Lake it's not guaranteed that Intel is selling any 6+16 parts or anything of the sort. And that configuration would be ridiculous anyway. They'd be better off with 12+16 using existing CPU chiplets if they're going that route.

Royal core will have 4 or 8 way smt it will be big.. jim keller influenced

That worked so well for POWER. Can't wait to see Intel do it!

(that's a joke)

 

[Exist50]

There's also Panther cove after Lion so it seems Royal is at least 3 generations away

Unless we get a big core update concurrently with Royal, which I think is a distinct possibility.

 

[Exist50]

They'd be better off with 12+16 using existing CPU chiplets if they're going that route.

MTL's SoC doesn't support multiple compute dies, so that option is off the table.

 

[DrMrLordX]

MTL's SoC doesn't support multiple compute dies, so that option is off the table.

Really? Why not?

 

[mikk]

There's also Panther cove after Lion so it seems Royal is at least 3 generations away

In the reddit leak Panther Cove sounded like it is Royal Core project.

Nova Lake (Panther Cove [tentative]/ Darkmont) 2025 - This will mark the biggest architectural change in CPU architecture since the Core architecture is introduced in 2006. Intel is working to build entirely new architecture from the ground up much like Ryzen with up to 50% CPU performance improvement from the lunar lake. This is also the reason why Glenn Hinton returned.

There are rumors about Panther Lake after Arrow Lake, Panther Lake and Panther Cove would make sense though.

 

[jpiniero]

Quite the opposite, the SKU lineup and earlier rumors from Raichu and kopite7kimi suggests it's the same generation.

TSMC said that N3 is only going to be 4-6% of revenue next year (the wording TSMC used was "mid single digit"). That doesn't seem like a lot, esp when you factor in how expensive the wafers must be. Arrow Lake would not be a small order by any means so it doesn't seem like it would be fabbed in 2023. Timing doesn't appear to work out for 14th Gen.

 

[mikk]

TSMC said that N3 is only going to be 4-6% of revenue next year (the wording TSMC used was "mid single digit"). That doesn't seem like a lot, esp when you factor in how expensive the wafers must be. Arrow Lake would not be a small order by any means so it doesn't seem like it would be fabbed in 2023. Timing doesn't appear to work out for 14th Gen.

Arrow Lake is a 2024 generation not 2023 and ARL-S will be limited to i7 and i9 by the looks of it.

 

[dangerman1337]

*One* Possibility I just thought today is that *maybe* ARL-S on TSMC *N3E* is potentially very late next year along with MTL-S on Desktop making up 14th gen and then 20A ARL-S+ in late 2024?

I just find the idea that Intel won't do a 8P Core SKU next year that isn't only Raptor Lake a bit weird.

 

[jpiniero]

Arrow Lake is a 2024 generation not 2023 and ARL-S will be limited to i7 and i9 by the looks of it.

So you agree with me then. 14th Gen would be 23/24 products. 15th Gen would be 24/25 products. You would be talking about chiplets that would be delivered to Intel in say the middle of 24.

 

[mikk]

So you agree with me then. 14th Gen would be 23/24 products. 15th Gen would be 24/25 products. You would be talking about chiplets that would be delivered to Intel in say the middle of 24.

No not really, I'm expecting Intel brings MTL-S and ARL-S in a 14th generation lineup during 2024 and more likely H1 2024. The leaked SKUs somehow confirms it. We could argue whether they will launch exactly at the same time or they might launch MTL and ARL 1-2 quarters separated. In this case they could launch MTL-S closer to early H1 and ARL-S closer to early H2.

 

[dullard]

So Meteor Lake and Arrow Lake will have different P cores and are supposed to launch at the same time? if that's the case it makes little sense to me (or I misunderstood something).

1) Intel has been saying for the last 3 years that they are no longer putting all of their eggs in one basket. That is because they were burned quite badly when they failed on executing that one basket. Instead, Intel is going in on multiple routes for many of their products. https://images.anandtech.com/doci/15217/20190916 SPIE Photomask and EUVL Plenary - Phillips v23-VRL2 distribute-page-019.jpg

2) Meteor Lake and Arrow Lake are not launching at the same time. But, there will be significant overlap.

3) Intel does not yet have enough EUV capability to do a whole product lineup launch on Meteor Lake. Focusing on smaller chips for Meteor Lake and waiting a few months for the larger chips on Arrow Lake helps solve that problem.

4) Meteor Lake has a lot of changes. Changes in design, major packaging changes, new features, etc. A small launch on Meteor Lake gives them time to work out a lot of the kinks before Arrow Lake comes out.

 

[jpiniero]

launch MTL-S closer to early H1 and ARL-S closer to early H2.

That's what I mean about the timing. That's too long of a gap between Raptor Lake and 2H 24. There would have to be a product released inbetween the two and that product would be the 14th Gen.

 

[trexfromouterspace]

There's also Panther cove after Lion so it seems Royal is at least 3 generations away

Not relevant, it's different teams. Intel will use whichever one is ready and/or better.

 

[igor_kavinski]

There would have to be a product released inbetween the two and that product would be the 14th Gen.

What if 14th gen is a slightly tweaked Raptor Lake on Intel 4? Having a new architecture on a new unproven process is too much of a risk, no?

 

[eek2121]

1) Intel has been saying for the last 3 years that they are no longer putting all of their eggs in one basket. That is because they were burned quite badly when they failed on executing that one basket. Instead, Intel is going in on multiple routes for many of their products. https://images.anandtech.com/doci/15217/20190916 SPIE Photomask and EUVL Plenary - Phillips v23-VRL2 distribute-page-019.jpg

2) Meteor Lake and Arrow Lake are not launching at the same time. But, there will be significant overlap.

3) Intel does not yet have enough EUV capability to do a whole product lineup launch on Meteor Lake. Focusing on smaller chips for Meteor Lake and waiting a few months for the larger chips on Arrow Lake helps solve that problem.

4) Meteor Lake has a lot of changes. Changes in design, major packaging changes, new features, etc. A small launch on Meteor Lake gives them time to work out a lot of the kinks before Arrow Lake comes out.

Intel has plenty of EUV capacity.

The last credible rumor I read:

Meteor Lake is on Intel 4. The GPU will be on a TSMC node.

Repeat with Arrow Lake and Intel 3. I doubt we will see client designs on an external node any time soon.

I also doubt Arrow Lake will be any part of 14th gen. I strongly suspect MTL-S will launch around this time next year, and Arrow Lake S will launch as 15th gen around the same time in 2024.

As for the 6+16 rumor (vs 8+16 rather than 6+8), it makes sense to me, though it is surprising. Reduced ring stops and a smaller die. It is a win/win. The rumor could be wrong though.

I could be wrong, of course, but I suspect I will be more right than wrong. We will see.

 

[Exist50]

Really? Why not?

My understanding is that the die to die interfaces are custom for each one. But even if that wasn't the case, a 6+8 CPU chiplet has radically different power rail and physical size requirements vs a small graphics one. Really doubt the socket would be flexible enough to accomodate either.

In the reddit leak Panther Cove sounded like it is Royal Core project.

The leaker probably got confused. Panther Cove is not Royal either. Which would probably mean we don't see it till Nova Lake.

 

[lightisgood]

I'm expecting Intel brings MTL-S and ARL-S in a 14th generation lineup during 2024 and more likely H1 2024.

Pat said "On Intel 4, we are progressing towards high-volume-manufacturing and will tape out the
production stepping of Meteor Lake in Q4".


"production stepping" is probably QS (quality sample).
This means MTL launching is set at 3Q23 ~ end of 23.

I guess that MTL launch at Sep. or Oct. next year.
MTL-P/M might be prioritized for first semester.


[post script]
sorry... you said about MTL-'S'.
Ignore my reply, please.
[end of script]

 

[DrMrLordX]

My understanding is that the die to die interfaces are custom for each one. But even if that wasn't the case, a 6+8 CPU chiplet has radically different power rail and physical size requirements vs a small graphics one. Really doubt the socket would be flexible enough to accomodate either.

Hmm. The 6+8 mockups we've seen to date show that the largest chiplet is the SoC, and that the iGPU and compute chiplets are quite a bit smaller.

 

[Exist50]

Hmm. The 6+8 mockups we've seen to date show that the largest chiplet is the SoC, and that the iGPU and compute chiplets are quite a bit smaller.

Sure, but the GPU chiplet is still way smaller (and certainly much shorter) than the CPU one. And that doesn't change the fact that they'd use different power rails.

 

[DrMrLordX]

Sure, but the GPU chiplet is still way smaller (and certainly much shorter) than the CPU one. And that doesn't change the fact that they'd use different power rails.

Well right, the board would require a pretty significantly complicated VRM to handle two compute chiplets. Unless they're not meant to run high-power as they are. That is unless Intel uses some kind of IVR.

 

[inf64]

1) Intel has been saying for the last 3 years that they are no longer putting all of their eggs in one basket. That is because they were burned quite badly when they failed on executing that one basket. Instead, Intel is going in on multiple routes for many of their products. https://images.anandtech.com/doci/15217/20190916 SPIE Photomask and EUVL Plenary - Phillips v23-VRL2 distribute-page-019.jpg

2) Meteor Lake and Arrow Lake are not launching at the same time. But, there will be significant overlap.

3) Intel does not yet have enough EUV capability to do a whole product lineup launch on Meteor Lake. Focusing on smaller chips for Meteor Lake and waiting a few months for the larger chips on Arrow Lake helps solve that problem.

4) Meteor Lake has a lot of changes. Changes in design, major packaging changes, new features, etc. A small launch on Meteor Lake gives them time to work out a lot of the kinks before Arrow Lake comes out.

Thanks. Is Arrow Lake going to use the same P core uarchitecture as Meteor Lake but on a different node?

 

[lightisgood]

Is Arrow Lake going to use the same P core uarchitecture as Meteor Lake but on a different node?

Maybe, No.

Ribbon FET and Power Via, Intel 20A, bring significant higher transistor budget and GHz headroom.
Lion's goal should be set more IPC or more GHz, Lion's uarch will inevitably change.

 

[Timmah!]

but but MLID told me that Arrow Lake will have 32 e cores....

No to defend MLID, but i dont think WCCFtech is more reliable in this particular case.

That said, if this is true, at least i know i wont be interested in Intel mainstream platform for another couple of years. If i wanted 8p +16e CPU, i would get RPL now.

 

[jpiniero]

No to defend MLID, but i dont think WCCFtech is more reliable in this particular case.

WTFTech will post anything they get their hands on. Some of it might be legit.

The Youtubers in general just make up stuff.