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.

 

[AcrosTinus]

This is not correct. Statistically, many consumers are not starved, but there are quite a few like me, and many others who really could use threadripper in an affordable package, but can not get.

I added "average" consumer, no one needing a threadripper is a average consumer

 

[511]

Not that I have been following this as closely as I used to, but a little research dug up this:

AMD Turin - 192 cores Zen 5c and Zen 5 variants (I think) - release Q4 2024
Intel Sierra Forest - 144 E cores - Release June 2024
Intel Sierra Forest - 288 E cores - Release Q4 2024 (Intel hasn't been very good about meeting their server roadmap lately though)
Intel Granite Rapids - 2025

So Zen 5 will go against GNR and Turin dense vs SRF but we have Clearwater Forest as well to complete 288 stronger Chadmont cores launching Q3

 

[fastandfurious6]

Big Win for Intel

Intel and AWS Expand Strategic Collaboration, Helping Advance U.S.-Based Chip Manufacturing

www.intc.com

Amazon is one of the worst customers they can have

 

[511]

Amazon is one of the worst customers they can have

Why lmao anyway intel won't care as long as they get orders for their Fab
It is all about $$ universal truth

 

[DrMrLordX]

AVX-512 can accelerate a ton of workloads, even the web browser. Hopefully someone will drop a build of Firefox that supports it.

I would love to see someone do a Firefox fork (Palemoon and Waterfox, where are you?) with an eye towards SIMD optimizations. It took years to even get Firefox to support 64bit mode, so I'm not holding out much hope that they'll optimize for performance in that fashion.

I hope all you folks are doing well. I dropped by briefly to see what is new. I have some medical issues going on, so I haven’t been on the forum very much lately.

Doing well enough, thanks. Keep up with us, the main site shut down but the forums are still going strong. For now!

 

[OneEng2]

So Zen 5 will go against GNR and Turin dense vs SRF but we have Clearwater Forest as well to complete 288 stronger Chadmont cores launching Q3

Everything I can find says 2025.... but it seems unlikely since we are now seeing Intel push back the server processors that they originally said would be released in 2024 to 2025. Also, this processor was supposed to use GAA and BPD. That seems like a lot of new stuff in a very short time window to me.

Perhaps I am being a bit pessimistic on Intel. I have had the shine rubbed off of that penny since the 10nm debacle that drug on forever. Then we had Intel talking about 5 die shrinks in 4 years ..... and now they are moving production to TSMC. It has the appearance that Intel has lost its lead in process technology, and isn't getting it back any time soon. Of course, I thought P4 was going to hurt them much more than it did. Turned out just to be a bump in the road.... but that was just a bad processor design. They still had an undisputed process advantage then.

My contention (guess) is that things will be quite a bit different for Intel without a huge process advantage. My current comparison is that Arrow Lake is competitive performance wise with Zen 5, but it does so at the cost of a more expensive process. This shows up in Intel's bottom line.

Now, I have been doing this way too long to say anything silly like "Intel is doomed". They are a power to be reconned with for sure. I am more thinking that Intel will have much more trouble than they did in the past as the landscape has changed for them in a more fundamental way.

Looking back at the server roadmaps for Intel, it seems like it has been moving back quite a bit. Perhaps this is true for AMD as well, but then, AMD managed to gain a huge upper hand in this segment and Intel (again) is playing catch-up. Can Intel actually "catch up" without a 1 to 2 generation die shrink process advantage? Can Intel catch up in AI processing? Seems like they have managed to get caught with their pants down in a number of computing area technologies, and now they have to "catch up" without the endless supply of cash they once enjoyed at a time when they are cutting back on resources as well.

I'm not counting them out by any means though. They have a way of turning things around in inventive ways (technological, political, and market wise). I would say that it seems like they have more "turning around" to do than they have had in the past though.

 

[9949asd]

Everything I can find says 2025.... but it seems unlikely since we are now seeing Intel push back the server processors that they originally said would be released in 2024 to 2025. Also, this processor was supposed to use GAA and BPD. That seems like a lot of new stuff in a very short time window to me.

Perhaps I am being a bit pessimistic on Intel. I have had the shine rubbed off of that penny since the 10nm debacle that drug on forever. Then we had Intel talking about 5 die shrinks in 4 years ..... and now they are moving production to TSMC. It has the appearance that Intel has lost its lead in process technology, and isn't getting it back any time soon. Of course, I thought P4 was going to hurt them much more than it did. Turned out just to be a bump in the road.... but that was just a bad processor design. They still had an undisputed process advantage then.

My contention (guess) is that things will be quite a bit different for Intel without a huge process advantage. My current comparison is that Arrow Lake is competitive performance wise with Zen 5, but it does so at the cost of a more expensive process. This shows up in Intel's bottom line.

Now, I have been doing this way too long to say anything silly like "Intel is doomed". They are a power to be reconned with for sure. I am more thinking that Intel will have much more trouble than they did in the past as the landscape has changed for them in a more fundamental way.

Looking back at the server roadmaps for Intel, it seems like it has been moving back quite a bit. Perhaps this is true for AMD as well, but then, AMD managed to gain a huge upper hand in this segment and Intel (again) is playing catch-up. Can Intel actually "catch up" without a 1 to 2 generation die shrink process advantage? Can Intel catch up in AI processing? Seems like they have managed to get caught with their pants down in a number of computing area technologies, and now they have to "catch up" without the endless supply of cash they once enjoyed at a time when they are cutting back on resources as well.

I'm not counting them out by any means though. They have a way of turning things around in inventive ways (technological, political, and market wise). I would say that it seems like they have more "turning around" to do than they have had in the past though.

Intel will make the clearwater forest using n3e in 2025. that’s the first server chip to compete with amd.

 

[cebri1]

Intel will make the clearwater forest using n3e in 2025. that’s the first server chip to compete with amd.

No! CFW is 18A. And current solutions already compete with AMD.

 

[OneEng2]

Intel will make the clearwater forest using n3e in 2025. that’s the first server chip to compete with amd.

So both Turnin dense and Clearwater Forest will be using N3E. Should be an interesting exercise and comparison. First time in my memory that an Intel and AMD chip were both produced on the same process.

 

[511]

Intel will make the clearwater forest using n3e in 2025. that’s the first server chip to compete with amd.

It is fully in-house they have said it in the white paper lol

https://www.intel.com/content/dam/www/central-libraries/us/en/documents/2024-02/intel-tech-clearwater-wp.pdf

Same as GNR/SRF lol i bet your source for N3E is MLID 😂🤣

 

[511]

So both Turnin dense and Clearwater Forest will be using N3E. Should be an interesting exercise and comparison. First time in my memory that an Intel and AMD chip were both produced on the same process.

Nope N3E is for GPU FLS and Xe3

 

[OneEng2]

No! CFW is 18A. And current solutions already compete with AMD.

Then Intel has a great deal riding on 18A.

New core, BSPD, GAA, Die shrink. Seems like a lot between now and a year from now. Not sure I would bet my 401K on that one .

 

[511]

Then Intel has a great deal riding on 18A.

New core, BSPD, GAA, Die shrink. Seems like a lot between now and a year from now. Not sure I would bet my 401K on that one .

You won't but intel has already betted everything on 18A 🤣

 

[511]

Everything I can find says 2025.... but it seems unlikely since we are now seeing Intel push back the server processors that they originally said would be released in 2024 to 2025. Also, this processor was supposed to use GAA and BPD. That seems like a lot of new stuff in a very short time window to me.

When was the pushback EMR/SRF were on time SRF is quite good for it's market

Perhaps I am being a bit pessimistic on Intel. I have had the shine rubbed off of that penny since the 10nm debacle that drug on forever. Then we had Intel talking about 5 die shrinks in 4 years ..... and now they are moving production to TSMC. It has the appearance that Intel has lost its lead in process technology, and isn't getting it back any time soon. Of course, I thought P4 was going to hurt them much more than it did. Turned out just to be a bump in the road.... but that was just a bad processor design. They still had an undisputed process advantage then.

No comments on 10nm it is/was a disaster but at least they got their senses back and are dual sourcing

My contention (guess) is that things will be quite a bit different for Intel without a huge process advantage. My current comparison is that Arrow Lake is competitive performance wise with Zen 5, but it does so at the cost of a more expensive process. This shows up in Intel's bottom line.

Agreed for the cost no one has idea unless someone measures everything but for volume it will be Intel

Now, I have been doing this way too long to say anything silly like "Intel is doomed". They are a power to be reconned with for sure. I am more thinking that Intel will have much more trouble than they did in the past as the landscape has changed for them in a more fundamental way.

Yes

Looking back at the server roadmaps for Intel, it seems like it has been moving back quite a bit. Perhaps this is true for AMD as well, but then, AMD managed to gain a huge upper hand in this segment and Intel (again) is playing catch-up. Can Intel actually "catch up" without a 1 to 2 generation die shrink process advantage? Can Intel catch up in AI processing? Seems like they have managed to get caught with their pants down in a number of computing area technologies, and now they have to "catch up" without the endless supply of cash they once enjoyed at a time when they are cutting back on resources as well.

Depends on scenarios Intel still has huge SW to complement their processor also "AI" Performance is still better on Intel

I'm not counting them out by any means though. They have a way of turning things around in inventive ways (technological, political, and market wise). I would say that it seems like they have more "turning around" to do than they have had in the past though.

Yes It's how Andy Grove turned Intel from memory to cpu company btw that guy was nuts when he was changing buisness strategy he went to the guys and the guys wo didn't agreed with him on his strategy were fired on site

 

[cebri1]

Then Intel has a great deal riding on 18A.

New core, BSPD, GAA, Die shrink. Seems like a lot between now and a year from now. Not sure I would bet my 401K on that one .

☝️

 

[9949asd]

It is fully in-house they have said it in the white paper lol

https://www.intel.com/content/dam/www/central-libraries/us/en/documents/2024-02/intel-tech-clearwater-wp.pdf

Same as GNR/SRF lol i bet your source for N3E is MLID 😂🤣

No, an intel employee said that. Who the heck will watch MLID.

 

[Hitman928]

No, an intel employee said that. Who the heck will watch MLID.

Granted, Intel pulled a bait and switch with ARL and 20a, but we had lots of leaks for a while telling us that ARL was actually an N3B product outside a small volume of 20a chips. There have been no such leaks that I've seen for 18a/CWF. Assuming, for the moment, that an actual Intel employee said that CWF is really on N3E, are you sure he'd even be in a position to know? What is the explanation for why Intel is publicly saying that CWF has taped out on 18a?

 

[511]

Are you sure cause you can't dual source this time unless you have the design ready anyway i am not buying this N3E Clearwater Rumour cause this is from official presentation at a conference

 

[511]

We have reach a stage where official info is not official according to some people bruh you can't lie on such a large scale for Intel 18A They have been selling it to investor as their lead product stop spreading lies on things that are confirmed

 

[DavidC1]

Then Intel has a great deal riding on 18A.

New core, BSPD, GAA, Die shrink. Seems like a lot between now and a year from now. Not sure I would bet my 401K on that one .

Clearwater Forest is not a new core. The core is going to be used in both Lunarlake and Arrowlake. Clearwater Forest's core is a small update.

Also it's redundant to say BSPD, GAA, and Die Shrink. They are all part of the new process technology.

 

[511]

Clearwater Forest is not a new core. The core is going to be used in both Lunarlake and Arrowlake. Clearwater Forest's core is a small update.

Also it's redundant to say BSPD, GAA, and Die Shrink. They are all part of the new process technology.

It is a updated core on entirely new node with GAA and BSPD so i would call it new otherwise crestmont wouldn't be a new core ?

 

[jdubs03]

Remember it’s supposed to be Darkmont (revised Skymont) that will be in CWF.

 

[naukkis]

Intel is about to go out of business because their fab has failed to execute time after time. If you were head of Intel do you still make important products to not yet production ready internal process without making back-up plan - dual source your products? I think that now every Intel product is dual sourced and tsmc version will be released on time if internal fab fails - which is more than likely.

 

[9949asd]

We have reach a stage where official info is not official according to some people bruh you can't lie on such a large scale for Intel 18A They have been selling it to investor as their lead product stop spreading lies on things that are confirmed

Just wait till intel to announce🤪they just did changed the ARL u5 u3 from 20a to full n3b.

 

[9949asd]

Just wait till intel to announce🤪they just did changed the ARL u5 u3 from 20a to full n3b.

And I’m really think intel use tmsc in the next 3 years will be the only way to make things work.