Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads

[Tigerick]

Wildcat Lake (WCL) Specs

Intel Wildcat Lake (WCL) is upcoming mobile SoC replacing Raptor Lake-U. 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 Q1 2026.

		Intel Raptor Lake U	Intel Wildcat Lake 15W?	Intel Lunar Lake		Intel Panther Lake 4+0+4
Launch Date		Q1-2024	Q2-2026	Q3-2024		Q1-2026
Model		Intel 150U	Intel Core 7	Core Ultra 7 268V		Core Ultra 7 365
Dies		2	2	2		3
Node		Intel 7 + ?	Intel 18-A + TSMC N6	TSMC N3B + N6		Intel 18-A + Intel 3 + TSMC N6
CPU		2 P-core + 8 E-cores	2 P-core + 4 LP E-cores	4 P-core + 4 LP E-cores		4 P-core + 4 LP E-cores
Threads		12	6	8		8
Max Clock		5.4 GHz	?	5 GHz		4.8 GHz
L3 Cache		12 MB		12 MB		12 MB
TDP		15 - 55 W	15 W ?	17 - 37 W		25 - 55 W
Memory		128-bit LPDDR5-5200	64-bit LPDDR5	128-bit LPDDR5x-8533		128-bit LPDDR5x-7467
Size		96 GB		32 GB		128 GB
Bandwidth				136 GB/s
GPU		Intel Graphics	Intel Graphics	Arc 140V		Intel Graphics
RT		No	No	YES		YES
EU / Xe		96 EU	2 Xe	8 Xe		4 Xe
Max Clock		1.3 GHz	?	2 GHz		2.5 GHz
NPU		GNA 3.0	18 TOPS	48 TOPS		49 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.

 

[Covfefe]

I have a theory that if you add everything needed to bring the performance of an E Core up to the level of a P core (or a Zen 6 core) you pretty much end up with a P core don't you?

No, not if the new CPU design is better than the old one. Your whole argument seems to hinge on the idea that all CPU cores are the same, and it's impossible to create a good one or a bad one.

Lion Cove is terrible. It has much worse PPA than Zen5. I don't know why you lump the p-core and Zen6 together. If a Skymont successor took over and simply matched the competing AMD CPU core, that would be a huge step up from the current situation.

 

[OneEng2]

No, not if the new CPU design is better than the old one. Your whole argument seems to hinge on the idea that all CPU cores are the same, and it's impossible to create a good one or a bad one.

Lion Cove is terrible. It has much worse PPA than Zen5. I don't know why you lump the p-core and Zen6 together. If a Skymont successor took over and simply matched the competing AMD CPU core, that would be a huge step up from the current situation.

It's an interesting concept I suppose.

I guess I believe that in this day and age, there aren't any mysterious CPU architectures that magically work better than everything that came before it.

I believe that Apple, Intel, and AMD all have equivalent engineering teams and tools. The difference is what you target your architecture to do and what things you decide to prioritize and what things you decide to give up.

I believe you can't say I want it all and actually get it all. If you say I want a core that is very power efficient, you can't also say I want a core that clocks higher than the competition.

You can't say I want the core to be very small AND I want 4 way SMT, AVX512, etc, etc.

I do agree that Lion Cove appears to have lower PPA than Zen 5, although I think ARL in general gets a pretty bad rap on the basis of its poor showing in latency sensitive applications (which is mostly a ring bus issue IMO vs a core problem).

I think it is a pretty tall order to take ANY derivative of Skymont and make it compete with Zen 5 across the board. I think you can make it do some things better, but at the expense of doing other things worse.

I just don't see getting something for nothing in engineering.

 

[511]

https://twitter.com/x/status/1986086104286597269

 

[Josh128]

https://twitter.com/x/status/1986086104286597269

5.1GHz for unobtainium SKUs only? 18A not looking great boys.

 

[dullard]

5.1GHz for unobtainium SKUs only? 18A not looking great boys.

Lets put both important pieces in the picture: turbo frequency AND turbo power levels used to reach that frequency (PL2).

• Alder Lake: 12900H 5.0 GHz top turbo, 115 W PL2
• Meteor Lake: 185H 5.1 GHz top turbo, 115 W PL2
• Lunar Lake: 288V 5.1 GHz top turbo, 37 W PL2
• Arrow Lake: 285H 5.4 GHz top turbo, 115 W PL2
• Panther Lake rumor: 388H 5.1 GHz top turbo, 45 W PL2 https://wccftech.com/intel-panther-...eak-core-ultra-300-16-cores-12-xe3-gpu-cores/

So, roughly the same top turbo speed with much lower power (other than lunar lake which fewer cores). What would a top frequency oriented chip get at 115 W?

 

[511]

5.1GHz for unobtainium SKUs only? 18A not looking great boys.

let the chip come out bruh

 

[gdansk]

let the chip come out bruh

I was told the chips were shipping for revenue. I think they're open for criticism then.

 

[dullard]

I was told the chips were shipping for revenue. I think they're open for criticism then.

Seems like you (and others) are looking to criticize without performance or price information.

 

[511]

I was told the chips were shipping for revenue. I think they're open for criticism then.

Yeah they are but to OEMs

 

[gdansk]

Seems like you (and others) are looking to criticize without performance or price information.

And you think a fmax regression is unworthy of criticism? Or that it simply isn't real?

 

[511]

And you think a clock speed regression is unworthy of criticism? Or that it simply isn't real?

depends on the comparison if you are comparing to LNL there is no clock regression but if you are to Arrow than their is

 

[dullard]

And you think a clock speed regression is unworthy of criticism? Or that it simply isn't real?

How long is it realistically at the top speed (tau)? What power is used for 1 core vs all core top speed? What performance do we get at that top speed (that is clock speed without IPC is useless)? None of those pieces of information are known. So, how can anyone honestly criticize? Panther Lake may be total crap--but you don't know yet. Somehow you think it is worthy of criticism though.

As for is it real or not, the only Intel mobile chip with higher clock speed requires 2x more power for 300 MHz more. And that is only a couple of Intel mobile chips that do it. Most top out at the exact same 5.1 GHz.

 

[gdansk]

How long is it realistically at the top speed (tau)? What power is used for 1 core vs all core top speed? What performance do we get at that top speed (that is clock speed without IPC is useless)? None of those pieces of information are known. So, how can anyone honestly criticize? Panther Lake may be total crap--but you don't know yet.

I didn't say it was crap. Simply that it is open to criticism because fmax dropped and it is shipping (i.e. that's where the specs come from now).

 

[511]

https://twitter.com/x/status/1986086104286597269

 

[dullard]

I didn't say it was crap. Simply that it is open to criticism because fmax dropped and it is shipping (i.e. that's where the specs come from now).

Criticism without full information is ignorance at best. Whether you call it crap or not, any criticism is invalid at this point. Even with shipped chips, the final BIOS hasn't shipped. GPU performance went up almost 10% last week alone.

Intel Core Ultra X9 388H "Panther Lake" iGPU Scores 33% Increase Over "Lunar Lake"

Laptop Review Club has released early benchmark data indicating that Intel's Core Ultra 300 series, known as "Panther Lake," represents a significant advancement in integrated graphics. The standout data comes from 3DMark Time Spy tests linked to the flagship Core Ultra X9 388H, which reportedly...

www.techpowerup.com

 

[gdansk]

Criticism without full information is ignorance at best.

No, it's objectively correct at worst. Fmax dropped relative to previous products. Some here have been in denial about it. But it's happening. You can say the product is better overall. That's a separate issue. It is a questionable showing for 18A.

 

[dullard]

No, it's objectively correct at worst. Fmax dropped relative to previous products. Some here have been in denial about it. But it's...it's over

Saved for posterity. And to show your unobjective colors. Frequency without IPC is a useless statistic. Frequency and IPC without consideration for power is ignorance (especially for mobile). All three without final BIOS and drivers shows nothing about technology.

 

[gdansk]

Saved for posterity. And to show your unobjective colors. Frequency without IPC is a useless statistic.

All I said is fmax dropped and you can criticize that. You know that IPC would be better applied if PTL could clock higher but alas... 18A can't yet. Sad.

And it's totally okay to say that sucks. It's a possibly good product being turned far less impressive because it couldn't reach a high enough clock rate.

 

[dullard]

All I said is fmax dropped and you can criticize that. You know that IPC would be better applied if PTL could clock higher but alas... 18A can't yet. Sad.

And it's totally okay to say that sucks.

What is 18A's Fmax at 115 W?

 

[Meteor Late]

Intel is not getting massive IPC increases to justify the stagnant clock speeds, it's as simple as that.
There is no justification for having the same clock speeds as Intel 4's Meteor Lake, none.

 

[Meteor Late]

Lets put both important pieces in the picture: turbo frequency AND turbo power levels used to reach that frequency (PL2).

• Alder Lake: 12900H 5.0 GHz top turbo, 115 W PL2
• Meteor Lake: 185H 5.1 GHz top turbo, 115 W PL2
• Lunar Lake: 288V 5.1 GHz top turbo, 37 W PL2
• Arrow Lake: 285H 5.4 GHz top turbo, 115 W PL2
• Panther Lake rumor: 388H 5.1 GHz top turbo, 45 W PL2 https://wccftech.com/intel-panther-...eak-core-ultra-300-16-cores-12-xe3-gpu-cores/

So, roughly the same top turbo speed with much lower power (other than lunar lake which fewer cores). What would a top frequency oriented chip get at 115 W?

That's completely irrelevant, because those power levels are only reached in multithreaded apps, none of these are going to consume 45W or more in 1T scenarios, which is where those frequencies happen. that is to say, Panther Lake doesn't have lower fmax just because it has a lower PL2, that's not how it works.

 

[gdansk]

View attachment 133221

https://twitter.com/x/status/1986086104286597269

See now this I really hope is wrong. 4+0+4 should not have 25W/55W...

 

[511]

See now this I really hope is wrong. 4+0+4 should not have 25W/55W...

Yeah it shouldn't be more than 37W cause it would never ever reach 55W like ever

 

[511]

Intel is not getting massive IPC increases to justify the stagnant clock speeds, it's as simple as that.
There is no justification for having the same clock speeds as Intel 4's Meteor Lake, none.

Why not .... It's not the same uArch as MTL it's bit down than my expectations though but Still fine.

 

[511]

All I said is fmax dropped and you can criticize that. You know that IPC would be better applied if PTL could clock higher but alas... 18A can't yet. Sad.

And it's totally okay to say that sucks. It's a possibly good product being turned far less impressive because it couldn't reach a high enough clock rate.

You are forgetting 30-40% reduction in power for the same ST which is pretty big this means you can have LNL st from like 15W to 9W that's big