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.

 

[Magio]

The 30 W Ultra 9 288V exists solely for giving the iGPU more oomph, and will therefore be used mainly in gaming-centric designs.

Also, the Prestige 13 Meteor Lake had pretty meh cooling.

I wouldn't hold my breath for any magical MT improvements. 12-13K in GB6 MT is the upper limit of what it may achieve.

Also, there is no '30 W sustained' or, for that matter any fixed power limit in Intel Mobile CPU land unless you disable DTT using Throttlestop. So comparisons of MT score, especially on different laptop models, is meaningless.

Of course I'm not expecting a 2x (or even 1.5x) from the 17W parts in CPU benchmarks, but a measurable improvement would be welcome. It's highly unlikely to compete with a 28W Strix part in MT CPU tasks, though.

And it's not a hard coded guarantee, but you'd expect a good design around a 30W PL1 part to at least be able to sustain close to 30W, just like you'd expect those that will ship the 17W SKUs to sustain roughly 17W with short bursts at 30W.

 

[FlameTail]

PTL-U = 4P + 4LPE
PTL-H = 4P + 8E + 4LPE

Clearly, PTL-U is the spiritual successor of Lunar Lake.

 

[511]

PTL-U is just a cheaper LNL variant from Intel i am more excited about PTL-H with 4+8+4 + 12 Xe3 cores if it can give LNL battery with 3050-3060 mobile Gpu performance it will be solid but it will be bandwidth starved if it's still at 128bit i wish they make the mc support 192bit as an option

 

[Magio]

The thing with PTL-U is it's hard to imagine it be that compelling an option at the top end when that Dell leaked roadmap says the early 2026 XPS 13 redesign will still ship with LNL (and Oryon V2), only getting a refresh when NVL-U drops in early 2027.

 

[511]

Those are leaked roadmap and are bound to change they are 2-3+ year in future crazy things can happen in that time

 

[511]

Those are leaked roadmap and are bound to change they are 2-3+ year in future crazy things can happen if PTL-U is pin compatible with LNL. which it should be maybe?

 

[H433x0n]

Using basic napkin math, it appears Lunar Lake could have an IPC advantage against Zen 5.

Zen 5 has +7% int and +12% fp advantage over Redwood Cove and Lion Cove is advertised as having a +14% IPC increase over Redwood Cove.

 

[Henry swagger]

View attachment 104153

Using basic napkin math, it appears Lunar Lake could have an IPC advantage against Zen 5.

Zen 5 has +7% int and +12% fp advantage over Redwood Cove and Lion Cove is advertised as having a +14% IPC increase over Redwood Cove.

Zen 5 competition will be arrow lake hx and h not lunar lake.. lunar lake competition is m2 m3

 

[SteinFG]

Zen 5 competition will be arrow lake hx and h not lunar lake.. lunar lake competition is m2 m3

AMD's Zen 5 line is their most efficient, they will definitely compete with lunar lake. Especially Kraken. Kraken and Lunar wil have comparable ST numbers, and Kraken will win by some% in MT. They'll probably be close in efficiency. Gaming perf should be similar too, I hope.

 

[FlameTail]

AMD's Zen 5 line is their most efficient, they will definitely compete with lunar lake. Especially Kraken. Kraken and Lunar wil have comparable ST numbers, and Kraken will win by some% in MT. They'll probably be close in efficiency. Gaming perf should be similar too, I hope.

Lunar Lake will crush Kraken in gaming, becuase LNL has a fatter GPU.

 

[exquisitechar]

View attachment 104153

Using basic napkin math, it appears Lunar Lake could have an IPC advantage against Zen 5.

Zen 5 has +7% int and +12% fp advantage over Redwood Cove and Lion Cove is advertised as having a +14% IPC increase over Redwood Cove.

Looks like it. That int performance is pretty bad considering the breadth of the changes in Zen5. People were right about AMD's IPC claims/questionable benchmark choices, it's the first time that they aren't representative of overall performance at all and it shows in SPEC.

 

[KompuKare]

Lunar Lake will crush Kraken in gaming, becuase LNL has a fatter GPU.

Yet ARC A770 is far fatter than the GPUs it competes with. Now Intel's GPUs have improved since ARC but fatter/bigger dies not automatically mean it is faster.

 

[511]

Yet ARC A770 is far fatter than the GPUs it competes with. Now Intel's GPUs have improved since ARC but fatter/bigger dies not automatically mean it is faster.

100% agree LNL will tell us will Xe2 be good or flop cause they share the whole architecture unlike MTL-ARL and the original Xe

 

[Ghostsonplanets]

PTL-U is just a cheaper LNL variant from Intel i am more excited about PTL-H with 4+8+4 + 12 Xe3 cores if it can give LNL battery with 3050-3060 mobile Gpu performance it will be solid but it will be bandwidth starved if it's still at 128bit i wish they make the mc support 192bit as an option

PTL should adopt LNL 8533 LP5X memory controller (Or even be improved to support 9600/10700T). Besides, Intel has a lot of cache (8MB) for the iGPU. So their designs aren't as memory bandwidth starved as the competitors.


Intel problem really was that their GPU IP nor their software stack were ready for the prime time. Which is something I suppose will be fixed by the time Panther Lake comes to market. Because, otherwise, they seem to be much more serious and giving more area consideration towards iGPU performance than AMD for example.

 

[Abwx]

View attachment 104153

Using basic napkin math, it appears Lunar Lake could have an IPC advantage against Zen 5.

Zen 5 has +7% int and +12% fp advantage over Redwood Cove and Lion Cove is advertised as having a +14% IPC increase over Redwood Cove.

Intel s 14% is comparable to AMD s 16%, that is, an average of various softwares, not only Spec, you are implying here that LNL will also have 14% uplift in Spec.

 

[ErichZannMusician]

Intel s 14% is comparable to AMD s 16%, that is, an average of various softwares, not only Spec, you are implying here that LNL will also have 14% uplift in Spec.

They're not.

AMD's +16% is not comparable to Intel's +14 because they use different software to arrive at those numbers.

 

[511]

Btw an interesting thing with the LNC slide in LNL it shows that Intel 4 is superior to N3B at high power assuming +14% IPC it means intel was not lying about their Intel 4/3 providing same P/W as N3 derivative

 

[511]

PTL should adopt LNL 8533 LP5X memory controller (Or even be improved to support 9600/10700T). Besides, Intel has a lot of cache (8MB) for the iGPU. So their designs aren't as memory bandwidth starved as the competitors.


Intel problem really was that their GPU IP nor their software stack were ready for the prime time. Which is something I suppose will be fixed by the time Panther Lake comes to market. Because, otherwise, they seem to be much more serious and giving more area consideration towards iGPU performance than AMD for example.

Yeah as Xe2 will give us a glimpse of their effort too bad BK ran the company to ground no innovation nothing It took gelsinger to correct predecessors mistake he is just dumping money in fabs and their core buisness

 

[Abwx]

Btw an interesting thing with the LNC slide in LNL it shows that Intel 4 is superior to N3B at high power assuming +14% IPC it means intel was not lying about their Intel 4/3 providing same P/W as N3 derivative

LNL CPU use TSMC s N3 while MTL CPU use Intel 4.

 

[DavidC1]

Using basic napkin math, it appears Lunar Lake could have an IPC advantage against Zen 5.

Zen 5 has +7% int and +12% fp advantage over Redwood Cove and Lion Cove is advertised as having a +14% IPC increase over Redwood Cove.

Where's the Wave that claims that Arrowlake is awesome now? It comparatively sucks much less than 2 days ago isn't it?



I do speculate though Lion Cove may be similar in that FP gains are going to be greater than Int. No, it's practically guaranteed with the 4 FP units.

They'll probably be close in efficiency.

Efficiency at this level is pretty much battery life, which Lunarlake should be significantly better at.

 

[lightisgood]

Intel’s Next-Generation Core Ultra Launch Event on Sept. 3​

Media Alert: Intel’s Next-Generation Core Ultra Launch Event on Sept....

Intel introduces next-gen Intel Core Ultra processors, code-named Lunar Lake, revealing breakthroughs in efficiency, compute and AI performance.

www.intel.com

 

[SiliconFly]

-

 

[511]

LNL CPU use TSMC s N3 while MTL CPU use Intel 4.

I know but if we take 14% IPC as avg uplift at higher curver Intel 4 is better than N3B this method has its flaws

 

[Abwx]

I know but if we take 14% IPC as avg uplift at higher curver Intel 4 is better than N3B this method has its flaws

That doesnt work like this, IPC has nothing to do with a process perf/watt.

 

[DavidC1]

I know but if we take 14% IPC as avg uplift at higher curver Intel 4 is better than N3B this method has its flaws

Intel sucks on mobile for this very reason. The whole company has been in the mindset of highest performance transistors for 30 years.

Therefore it is natural that at the peak clocks, Intel processes are better. But in Lunarlake and Server products, TSMC is better.