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 7 360	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	4.8 GHz	5 GHz		4.8 GHz
L3 Cache		12 MB	6 MB	12 MB		12 MB
TDP		15 - 55 W	15 - 35 W	17 - 37 W		25 - 55 W
Memory		128-bit LPDDR5-5200	64-bit LPDDR5x-7467	128-bit LPDDR5x-8533		128-bit LPDDR5x-7467
Size		96 GB	48 GB	32 GB		128 GB
Bandwidth		83 GB/s	60 GB/s	136 GB/s		120 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	2.6 GHz	2 GHz		2.5 GHz
NPU		GNA 3.0	17 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.

 

[511]

https://twitter.com/x/status/1886818040056365063

ARL-H Deep Dive

TSMC救不了Intel：酷睿Ultra 7 255H（Arrow Lake H45）测试 | David Huang's Blog

 

[Meteor Late]

Very bad efficiency in single core, worse than Strix Point with a better node, except at the top of the curve.

 

[igor_kavinski]

258V looks most impressive there.

After that, HX370 is great but then its curve becomes similar to 255H. Almost like they are copying each other's work.

 

[Meteor Late]

258V looks most impressive there.

Not necessarily, we are using a better node and Memory on Package to achieve parity in the vast majority of the curve with the competition.

 

[igor_kavinski]

we

Current or ex-Intel engineer? 🙂

 

[desrever]

Very bad efficiency in single core, worse than Strix Point with a better node, except at the top of the curve.

I wouldn't call Intel 3 a better node than TSMC N4

 

[Hitman928]

I wouldn't call Intel 3 a better node than TSMC N4

ARL-H is using TSMC’s N3(B).

 

[SiliconFly]

.

 

[igor_kavinski]

255H beats HX370 at high frequencies! Nice.

Seems like a very bad missed opportunity to me. If that 258V line could've been extended a bit further to 14W without any regression, that would've been awesome. Guess they didn't have the foresight to do something like Strix Halo.

 

[cannedlake240]

View attachment 116268
DMR Launching Q2/Q3 26?

Idk how impactful DMR will be without smt and probably slightly lower core count... It might narrow the gap vs Epyc in some scenarios but lose ground in others, as opposed to Granite which was a clear improvement over prior gens in all aspects. Xeon L3 has the same latency issues seen on client too. Guess 2028 is the new 2025, Jaguar, Coral etc is their next chance at catching up.

 

[dttprofessor]

AMD's good days in mid-to-high-end notebooks are over, even with an 8-10% IPC limitation on Pcore@285H (compared to 258V), ULTRA2 still has considerable competitiveness. The only competitive products AMD currently has in the laptop market are the low-end 7845H and 8845H.

 

[desrever]

ARL-H is using TSMC’s N3(B).

Oh I see, I totally missed this, thought they were the same as the Ultra U series.

 

[adroc_thurston]

AMD's good days in mid-to-high-end notebooks are over, even with an 8-10% IPC limitation on Pcore@285H (compared to 258V), ULTRA2 still has considerable competitiveness.

Meds.

Guess they didn't have the foresight to do something like Strix Halo.

THEY DID. ARL-H 683 was canned.

 

[poke01]

ARL-H/HX is meh. PTL-H is where the fun begins

 

[adroc_thurston]

PTL-H is where the fun begins

Oh lmao no, the CPU part there is completely unremarkable.
Now, GPU? yea.

 

[poke01]

Oh lmao no, the CPU part there is completely unremarkable.
Now, GPU? yea.

Doesn’t PTL also get rid of MTL design junk

 

[adroc_thurston]

Doesn’t PTL also get rid of MTL design junk

How does that stop the cores (and the L3) from Not Sucking?

 

[SiliconFly]

.

 

[gdansk]

Improvements in design. Like how AMD went from Bulldozer to Zen.

PTL has a totally different uarch?

 

[adroc_thurston]

But for some reason, NVL is again switching back to MTL/ARL tile layout (mostly) with some modifications/improvements

Cost.

Funny thing is, Zen 6 is also expected to follow similar Intel tile designs.

It follows Strix-halo.

PTL is expected to bring in better performance due to the new design compared to ARL-H.

w r o n g.

Improvements in design. Like how AMD went from Bulldozer to Zen.

what the-

 

[SiliconFly]

.

 

[adroc_thurston]

Nope.

Yeah.

Nope.

Yeah. The selling point of PTL-H is the iGFX.

Thats the-

PTL does not introduce any new cores.

 

[SiliconFly]

.

 

[poke01]

So apart from iGPU, PTL is not exciting?
Hopefully NVL-H is better otherwise AMD gonna eat Intel for lunch with their mobile designs.

 

[dullard]

PTL does not introduce any new cores.

Last rumor that I saw was more LP-E cores (4 instead of 2). And all cores get upgrades. Cougar Cove P cores instead of Lion Cove. E cores get a few optimizations, but aren't much different. LP-E cores should be able to be ran much faster (and hopefully will then be functionally usable) when not on TSMC 6. And that isn't even counting if there are power efficiency gains from 18A.

What evidence do you have that this is no CPU improvement over ARL-H?