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.

 

[poke01]

X86 laptops won't get there unless AMD/Intel decides to make their own chassis with their own OS and everything Apple optimizes at every level x86 vendors need to make their core more client focused as well they are quite a bit lacking in resourcing in terms of their cores

AMD has their own Strix halo box coming soon but if AMD/intel made their own laptops and everything else that be really good.

 

[511]

Yeah, pricing is the determining factor.

funny thing is in Australia at least I can’t buy any Panther lake laptops but the dell XPs online.

I would be interested to know how it is in other regions.

I checked bunch of different Region Europe only got like 1 PTL laptop US has the most

 

[511]

AMD has their own Strix halo box coming soon.

that's nice

 

[DavidC1]

X86 laptops won't get there unless AMD/Intel decides to make their own chassis with their own OS and everything Apple optimizes at every level x86 vendors need to make their core more client focused as well they are quite a bit lacking in resourcing in terms of their cores

Intel/AMD has a mindset problem too. Remember they are the source of the "fountain". The entire ecosystem is a problem. This is why the x86 ecosystem should focus on more repairability and modularity even on laptops, because competing with Apple in price and how premium it is, is a losing battle. They should be foremost on pushing standards like LPCAMM.

Like a common phrasing says, it's also a skill issue. They just suck at it. I don't know how people can justify the x86 chips are anywhere near close ARM ones. It's completely embarassing a same chip that can fit in a device that is carried in a regular pocket and perfectly fits in your hand can get anywhere near close a desktop chip with a heatsink large and heavy as an entire ultrabook. The argument for decades was that ARM inherently was slower and couldn't catch up in single thread.

I checked bunch of different Region Europe only got like 1 PTL laptop US has the most

It always takes about until March/April at least to get somewhat wide availability. Mainstream availability is June onwards, and peak is near to the announcement of next year's part, so like August. This is part of the reason Intel had a process lead. Their processor was available on Day 1 of announcement. You can buy everything on Day 1 of release for a desktop.

 

[Win2012R2]

X86 laptops won't get there unless AMD/Intel decides to make their own chassis with their own OS and everything Apple optimizes at every level x86 vendors need to make their core more client focused as well they are quite a bit lacking in resourcing in terms of their cores

That will never happen.

 

[DavidC1]

That will never happen.

They need to worry about fixing their hardware before worrying about things that are way out of their core skillset.

 

[poke01]

Linux is open source they can use that, they just need to care like Valve does.

 

[511]

It always takes about until March/April at least to get somewhat wide availability. Mainstream availability is June onwards, and peak is near to the announcement of next year's part, so like August. This is part of the reason Intel had a process lead. Their processor was available on Day 1 of announcement. You can buy everything on Day 1 of release for a desktop.

Desktop are different thing with laptops even if Intel ships the chips there are additional steps to make laptops which take like couple of months.

Intel/AMD has a mindset problem too. Remember they are the source of the "fountain". The entire ecosystem is a problem. This is why the x86 ecosystem should focus on more repairability and modularity even on laptops, because competing with Apple in price and how premium it is, is a losing battle. They should be foremost on pushing standards like LPCAMM.

There is no ecosystem like x86_64 in terms of CPU ISA like there is no comparison in terms of documentation and stuff available vs like an ARM chips you can get a chip over which you have total control ARM has superior Hardware in terms of Integer Cores and perf/watt for sure but when it comes to everything ARM is not close.

Like a common phrasing says, it's also a skill issue. They just suck at it.

LMAO x86 Vendors don't have skill issue they have different design methodology in terms of how they go about designing their cores

They need to worry about fixing their hardware before worrying about things that are way out of their core skillset.

Their Hardware is fine if anything they have the most robust hardware in terms of support like why do you think NVIDIA felt the need for X86 CPU in NVL Chassis.

 

[poke01]

They should be foremost on pushing standards like LPCAMM

LPCAMM is supported by Panther Lake.

 

[511]

Linux is open source they can use that, they just need to care like Valve does.

Intel is one of the 5 top most contributer to the Linux lol

 

[poke01]

laptops which take like couple of months.

I mean it’s impressive that Intel releases new laptop chips yearly. AMD doesn’t do that and Nvidia sucks at mobile.

Laptops are hard.

 

[511]

I mean it’s impressive that Intel releases new laptop chips yearly. AMD doesn’t do that and Nvidia sucks at mobile.

Laptops are hard.

Mobile Vendors deserves praise as well for yearly releases

 

[Hulk]

Just like every other product, things are priced at what the vendor believes the market will bear. I do believe Dell has lost their minds with this pricing but so has Porsche with the 911 yet they are selling every one they can produce.

The problem with pricing too high and lowering the price later is that sometimes you can't "run away" from that initial price fast enough.

Covid caused insane price run-ups. Vendors loved it and are refusing to back down pricing and why if people continue to buy at Covid pricing?

The Asus ProArtPX13 that I am typing on right now is listed for $1800 at Best Buy. I caught it on sale last summer for $1200. HX370, 32GB, 1TB, OLED, 4050. Functionally the equivilant of Panther Lake at half the cost. I can live with the battery life not being best-in-class.

Let's be smart with our money, that's how we push back against insane pricing.

 

[511]

Initial launch stuff is always expensive

 

[coercitiv]

Initial launch stuff is always expensive

Dell in particular tends to be one of the more expensive PC OEMs.

PS: It's fascinating for me how they managed to remain a top choice for value monitors for my use case, yet I haven't considered a Dell laptop in over a decade.

 

[Tachyonism]

AMD has their own Strix halo box coming soon

Still run Windows so meh. Plus the level of hardware integration as a whole is still far behind what Apple currently does.

They need to worry about fixing their hardware before worrying about things that are way out of their core skillset.

I mean, it's not that much. What they need to do is:

1) sending patches to ensure Linux can runs on those SoCs well (and fast too), which already happens.
2) ensuring OEMs parts and hardware have good Linux compatibility. This is the hard part, OEMs dont care at all.
3) throwing money around and pushing app devs to make their app available on Linux. This is probably the hardest part, but I found the current Linux ecosystem to cover all my daily needs, even gaming.

Also Linux is way more optimized compared to Windows, as long as there is proper driver (looking at you NVidia).

 

[poke01]

Still run Windows so meh

It runs both. AMD would have to provide first class Linux support for the device

 

[511]

I mean, it's not that much. What they need to do is:

1) sending patches to ensure Linux can runs on those SoCs well (and fast too), which already happens.
2) ensuring OEMs parts and hardware have good Linux compatibility. This is the hard part, OEMs dont care at all.
3) throwing money around and pushing app devs to make their app available on Linux. This is probably the hardest part, but I found the current Linux ecosystem to cover all my daily needs, even gaming.

Also Linux is way more optimized compared to Windows, as long as there is proper driver (looking at you NVidia).

Well there is only one choice than AMD GPUs for CPU both are fine Intel while having open source drive has issues compared to windows and NVIDIA doesn't care

 

[MerryCherry]

So Medusa has only 2 LP cores?

Is that enough?

Intel's Meteor Lake/Arrow Lake with 2 LP cores was a flop. Lunar Lake and Panther Lake have four.

Apple had 2 E-cores only in M1 Pro/M1 Max. Every other M chip has had atleast 4 E-cores.

Qualcomm with snapdragon x2 also has 6 'E-cores'.

Edit: sorry wrong thread

 

[511]

So Medusa has only 2 LP cores?

Is that enough?

Intel's Meteor Lake/Arrow Lake with 2 LP cores was a flop. Lunar Lake and Panther Lake have four.

Apple had 2 E-cores only in M1 Pro/M1 Max. Every other M chip has had atleast 4 E-cores.

Qualcomm with snapdragon x2 also has 6 'E-cores'.

depends weather it gets SLC or something MTL LP-E didn't have that so it hurt the performance and they were Skylake tier a Z6 tier 2C/4T shouldn't have issue for idle or video playback or other apps

 

[Fjodor2001]

depends weather it gets SLC or something MTL LP-E didn't have that so it hurt the performance and they were Skylake tier a Z6 tier 2C/4T shouldn't have issue for idle or video playback or other apps

What about the perhaps most common use case: web browsing?

Ideally you'd like to be able to do that without waking up the non-LPE cores. Or briefly wake up the non-LPE cores / (i)GPU to render the web page quickly, and then they can go to sleep while the user is reading the web page only relying on the LPE cores.

Possibly 4 LPE cores could be useful for this?

 

[adroc_thurston]

Is that enough?

4t so yeah.

Intel's Meteor Lake/Arrow Lake with 2 LP cores was a flop. Lunar Lake and Panther Lake have four.

Apple had 2 E-cores only in M1 Pro/M1 Max. Every other M chip has had atleast 4 E-cores.

Qualcomm with snapdragon x2 also has 6 'E-cores'.

they're wee little babies and this is an actual lobotomized Zen6. with SMT and all.

 

[511]

4t so yeah.

they're wee little babies and this is an actual lobotomized Zen6. with SMT and all.

As long as it gets something besides L2 maybe 2-4MB L3

What about the perhaps most common use case: web browsing?

Ideally you'd like to be able to do that without waking up the non-LPE cores. Or briefly wake up the non-LPE cores / (i)GPU to render the web page quickly, and then they can go to sleep while the user is reading the web page only relying on the LPE cores.

Possibly 4 LPE cores could be useful for this?

Well Zen 6 2C/4T ought to be enough but have to see what do they get

 

[adroc_thurston]

As long as it gets something besides L2 maybe 2-4MB L3

They can't really ship L3-less configs.
L2 macro has no fabric ports available.