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.

 

[moinmoin]

What is ADM? Adamantine?

Why does Adamantine need an abbreviation? You Intel fellows love abbreviations, eh?

WDYM? ISHYGDDT

 

[FlameTail]

WDYM? ISHYGDDT

FTLOGP

 

[mikk]

OK, here comes some leaks not from me but credential leaker about ARL & PTL:

• ARL-H (6P+8E) comes with iGPU (Build by newer process, N4P) 8 Xe1 LPG+ cores clocking at 2.3GHz
• ARL-HX (8P+16E, H and HX's tCPU are different dies) comes with 4 Xe cores
• SoC has been updated to support WiFi7, still on N6
• PTL-H's tCPU integrates with CPU (4P+8E+4LPe without HT), SoC and MC
• Here comes interesting part, PTL's tGPU comes with 12 Xe3 LPG core clocking at 2.5GHz. Yeah it is GT3 with 7.7 TF, slightly higher than Strix Point. And they are built on Intel 3+ process. Finally, IFS build tGPU.
• ADM so far is dead.

How do you get 7.7 Tflops? If Xe3 uses SIMD16 ALUs like Xe2 it's double than that unless they lower the number of vector engines per Xe core.

 

[igor_kavinski]

Core Ultra laptop availability on Amazon UAE

$2717 ASUS Zenbook 14 Duo UX8406MA-OLEDI9IG, Inkwell Gray, Dual Touch Screen, intel Ultra 9 32GB 1TB SSD, WIN11 HOME, 14.0inch 3K OLED, Touch Screen, Soft-Eng-Arb-KB

$1005 Acer Swift Go 14 your AI PC featuring the Intel Core Ultra 5-125H 14 Cores Upto 4.50GHz/16GB LPDDR5X 6400MHz/512GB SSD/Intel ARC Graphics/14" 2.8K OLED 100% sRGB/WiFi-6E/FP/BL KB/Win11 Home/Silver

$1331 Acer Swift Go 14 your AI PC featuring the Intel Core Ultra 7-155H 16 Cores Upto 4.80GHz/32GB LPDDR5X 6400MHz/1TB SSD/Intel ARC Graphics/14" 2.8K OLED 100% sRGB/WiFi-6E/FP/BL KB/Win11 Home/Silver

$3532 ROG Zephyrus G16 GU605MI-OLEDI9WPG Gaming Laptop, Intel Ultra 9 185H, 32GB 1TB SSD, NV RTX 4070, 8GB VRAM, WIN11 Pro, 16-inch 2.5K (2560 x 1600) 16:10, 240Hz, 1080 FHD, Webcam, Backlit-RGB-Eng-Arb-KB

That last one, why would anyone choose that instead of this:

$3695 ROG Strix SCAR 16 G634JZ-NM063W Off Black, Gaming Laptop, i9-13980HX, 32GB, 2TB SSD, NV RTX4080, WIN11 HOME, 16 inch QHD+ 16:10 (2560 x 1600, WQXGA) 240Hz, HD Webcam, RGB-Eng-Arb-KB + ROG Backpack

It would only make sense to buy the Core Ultra laptop with 4070 if it really did offer "ultra" level CPU performance. It doesn't so ASUS is just scamming honest, hardworking non-techies out of their money.

 

[DavidC1]

How do you get 7.7 Tflops? If Xe3 uses SIMD16 ALUs like Xe2 it's double than that unless they lower the number of vector engines per Xe core.

Remember, he's talking in Xe cores, not EUs.

8x Xe2 cores = 128EU by traditional definition.
12x Xe3 cores = 192EU x 2.5GHz x 16 Flops per EU = 7,680GFlops, or ~50% improvement over the 8 Xe2 core part, and the 128EU Meteorlake.

PTL-H's tCPU integrates with CPU (4P+8E+4LPe without HT), SoC and MC

This part makes no sense, so you might want to clarify that. tCPU is already a CPU...

Also, 780M's Flops aren't truly double 680M because it's done using co-issue while number of CUs remain the same at 12, while in post-Xe2's case it'll literally need double the transistors for compute.

PTL's iGPU is also GT3 only in name, as 192EU equivalent in 2026 won't be impressive by any means. If they had 16 or 24 Xe3 cores on the other hand, then GT3 would make sense.

It's like calling Icelake GT3 and Tigerlake GT4.

 

[Glo.]

How do you get 7.7 Tflops? If Xe3 uses SIMD16 ALUs like Xe2 it's double than that unless they lower the number of vector engines per Xe core.

12 Xe cores with 128 ALUs each = 1536 ALUs.

1536 x2 x 2.5 GHz = 7.68 TFLOPs.

 

[SiliconFly]

Only in LKF and PVC.

MTL is foveros. Its common knowledge. Even the video link says that!!!

Youtube link

Fyi, MTL is a very advanced form of Lakefield.

 

[SiliconFly]

Yes, MTL and ARL use more advanced packaging than Zen 4 or Zen 5 desktop (excluding the 3D cache variants which use more advanced packaging than what Intel currently has to offer). That's different than your original statement but if this was the point you were trying to make then, o.k. We'll see how much it actually helps them fairly soon.

Chiplets helped AMD immensely. I'm sure foveros will help Intel a lot.

But Intel probably has a big headache... LNC. If it doesn't deliver, nothing else can help them.

 

[SiliconFly]

I don’t get the hype over the packaging. There’s a high probability that the packaging will be the cause for ARL-S having poor gaming performance.

AMD APUs are chiplets in packaging. If it's good for gaming, so should ARL be.

Chiplets, tiles, whatever you want to call it is always a compromise solution in client products. It’s not a feature or impressive tech - it’s there to save money at the expense of a worse product.

Chiplets aren't cheaper by any means. Lisa sue herself said that their cost advantage comes only in economies of scale. Meaning large volume. Monolithic is cheaper for smaller cpus. Chiplets accelerate development, validation times, time to market, IP reuse, etc. Thats why it's all chiplets now a days.

 

[SiliconFly]

.

 

[igor_kavinski]

But Intel probably has a big headache... LNC.

How is it a headache? Any link to a rumor saying that it is underperforming, not hitting performance targets or yields are lower than expected?

 

[SiliconFly]

.

 

[SiliconFly]

.

 

[dullard]

Just out of curiosity, if so many are available, why is it that no major pc review sites have gotten their hands on retail MTL laptops for reviews!

This is just my own speculation, but I suspect it is a combo of a few problems that Intel had.

1) Not a lot of production capacity for the Intel 4 tile. Time will fix that, but it just isn't a mass produced tile yet. For example just about everything out there is the 155H. A full review of Meteor Lake itself should cover the range of CPUs, but they just don't seem to be available yet. And the 155H isn't particularly a "sexy" model to review--not the top of the line, not the cheapest, not the lowest power, not the highest power, etc.

2) Not a lot of advanced packaging capacity yet. Fab 9 which is the first to do mass production of Intel's advanced packaging just opened late January.

Combine #1 and #2 and there isn't a lot of incentive to have reviews of chips that are in limited quantities. Why advertise what you can't sell in any significant numbers? So free laptops were not being given to reviewers. Reviewers can and have purchased the laptops themselves, but this was mostly done by the smaller reviewers. See below for my speculations on the major review sites.

3) BIOS and vulnerability issues. There was the discussion of the BIOS fix last month, but just yesterday there were a slew of vulnerabilities patched. Why review something when an imminent patch could completely change the results? https://www.tomshardware.com/pc-com...ware-thunderbolt-xtu-chipset-drivers-and-more

4) Software isn't there yet. Much of Meteor Lake is the NPU--its AI functionality. But the software that would be helpful for this feature isn't ready yet. So a major review of something that you can't fully review might not really be a highly useful review.

That said, some of the more major sites have just started tricking out reviews now that the BIOS and other vulnerabilities have been patched. https://www.tomshardware.com/laptops/ultrabooks-ultraportables/asus-zenbook-14-oled-ux3405m-review

 

[Abwx]

Just out of curiosity, if so many are available, why is it that no major pc review sites have gotten their hands on retail MTL laptops for reviews!

Actually the Asus Rog Zephirus was tested at NBC, so far there s no price, but the also tested AMD Hawk Point counter part is 2500€ or so despite a smaller screen, perfs are more or less comparable using the same dGPU :

https://www.notebookcheck.net/Asus-ROG-Zephyrus-G16-2024-review-The-gaming-laptop-now-with-G-Sync-OLED-and-Intel-Meteor-Lake.800175.0.html

https://www.notebookcheck.net/Asus-ROG-Zephyrus-G14-2024-review-The-gaming-multimedia-laptop-with-Ryzen-8000-and-G-Sync-OLED.800772.0.html

 

[Hitman928]

Chiplets helped AMD immensely. I'm sure foveros will help Intel a lot.

But Intel probably has a big headache... LNC. If it doesn't deliver, nothing else can help them.

Chiplets helped AMD because they could design 1 chiplet and use it across multiple product lines effectively (Desktop, WS, Server, some notebook and embedded). This saved AMD a lot of design and validation work and simplified their supply logistics. For Epyc and TR, where the silicon area was very large, it also saved production costs. Performance wise, though, going chiplets was a degrading factor. AMD put a lot of work into minimizing the chiplet impact on performance and overall, made it work quite well. Intel has yet to show that they can use a tiled approach and execute it well, even when using more advanced packaging than AMD.

 

[Hitman928]

Actually the Asus Rog Zephirus was tested at NBC, so far there s no price, but the also tested AMD Hawk Point counter part is 2500€ or so despite a smaller screen, perfs are more or less comparable using the same dGPU :

https://www.notebookcheck.net/Asus-ROG-Zephyrus-G16-2024-review-The-gaming-laptop-now-with-G-Sync-OLED-and-Intel-Meteor-Lake.800175.0.html

https://www.notebookcheck.net/Asus-ROG-Zephyrus-G14-2024-review-The-gaming-multimedia-laptop-with-Ryzen-8000-and-G-Sync-OLED.800772.0.html

One thing to note is that despite having similar GPU TDP ratings, the MTL version appears to be able to sustain about 23% higher GPU power than the Hawk Point version (80W vs 65W) during their Witcher 3 test. It may just be how they are balancing the powers for the different models but it seems like there is some difference in how much GPU power is used from each model but the MTL model is able to sustain much higher system power as well.

Edit:

Witcher 3 example shows difference in power draw which should be coming from larger GPU power draw on the MTL model.

 

[Abwx]

They displayed the Witcher 3 curves for the GPU, in the 16" MTL laptop the 4070 use about 80W while in the 14" Hawk Point that s about 63W, guess that the actual TDP is configured according to the laptop size.

Measured at the main with external monitor the powers are respectively 122W and 105W with this game.

On CB R23 that s 103W with a peak at 115W at the start for MTL and a continuous 92W for the 8945HS, also measured at the main with external monitor.

 

[igor_kavinski]

Intel has yet to show that they can use a tiled approach and execute it well, even when using more advanced packaging than AMD.

What's stunning is that even with their huge R&D budget, what were they actually spending money researching during the BKrzanich era? I'm pretty sure that they internally laughed off the Zen/Zen2 performance gains and considered them just a fluke. Zen 3 was when they got shocked to their core and realized that AMD was on a crazy flawless execution path but by then it was too late to do anything radical about it.

They've been reaching into their node refinement jar to keep themselves from failing spectacularly compared to each new Ryzen but even that is almost empty now and they are faced with launching ARL on an unproven node so the chances of messing up are sky high. Had they taken Ryzen seriously from day one and considered it the end of their "lazily enjoying doughnut and coffee" days, they would not be in their current predicament. Had MTL development started (or accelerated at lightspeed) when the original Zen launched, we could've been looking at a serious M3 competitor from Intel.

Intel's failures deserve to be researched, released in book form and serve as a cautionary tale for anyone looking to head a global corporation to avoid the same missteps that led to Intel's downfall.

 

[dullard]

Intel's failures deserve to be researched, released in book form and serve as a cautionary tale for anyone looking to head a global corporation to avoid the same missteps that led to Intel's downfall.

You neglected the big issues: terrible management, infighting of groups, overly unrealistic goals, underfunding of production equipment, a chip shortage taking up that limited production equipment, and thus no ability to realistically test the research on the production lines. You can have all the R&D money in the world, but if you can't put the research into many iterations of actual chips to properly test your ideas, you have nothing.

 

[SiliconFly]

.

 

[igor_kavinski]

https://www.notebookcheck.net/Asus-ROG-Zephyrus-G16-2024-review-The-gaming-laptop-now-with-G-Sync-OLED-and-Intel-Meteor-Lake.800175.0.html

https://www.notebookcheck.net/Asus-ROG-Zephyrus-G14-2024-review-The-gaming-multimedia-laptop-with-Ryzen-8000-and-G-Sync-OLED.800772.0.html

Damn shame that the Crater Lake laptop can't touch the Ryzen laptop even with a larger battery.


The first three tests the MTL SoC was supposed to excel at and even in the last test, the Ryzen laptop gives it a run for its money.

But don't worry, Intel. Your fans won't let you sweat. They will keep you nice and cool with their disposable dollars!

 

[igor_kavinski]

It was said that the engineering & foundry teams were not even able to directly communicate their ideas/issues with these dimwits not only due to their lacks of understanding of tech, but also because they thought that engineers were beneath them!

Yeah my COO is just like that. Thinks of these people as servants for hire and ignores their suggestions. I keep wondering how long before we post a loss. Then it's gonna be him pointing fingers at everyone but himself.

 

[dullard]

Yeah my COO is just like that. Thinks of these people as servants for hire and ignores their suggestions. I keep wondering how long before we post a loss. Then it's gonna be him pointing fingers at everyone but himself.

It is truly demoralizing when leaders are too hard headed to listen. I've had the same experience myself -- although in biotech. This article sums my experience up completely: https://www.npr.org/sections/thetwo... ago, as the,managers and NASA overruled them.

 

[PJVol]

Damn shame that the Crater Lake laptop can't touch the Ryzen laptop even with a larger battery.

Can't believe it took 235 pages to figure out what this CrapLake is, lol