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.

 

[Hitman928]

Use of different power profiles

Not sure what you mean, he wanted to test all three at the same power limit so he had to manually limit the MTL and STX laptops. . .

and using XeSS for Wukong instead of FSR

Why is this an issue?

 

[Hitman928]

The issue is that we are seeing differences of 20W under load and people are attributing it to the OLED display.

The display is completely different and has 2x the max frame rate as well as being larger. It probably doesn’t explain the whole difference, but it’s a huge chunk of it. Lots of other little efficiency points can add up to the rest, or close to it.

 

[Hitman928]

That's just PPT difference and is an OEM preference.

PPT limits for the SoCs were equalized for this test.

 

[controlflow]

Why is this an issue?

XeSS on LNL's GPU is done using XMX which produces higher image quality vs the DP4a path on other GPUs. I suppose that makes it a not too apples to apples comparison arguably.

 

[Dayman1225]

AMD basically goes one server gen per 2 years. Intel goes one per year - 2023 Sapphire + Emerald; 2024 Granite; 2025 Diamond

Intel is gonna be server leader from 2025 Diamond Rapids launch 'till late 2026 Venice launch.

DMR is not even close to a 2025 product

 

[controlflow]

Lion Cove: Intel’s P-Core Roars

Intel’s mobile CPUs have undergone massive changes over the past couple generations as Intel defends its laptop market against AMD, Qualcomm, and to a lesser extent Apple.

chipsandcheese.com

The conclusion there is certainly different than the general sentiment here towards Lion Cove.

"
Lion Cove is a much improved architecture compared to Redwood Cove, and shows Intel still has potent engineering muscle despite recent setbacks. Traditionally Intel delivered significant architecture changes during a “tock” in a tick-tock cycle. That reduces risk by separately handling process node and architecture changes. Lunar Lake not only combines a new architecture with a move to a new node, but also drops system level changes on top. At a time when Intel’s facing increased pressure from all sides, a move like Lunar Lake is a sign that Intel can adapt and survive.

Intel’s upcoming Arrow Lake desktop CPU will let Lion Cove stretch its legs with more cache and a larger power budget. Lower latency DDR5 should improve performance even further. After seeing Lion Cove perform well in a mobile form factor, I’m optimistic about what the same architecture can do on desktop. Recently Intel has is sitting on a rather unstable foundation right now with Raptor Lake, and Arrow Lake’s release will be a great time to put the company’s high performance chips back on stable footing.

"

 

[Bouowmx]

What would be the first server/mobile/desktop arch with APX?

Linux 6.10 x86 Instruction Decoder Prepares For APX & Other New Intel Instructions - Phoronix

www.phoronix.com

 

[MarkPost]

Can you understand What he’s saying?View attachment 108335View attachment 108336

dude, thats the whole laptop! not SOC. SOC is fixed to 28W for both.

Thats the reason why any battery life or efficiency comparison between laptops with different components is just misleading and useless (if your goal is to compare SOCs in that regard).

So for that, you need to compare the exact laptop model with the exact other components, SOC apart.

Its not that hard to understand.

 

[AMDK11]

Edit:

https://i0.wp.com/chipsandcheese.com/wp-content/uploads/2024/09/image-2.png?resize=1536%2C964&ssl=1

https://i0.wp.com/chipsandcheese.com/wp-content/uploads/2024/09/redwood_cove-1.png?w=1348&ssl=1

 

[9949asd]

dude, thats the whole laptop! not SOC. SOC is fixed to 28W for both.

Thats the reason why any battery life or efficiency comparison between laptops with different components is just misleading and useless (if your goal is to compare SOCs in that regard).

So for that, you need to compare the exact laptop model with the exact other components, SOC apart.

Its not that hard to understand.

What are you talking about? Are you saying 2 more inches screen will use 20+w??? I am so tired about this.

 

[poke01]

What are you talking about? Are you saying 2 more inches screen will use 20+w??? I am so tired about this.

ROG Zephyrus G16 have dGPU so this impacts total system power draw but they should be disabled right?
So yeah where is this extra 20+w difference coming from if all CPUs were limited to 28W, this may mean that AMD's Strix and MTL iGPU is more power hungry then.

 

[NTMBK]

Yet Intel's bottom line is bleeding red while AMD's books are flush. Can you explain this?

AMD is making huge inroads in the high margin datacenter market, even though they're not really getting anywhere in laptops. (Since they make much higher profits for the same number of wafers, it makes sense that they prioritise that market.)

 

[Hitman928]

ROG Zephyrus G16 have dGPU so this impacts total system power draw but they should be disabled right?
So yeah where is this extra 20+w difference coming from if all CPUs were limited to 28W, this may mean that AMD's Strix and MTL iGPU is more power hungry then.

The iGPU power is part of the 28 W limit. If you check the CB result in the review, it’s clear the 28 W limit is in place. It is not that crazy that a near 20 W difference can exist at the system level between these laptops, outside of the SoC power.

 

[KompuKare]

What are you talking about? Are you saying 2 more inches screen will use 20+w??? I am so tired about this.

I think that's mostly a strawman though.
I think what people have been saying repeatedly is that the extra 2 inches on an OLED explains SOME of the difference, not ALL of the differences.

Anyway didn't this start with those WHOLE system power differences with the synthetic scores to make it out like Lunar Lake's was some iGPU miracle?

Because while BMG seems to translate synthetics into gaming performance a lot better than any previous Intel iGPU, there is still a huge difference between synthetics and gaming. At least on the CB review where the synthetic Steel Nomad score was almost double the Strix score, but in games they were almost tied. This may mean that Intel's driver team may be able to squeeze out more performance eventually, on the other hand BG3 and Cyberpunk are both games for which Intel's driver team should already have optimised extensively.

 

[511]

I think that's mostly a strawman though.
I think what people have been saying repeatedly is that the extra 2 inches on an OLED explains SOME of the difference, not ALL of the differences.

Anyway didn't this start with those WHOLE system power differences with the synthetic scores to make it out like Lunar Lake's was some iGPU miracle?

Because while BMG seems to translate synthetics into gaming performance a lot better than any previous Intel iGPU, there is still a huge difference between synthetics and gaming. At least on the CB review where the synthetic Steel Nomad score was almost double the Strix score, but in games they were almost tied. This may mean that Intel's driver team may be able to squeeze out more performance eventually, on the other hand BG3 and Cyberpunk are both games for which Intel's driver team should already have optimised extensively.

IMO
>Synthetic is best case scenario with proper Drivers
> Real games shows how good the game is actually optimized for
It's! Basically Potential vs Actual performance

 

[KompuKare]

IMO
>Synthetic is best case scenario with proper Drivers
> Real games shows how good the game is actually optimized for
It's! Basically Potential vs Actual performance

Theoretically, but my point was there is no way that Intel's drive team have left that much performance on the table for a game like Cyberpunk:
Steel Nomad is 191% Vs Strix
Cyberpunk is 98% Vs Strix.
Further optimisation might gain a few percentage points but almost double? Intel's driver team is not that incompetent.

Also, ComputerBase only tested two games and ComputerBase is notorious for not offending big companies - especially Intel - so neither BG3 nor Cyberpunk are unlikely to be some badly optimised outliner; they are more likely to be featured in any Intel reviewer's guide IMO.

 

[511]

Theoretically, but my point was there is no way that Intel's drive team have left that much performance on the table for a game like Cyberpunk:
Steel Nomad is 191% Vs Strix
Cyberpunk is 98% Vs Strix.
Further optimisation might gain a few percentage points but almost double? Intel's driver is not that incompetent.

Also, ComputerBase only tested two games and ComputerBase is notorious for not offending big companies - especially Intel - so neither BG3 nor Cyberpunk are likely to be some badly optimised outliner; they are more likely to be featured in any Intel reviewer's guide IMO.

Cyberpunk is fair game but i want some older games like shadow of war Deus Ex Mankind Divided Prey tested as well it can take time but i want many games from past 7-10 years nowdays we get Buggy mess in the name of video games and no youtuber tests crysis 🙂

 

[MarkPost]

What are you talking about? Are you saying 2 more inches screen will use 20+w??? I am so tired about this.

you missed @Josh128 post? https://forums.anandtech.com/thread...akes-discussion-threads.2606448/post-41306858



Here HX 370 has longer battery life. Different models so yet not a proper comparison, but more accurate than an usseless 14" vs 16"

 

[The Hardcard]

Better we compare HX 370 with Lunar Lake. Cos, if we try to compare HX 370 with ARL-H parts, it'll be a complete bloodshed!

It's time for Lisa Su to retire.

I’ll be comparing ARL-H to Strix Halo. It’ll be interesting to see how they land in performance, price point, and power consumption.

 

[Josh128]

Excellent article.

A snippet from the Chips & Cheese article:
"Intel’s upcoming Arrow Lake desktop CPU will let Lion Cove stretch its legs with more cache and a larger power budget. Lower latency DDR5 should improve performance even further. After seeing Lion Cove perform well in a mobile form factor, I’m optimistic about what the same architecture can do on desktop."

And like always, with ARL-S 285K final, I still think we can expect 15% to ~20% IPC uplift (over 14900K). And >10% to 15% overall ST performance uplift. I've been saying the same all along and I stand by it.

Many initially claimed no more than 5%. Some of them eventually revised it to 9% and some 10%. But I'm saying with very high confidence that it's definitely gonna be greater than 10% easily. Time will tell.

Zen 5 may get hammered!

Recent Geekbench 6 runs dont support the hammering part at all.

285K - Geekbench 6 CPU Search - Geekbench

265K - Geekbench 6 CPU Search - Geekbench

 

[SiliconFly]

.

 

[cebri1]

I think there's a small difference between ES & final silicon.

Those are probably QS, not ES.

 

[cebri1]

you missed @Josh128 post? https://forums.anandtech.com/thread...akes-discussion-threads.2606448/post-41306858

View attachment 108378


Here HX 370 has longer battery life. Different models so yet not a proper comparison, but more accurate than an usseless 14" vs 16"

I thought you didn’t care about battery, there are outlets everywhere.

 

[SiliconFly]

.

 

[MarkPost]

I thought you didn’t care about battery, there are outlets everywhere.

?