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.

 

[Hulk]

Remember the good old days when Intel didn't constantly update us on process status? They just put the product on the shelves and let it speak for itself.

18A status update #23 "launches" Q2 '25...

Does anybody remember Intel 20A, now THAT was a good launch.

Sorry, I AM an Intel fan. I just want them to DO something meaningful. Although the Battlemage release was meaningful I must admit. Just no volume.

 

[511]

Remember the good old days when Intel didn't constantly update us on process status? They just put the product on the shelves and let it speak for itself.

18A status update #23 "launches" Q2 '25...

Does anybody Intel 20A, now THAT was a good launch.

Sorry, I AM an Intel fan. I just want them to DO something meaningful. Although the Battlemage release was meaningful I must admit. Just no volume.

That was a different Intel I think after BK there were constant delays and stuff for almost everything

 

[igor_kavinski]

Am I reading Q1 2026 wrong? Does that mean no H2025 Launch

At least there should be a nice Core Ultra 300 launch by Q42025.

Intel can save some face by launching 4P+32E halo part for the rendering/video encoding/workstation crowd. That would catch AMD off-guard for sure, unless AMD has a similar design ready with Zen5c and they can ramp up volume in a matter of months.

 

[511]

At least there should be a nice Core Ultra 300 launch by Q42025.

Intel can save some face by launching 4P+32E halo part for the rendering/video encoding/workstation crowd. That would catch AMD off-guard for sure, unless AMD has a similar design ready with Zen5c and they can ramp up volume in a matter of months.

Not happening I wonder how powerful Nova Lake will be though it's got 16+32+4 Config

 

[igor_kavinski]

Not happening

I wouldn't bet on that.

Intel’s new CEO gets a $1 million salary plus $68 million in bonuses

That looks like a lot of money.

www.tomshardware.com

They got him to invest his own money in Intel shares so he will try his best to put something competitive out ASAP otherwise he stands to lose a lot.

Even something as simple as bringing back HT in Arrow Lake (assuming it's just fused off) and AVX-512 (maybe re-using the server compute tile in consumer) could be huge things that would improve the value proposition of Intel's CPU line-up and competitiveness.

 

[adroc_thurston]

18A status update #23 "launches" Q2 '25...

Ahhh Clearwater Forest.
What a poor and tortured part.

 

[jdubs03]

I wouldn't bet on that.

Intel’s new CEO gets a $1 million salary plus $68 million in bonuses

That looks like a lot of money.

www.tomshardware.com

They got him to invest his own money in Intel shares so he will try his best to put something competitive out ASAP otherwise he stands to lose a lot.

Even something as simple as bringing back HT in Arrow Lake (assuming it's just fused off) and AVX-512 (maybe re-using the server compute tile in consumer) could be huge things that would improve the value proposition of Intel's CPU line-up and competitiveness.

Agreed. And yeah that’s definitely a good motivator when you’re helming a company that is in the doldrums.

 

[poke01]

On the scale to 1 to 10, how good is the iGPU in Panther lake?

 

[511]

On the scale to 1 to 10, how good is the iGPU in Panther lake?

From the rumors and stuff it's somewhere between 1.6X - 2X LNL iGPU

 

[511]

So according to AMD they are only 10% ahead in performance pef core?

 

[QuickyDuck]

So according to AMD they are only 10% ahead in performance pef core?

View attachment 119998

EPYC is ahead in performance because Xeon is power constrained.
Given enough power P core should easily surpass zen core, albeit lacking in efficiency.

 

[511]

EPYC is ahead in performance because Xeon is power constrained.
Given enough power P core should easily surpass zen core, albeit lacking in efficiency.

The P cores are not efficient Yes that is true but this is per core performance I doubt it's power constrained

 

[poke01]

From the rumors and stuff it's somewhere between 1.6X - 2X LNL iGPU

and how are they achieving that with limited bandwidth. more cache?

 

[adroc_thurston]

more cache?

2x the SLC and the new uarch.

 

[511]

and how are they achieving that with limited bandwidth. more cache?

Faster Memory Speeds as well

 

[poke01]

Faster Memory Speeds as well

Nope that stays the same as lunar lake

 

[Ghostsonplanets]

and how are they achieving that with limited bandwidth. more cache?

Aside from what everyone else said, PTL-H iGPU is also 50% wider than ARL-H Xe LPG and LNL Xe² LPG.

IMO:

PTL - H 12Xe³ - New iGP performance class
PTL - U 4/6Xe³ - =<~ MTL-H performance class
WCL - 2Xe³ - ~= MTL-U performance

 

[511]

What a deal Non US people don't get good deals.

https://www.bestbuy.com/site/asus-vivobook-s-14-14-oled-laptop-copilot-pc-intel-core-ultra-5-16gb-memory-512gb-ssd-neutral-black/6595523.p?skuId=6595523&nrtv_cid=3fpu7f921haku&irgwc=1&ref=198&loc=Howl%20Technologies%2C%20Inc.&acampID=0&mpid=376373&affgroup=%22Content%22%2C%22Partners%20to%20Keep%22&intl=nosplash

 

[511]

For 12Xe3 we have many Improvements also there is clear PPA Improvement as well vs Xe2 here are all the changes that have been leaked.

> 2X iGPU cache
> 20% Clock Speed Improvement
> 1.5X the Xe Cores
> Major Improvements to the Arch with Xe3
> Size Increase of iGPU die from 39mm2 N3B to 54mm2 N3E(N3B is denser in SRAM and Logic vs N3E)
> Variable register allocation
> New Xe Slice organization with 6 Xe3 Cores

Did I miss something?

 

[gdansk]

What a deal Non US people don't get good deals.

https://www.bestbuy.com/site/asus-vivobook-s-14-14-oled-laptop-copilot-pc-intel-core-ultra-5-16gb-memory-512gb-ssd-neutral-black/6595523.p?skuId=6595523&nrtv_cid=3fpu7f921haku&irgwc=1&ref=198&loc=Howl%20Technologies%2C%20Inc.&acampID=0&mpid=376373&affgroup=%22Content%22%2C%22Partners%20to%20Keep%22&intl=nosplash

It was $450 earlier.

 

[511]

It was $450 earlier.

that is a steal NGL it's the best laptop for normal use at that price bar none

 

[Ghostsonplanets]

For 12Xe3 we have many Improvements also there is clear PPA Improvement as well vs Xe2 here are all the changes that have been leaked.

> 2X iGPU cache
> 20% Clock Speed Improvement
> 1.5X the Xe Cores
> Major Improvements to the Arch with Xe3
> Size Increase of iGPU die from 39mm2 N3B to 54mm2 N3E(N3B is denser in SRAM and Logic vs N3E)
> Variable register allocation
> New Xe Slice organization with 6 Xe3 Cores

Did I miss something?

Outside of sub-triangle opacity culling for better RTRT, no.

Where was the tile size of 12Xe³ shared? That's quite the increase in area dedicated for graphics.

 

[511]

Outside of sub-triangle opacity culling for better RTRT, no.

Where was the tile size of 12Xe³ shared? That's quite the increase in area dedicated for graphics.

If they add major changes with Xe3 than this area increase is justified example a 12 Xe2 would be slightly larger than 60mm2 N3B(considering the fact that the graphic tile consists of exclusively the cores and the caches)

https://twitter.com/x/status/1812898063502938260

 

[Ghostsonplanets]

If they add major changes with Xe3 than this area increase is justified example a 12 Xe2 would be slightly larger than 60mm2 N3B(considering the fact that the graphic tile consists of exclusively the cores and the caches)

https://twitter.com/x/status/1812898063502938260

View attachment 120016

Thank you!

I'm certainly interested in how Intel will improve PPA with PTH/WCL Celestial.

Battlemage already was a massive improvement with a 20 Xe² die being faster than a ACM 32 Xe die. And Lunar Lake was already very toe to toe with Ryzen AI HX 390 at lower TDP.

Perhaps we'll see the first performance leap from Intel over AMD on IGP space?

 

[511]

Thank you!

I'm certainly interested in how Intel will improve PPA with PTH/WCL Celestial.

Battlemage already was a massive improvement with a 20 Xe² die being faster than a ACM 32 Xe die. And Lunar Lake was already very toe to toe with Ryzen AI HX 390 at lower TDP.

Perhaps we'll see the first performance leap from Intel over AMD on IGP space?

The area efficiency sucks for battlemage sucks as a dGPU though they improve that as well with celestial and for second part strix halo exits but it is damm expensive vs Panther Lake also this new feature I found on the web 🤣