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.

 

[adroc_thurston]

As for GNR, are you seriously claiming that ~128c wouldn't even beat Genoa?

120 and yea, there's more to server chips than just raw core counts.
AmpereOne is 192c, but lol. Lmao even.
AMD makes better cores than anyone else in the industry and you will learn it the hard way!

If you have numbers for both, than post them

Well you've seen the GNR perf target from BHS-AP platform guide (it's 2x EMR which is lol/lmao), and Turin idk if I can say the numbers yet, but a lot faster than that even.

 

[Exist50]

120 and yea, there's more to server chips than just raw core counts.
AmpereOne is 192c, but lol. Lmao even.
AMD makes better cores than anyone else in the industry and you will learn it the hard way!

Well you've seen the GNR perf target from BHS-AP platform guide (it's 2x EMR which is lol/lmao), and Turin idk if I can say the numbers yet, but a lot faster than that even.

2x EMR would very soundly beat Genoa. And yes, there's more than core count, hence my comments about the Zen 5 hype train. Anyway, we shall see.

 

[jpiniero]

2x EMR would very soundly beat Genoa. And yes, there's more than core count, hence my comments about the Zen 5 hype train. Anyway, we shall see.

Has Intel actually followed through in selling any of the AP products that have been talked about?

 

[Exist50]

Has Intel actually followed through in selling any of the AP products that have been talked about?

Cascade Lake AP was a joke, but GNR-AP actually makes sense. No two CPUs in one socket nonsense. Just 3 compute dies / 12 memory channels instead of 2 dies / 8 channels. See no reason to doubt that'll reach the market.

 

[uzzi38]

2x EMR would very soundly beat Genoa. And yes, there's more than core count, hence my comments about the Zen 5 hype train. Anyway, we shall see.

"There's more than just core count" is exactly why Turin rumours exceeded my expectations.

Well, I don't think that GNR-AP would lose to or be around Genoa level (it better not given it's also a 500w part like Turin stuff), but I'm also pretty confident it won't touch Turin but rather sit somewhere in the middle of Genoa and Turin instead.

 

[Exist50]

"There's more than just core count" is exactly why Turin rumours exceeded my expectations.

Well, I don't think that GNR-AP would lose to or be around Genoa level (it better not given it's also a 500w part like Turin stuff), but I'm also pretty confident it won't touch Turin but rather sit somewhere in the middle of Genoa and Turin instead.

I'd guess a roughly 10-20% gap thanks to Zen 5 vs RWC+. But hopefully we start seeing leaks for real silicon start trickling out soon. Both should probably ship around a year from now.

 

[eek2121]

Looks like Intel is discontinuing NUC. Sad.

Hopefully partners pick up the slack. I would purchase an Arrow Lake based NUC if it had good perf/watt.

There will be Ultra 9 variants.

你也改名？2分钟看懂全新酷睿产品命名规则_哔哩哔哩_bilibili

你也改名？2分钟看懂全新酷睿产品命名规则, 视频播放量 87830、弹幕量 159、点赞数 2085、投硬币枚数 154、收藏人数 569、转发人数 537, 视频作者 金猪升级包, 作者简介 既不独立，也不客观，更不公正，相关视频：英特尔14代酷睿重大变革：最高频率超6.2GHz，i9/i7/i5更名为Ultra9/Ultra7/Ultra5【宅同学】，全网最全电脑CPU性能排行榜！！！，英特尔Ultra系列和i系列会有什么差异？，英特尔新一代CPU又改名了？玩游戏到底选Ultra还是14代i9?，Intel Ultra 7 265K对比i7-14700KF 评测...

www.bilibili.com

My understanding was that is more about the naming than actual SKUs coming out. I hope to be proven wrong, however.

View attachment 82884

Seems like QS sample can only hit up to 4.8GHz for P core and 2.2GHz for tGPU, hopefully Ultra 9 would turbo up to 5GHz....

You know this is a mobile (low power) chip, right?

I'm curious if the 65W number listed is PL2 or not. If so, should be interesting.

 

[adroc_thurston]

2x EMR would very soundly beat Genoa

lol no, EMR is barely-barely faster than SPR, just with less horrendous SNC1 penalties.

Anyway, we shall see.

Not much to see here, the hyperscale lined up for Turin (well, mostly -Dense) already.

I'd guess a roughly 10-20% gap thanks to Zen 5 vs RWC+

A lot, lot, lot and lot bigger.

Just 3 compute dies / 12 memory channels instead of 2 dies / 8 channels

"let's make an already big mesh, worse" is not exactly what Intel needs period but i digress

 

[H433x0n]

120 and yea, there's more to server chips than just raw core counts.
AMD makes better cores than anyone else in the industry and you will learn it the hard way!

When you say GNR-AP won’t outperform Genoa, what metric are you going by?

 

[Exist50]

lol no, EMR is barely-barely faster than SPR, just with less horrendous SNC1 penalties.

It's probably a bigger bump than you'd think. The Raptor Cove and L3 changes alone should be significant. It's coming too late to matter, but it should at least be double digit over SPR.

A lot, lot, lot and lot bigger.

If you're expecting far more than that out of Zen 5, you'll be disappointed.

 

[Abwx]

If you're expecting far more than that out of Zen 5, you'll be disappointed.

Perf/core is still unknown but SKUs will be 192-256 cores, dunno what Intel will use to face this massive increasement over current gen.

 

[Geddagod]

I knew my 200% IPC uplift over Zen 4 for Zen 5 estimate was accurate, and now it's been confirmed by reputable AMD leaker adroc_thurston. Thanks Anandtech!

 

[Geddagod]

Perf/core is still unknown but SKUs will be 192-256 cores, dunno what Intel will use to face this massive increasement over current gen.

It's 128 cores for the standard Zen 5 version

 

[Geddagod]

Intel Meteor Lake Xe-LPG iGPU reportedly boosts up to 2.2 GHz, slightly more compute power than Radeon 780M - VideoCardz.com

Intel Meteor Lake Xe-LPG graphics currently boost up to 2.2 GHz The competition in integrated graphics is heating up. Intel is in the final stages of preparing its Meteor Lake chips for launch later this year. The upcoming mobile Core Ultra series will introduce an entirely new architecture...

videocardz.com

Don't know if this has been posted yet, but iGPU clocks are nice. A little disappointed in the ST max frequency clock based on previous rumors that indicate it would reach >5GHz, but who knows maybe the QS isn't the -H series or something (though that might just be copium on my part lmao).
I think a ST regression is becoming more and more likely looking at clocks, but for the overall laptop, if they manage to reduce CPU power consumption during gaming, overall game fps might increase if they funnel some extra power over to the dGPU.

 

[adroc_thurston]

It's probably a bigger bump than you'd think. The Raptor Cove and L3 changes alone should be significant.

They're minor, I know EMR socket perf too.
It's a nothingburger that just makes SPR less actively miserable.

If you're expecting far more than that out of Zen 5, you'll be disappointed.

I literally know the numbers lol.

Perf/core is still unknown but SKUs will be 192-256 cores

Turin is 'just' 128c, Turin-D is 192c though.

I knew my 200% IPC uplift over Zen 4 for Zen 5 estimate was accurate, and now it's been confirmed by reputable AMD leaker adroc_thurston. Thanks Anandtech!

No need to get upset.
Zen5 really is the biggest jump since the OG Zen.

 

[Abwx]

It's 128 cores for the standard Zen 5 version

They wouldnt need to increase TDP up to 500-600W for the same core count as current products, moreover when using a smaller node...

 

[Exist50]

I literally know the numbers lol.

Ok, then why not post them so we can reference later? It's not like you have an existing track record in this forum.

 

[adroc_thurston]

They wouldnt need to increase TDP up to 500-600W for the same core count as current products, moreover when using a smaller node...

N4p is a nodelet that's barely an improvement over N5p and you gotta understand what those cores really are.

Ok, then why not post them so we can reference later?

Because people would scream at me.

It's not like you have an existing track record in this forum.

Yea cuz I'm a twitter tourist.

 

[Exist50]

Because people would scream at me.

Whatever number you post, I can assure you that someone here has already made a far worse prediction. Not to even mention Twitter...

And if you're right, it'll help take the rest of your claims and other predictions more seriously in the future.

 

[Abwx]

N4p is a nodelet that's barely an improvement over N5p and you gotta understand what those cores really are.

They still woudnt need to almost double the TDP...

 

[Geddagod]

No need to get upset.
Zen5 really is the biggest jump since the OG Zen.

I'm just trolling, but tbh GNR barely being competitive with Genoa? Bit of a wild claim, esp without a track record to back it up. And the Zen 5 hype sounds surreal, but ig we will see ¯\_(ツ)_/¯

 

[adroc_thurston]

Whatever number you post, I can assure you that someone here has already made a far worse prediction. Not to even mention Twitter...

People that post here know me, I ain't ever doing that.
Focus on client parts where Intel has a lot, lot more chances than in DC.

They still woudnt need to almost double the TDP...

It's 33% more cores with a wee bit more uncore and those cores are truely gigantic.
Without a real shrink it all gets rather taxing.

but tbh GNR barely being competitive with Genoa?

Well, yea, SPR is ~Milan, and Genoa is a heck lot faster than Milan.
Those are all SPECint rate quotes so doesn't apply to all workloads but matters still.

And the Zen 5 hype sounds surreal

The silicon sampled, there's no hype left.
It is what it is.

 

[H433x0n]

It's probably a bigger bump than you'd think. The Raptor Cove and L3 changes alone should be significant. It's coming too late to matter, but it should at least be double digit over SPR.

Glad to see I wasn’t fed pie-in-the-sky optimism, I was told EMR was a noticeable bump over SPR but I was afraid to post it for fear of being ridiculed.

I was quoted roughly >=20% at the same power envelope since there will be room for more clocks with some better power management and bloat removed.

 

[Geddagod]

at least double digits is a bit far from ~20%.
Isn't ~20% from MLID?
Wonder if EMR fixes L3 bandwidth problems SPR faces

 

[adroc_thurston]

at least double digits is a bit far from ~20%.

yea lol.

Wonder if EMR fixes L3 bandwidth problems SPR faces

No.
Bigger L3 but there are issues (kinda obvious ones really).

there will be room for more clocks with some better power management and bloat removed.

Mesh clock is literally going down so lol. lmao.