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.

 

[H433x0n]

posterity.

How about waiting until the 14th? If efficiency improves by 15% at 45W they’ve arguably won client mobile until Strix. If Intel gets MTL efficiency within margin of error of Phoenix they likely hold onto their market share for 2024.

They’ve got more volume, more resources for supporting OEMs and the ability to sell cheap lower end RPL dies.

 

[Abwx]

How about waiting until the 14th? If efficiency improves by 15% at 45W they’ve arguably won client mobile until Strix. If Intel gets MTL efficiency within margin of error of Phoenix they likely hold onto their market share for 2024.

They’ve got more volume, more resources for supporting OEMs and the ability to sell cheap lower end RPL dies.

The leak say 12k pts in CB R23 at 40W, same score as a 13700H@45W and a 7840U@30W, that fits in a +33% error margin, the 7840U@30W does exactly 12385 pts, so that put it at barely 28W for 12k pts.

 

[H433x0n]

The leak say 12k pts in CB R23 at 40W, same score as a 13700H@45W and a 7840U@30W, that fits in a +33% error margin, the 7840U@30W does exactly 12385 pts, so that put it at barely 28W for 12k pts.

Let’s wait until the 14th.

The 13700H is a 45W process being compared against a 28W MTL lower down the stack. It’s replacing the 1360P, not the 13700H.

 

[adroc_thurston]

https://twitter.com/x/status/1734409922710995348

yeah it's bad.
next.

 

[gdansk]

I'll wait a bit more. Laptops are all over the place based on cooling solutions and vendor-chosen power limits and boost times.
Plus that's a cutdown variant so maybe more cores will help.

 

[adroc_thurston]

Plus that's a cutdown variant so maybe more cores will help.

They're comparing the same tier SKUs with the exact same (well MTL has 2 more Atoms, they're on the SOC tile) core count.

 

[gdansk]

They're comparing the same tier SKUs with the exact same (well MTL has 2 more Atoms, they're on the SOC tile) core count.

So 2 more P cores and potentially 600MHz won't help relative to AMD?

 

[adroc_thurston]

So 2 more P cores and potentially 600MHz won't help relative to AMD?

The point it the thing being a perf regression over RPL.
ISO core count (-ish).
I haven't mentioned AMD anywhere, MTL ain't touching HWK1.

 

[H433x0n]

So 2 more cores won't help relative to AMD?

Doesn’t matter, every leak that’s not favorable will get spun to favor a narrative in the brand war™

 

[Khato]

https://twitter.com/x/status/1734409922710995348

yeah it's bad.
next.

What's bad exactly? The minor single core performance regression? Since in all other points the Ultra5-125H is showing better performance than the i5-13500H despite running well above its target TDP.

 

[adroc_thurston]

What's bad exactly?

It's a dieshrink tied to an overall a lot more expensive product that showcases a perf regression over the cheapness that is RPL.

 

[H433x0n]

What's bad exactly? The minor single core performance regression? Since in all other points the Ultra5-125H is showing better performance than the i5-13500H despite running well above its target TDP.

The iGPU performance looks good but even that I’m pretty skeptical of until we get proper data.

No clue why 65W was chosen either, at 65W the 13700H would appear to be close to Phoenix in efficiency and it’d be totally misleading. In this context it’s okay since it means you can dunk on MTL though.

 

[Saylick]

It's a dieshrink tied to an overall a lot more expensive product that showcases a perf regression over the cheapness that is RPL.

Seems like the iGPU will be its saving grace then.

 

[adroc_thurston]

Seems like the iGPU will be its saving grace then.

Not a particularly relevant one since Intel primarily lacks on CPU oomph/W and battery life and MTL solves kinda none of that.

 

[FlameTail]

So MLID was right about MTL efficiency?

 

[Philste]

So MLID was right about MTL efficiency?

Nobody knows of these are ES or final Silicon. Let alone the fact that 65W is probably too high to really compare the CPUs. MTL might bring 95% of that performance at 35W, where RPL only brings half of it. It's like Comparing Phoenix and Rembrandt at 95W. Both pretty much Max out at 65W and below, so at 95W the performance is roughly the same.

 

[Saylick]

Not a particularly relevant one since Intel primarily lacks on CPU oomph/W and battery life and MTL solves kinda none of that.

If so, I'm sure some one out there will highlight the dual encoders/decodes and cite it as a reason to buy MTL over HWK.

 

[Abwx]

MTL might bring 95% of that performance at 35W

That s impossible, 95% of the perf@65W is obtained in the 56-57W range, there s a curve posted by HXL, and it s apparently 95%@57W when looking at the slope from 28 to 65W.

 

[H433x0n]

That s impossible, 95% of the perf@65W is obtained in the 56-57W range, there s a curve posted by HXL, and it s apparently 95%@57W when looking at the slope from 28 to 65W.

The graph with no y axis and has no baseline comparison to 13500H.

 

[Khato]

It's a dieshrink tied to an overall a lot more expensive product that showcases a perf regression over the cheapness that is RPL.

Heh, and how much more expensive is it? I have my popcorn ready and waiting.

MTL regression on single thread performance versus RPL is indeed unfortunate. But last I checked RPL was already well on top in that particular metric anyway. It's efficiency (and by extension MT performance) where RPL is a bit behind what AMD has to offer. Where MTL lands in that pursuit? Well, hopefully you need only wait a few more days now. (Can at least hope that review samples are provided in advance of launch.)

 

[H433x0n]

MTL regression on single thread performance versus RPL is indeed unfortunate.

There’s MTL benchmark leaks that show the opposite with the Core 9 185H ST performance in line with expectations. In general ST performance will depend on clocks since IPC is relatively the same.

 

[Abwx]

The graph with no y axis and has no baseline comparison to 13500H.

The comparison with the 13500H is irrelevant in respect of what i said, wich is that this 4 + 8 need 56-57W to perform 95% of its 65W perf.

The graphs say that perf increase by 47.1% when power increase by 132%, that is , from 28W to 65W, from here you can extract the power/frequency curve in this segment :

Power = F^(ln(2.32)/ln(1.471)) = F^2.18

So at 0.95x the frequency power will be :

P = 65 x (0.95^2.18) = 65 x 0.894 = 58.1W

Obviously this chip was tailored to reach high TDPs, assuming this curve is accurate, wich is no surprise since current 4 + 8 can reach way higher powers than 65W.

 

[igor_kavinski]

Intel Ultra 5 125H and Ryzen 7 7840HS tested at 65W: Intel with higher iGPU score but falls short in CPU tests - VideoCardz.com

Intel Core Ultra 5 125H at 65W shows lower CPU score and higher iGPU score than AMD Ryzen 7 7840HS Intel’s upcoming Core Ultra 100H series is just around the corner. Intel is going to show off its new laptop series called Meteor Lake in two days. This will introduce a new name for the […]

videocardz.com

Seems Intel is back in the game, at least for the iGPU.

One primary reason for getting a Radeon laptop just went whoosh!

My condolences, @adroc_thurston

 

[FlameTail]

One primary reason for getting a Radeon laptop just went whoosh!

yeah screw AMD with their good APUs being barely available and mostly relegated to high end gaming laptops which have a dGPU anyways.

Intel will be our saviour thanks to their immense volume and influence.

Obligatory /s

 

[gdansk]

Intel Ultra 5 125H and Ryzen 7 7840HS tested at 65W: Intel with higher iGPU score but falls short in CPU tests - VideoCardz.com

Intel Core Ultra 5 125H at 65W shows lower CPU score and higher iGPU score than AMD Ryzen 7 7840HS Intel’s upcoming Core Ultra 100H series is just around the corner. Intel is going to show off its new laptop series called Meteor Lake in two days. This will introduce a new name for the […]

videocardz.com

Seems Intel is back in the game, at least for the iGPU.

One primary reason for getting a Radeon laptop just went whoosh!

My condolences, @adroc_thurston

View attachment 90294

If you scroll up a bit, he already posted this and says it means MTL is a flop.
In either case it's probably best to wait for testing at MTL's target power limits.
And if it's like other Intel GPUs it'll kill in benchmarks then flail around in actual games.