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.

 

[Geddagod]

Anandtech reviews.

I don't even think Anandtech showed that it was only on par with Skylake.

But yes, that was an approximation. It's a little better than Skylake but not as high as in the chart you posted in actual applications. Zen 3 would fit in between Sunny Cove and Skylake.

Why not, and which "actual applications"?

 

[poke01]

What’s the size of Intels P core in PTL?

Excluding cache.

 

[ToTTenTranz]

According to the internet and reviewers, CPU performance doesn't matter now.

What does ST CPU performance above e.g. Zen4 Phoenix matter on a thin&light laptop matter for, realistically?


On the system-wide TDP of such laptops (sub-100W?), games are almost always going to be GPU-limited. It's not like that CPU is pushing back a RTX 5090 32GB.

Are office apps going to feel snappier between a Ryzen 7 7840U and a Ryzen 9 HX470? Is web browsing going to load and scroll noticeably faster?


For the overwhelming majority of people, performance of a 6core Zen4 is more than enough. And from there, what will differentiate from other options is heat output and battery life during load, idle and sleep.
And then there's a subset that cares about iGPU performance, or dGPU performance if there's one present.

That's it.

 

[jpiniero]

What does ST CPU performance above e.g. Zen4 Phoenix matter on a thin&light laptop matter for, realistically?

Microsoft Microsofting.

 

[poke01]

No, according to reviewers a laptop is the more than just raw performance. It's a laptop, not a desktop. And so plenty of other factors are considered, like battery life, chassis temperatures, and cooling noise. All of which saw improvements.

And this is true. This is how a large portion of the laptop buying public judges a laptop, myself included.

Really? So I guess you didn’t buy any Coffelake, Ice lake, Tiger Lake, Arrow Lake, Raptor lake, Meteor Lake laptops right?

Cause those above architectures sucked at battery life, temps etc and were horrible to use in laptops.

 

[Hulk]

I don't even think Anandtech showed that it was only on par with Skylake.

Why not, and which "actual applications"?

Ooooh!! It's so hard to back down, but I checked my data and you are right!

Zen 3 is closer to Cypress Cove. Zen original is more like Skylake. I somehow confused Zen and Zen 3 in my head and didn't revisit my data.

Thanks for calling me out. Doesn't feel good to be wrong, but truth is truth.

I'll correct the chart later after the sting of this humiliation has subsided...

 

[DavidC1]

I'll correct the chart later after the sting of this humiliation has subsided...

You are funny so that's ok.

Really? So I guess you didn’t buy any Coffelake, Ice lake, Tiger Lake, Arrow Lake, Raptor lake, Meteor Lake laptops right?

Cause those above architectures sucked at battery life, temps etc and were horrible to use in laptops.

Eh, Meteorlake is no worse competitively. Sure it's behind ARM but only Lunar and Panther can get similar so in the x86 sense it wasn't too bad. Also the 14nm generation and Icelake did ok too.

No, according to reviewers a laptop is the more than just raw performance. It's a laptop, not a desktop. And so plenty of other factors are considered, like battery life, chassis temperatures, and cooling noise. All of which saw improvements.

And this is true. This is how a large portion of the laptop buying public judges a laptop, myself included.

Yup. I don't think the raw performance is what holds back laptop performance, at least for the responsiveness metric which most people care about, but the power management. My desktop became annoyingly slower with the power C state on required to get the CPU idle from 8W to 2W. It just felt slower in everything. A laptop, even the latest fancy uarch will feel slower. Because it needs to ramp up from the LFM 800MHz frequency, or wake it up from the deep C state.

Where are the sub-$1000 Pantherlakes? This is supposed to be a low-power SoC that doesn't even operate above 25W. It's like Lunar Lake all over again. Also how is Gorgon Point relevant when Pantherlake is competing with the now-quite-old Strix Point?

Because Gorgon Point is also 2026 mainstream.

It may be both of the client market is doomed this year because of AI slop pushing prices. You get your talking dog video for the price of 4x RAM and SSD prices plus 40% more utility bills. Yay!

 

[Hulk]

Update. Geddagod "reminded" me (schooled me appropriately) that Zen 3 is not like Skylake for ST, but more like Sunny Cove.
Here's the updated chart.
Zen 2/3 don't fit in cleanly. Zen 2 is perhaps a bit under Sunny Cove, Zen 3 a bit better.

 

[mvprod123]

https://twitter.com/x/status/2017390286666928298

 

[OneEng2]

This is always subjective and controversial but I've been tracking benches and reviews for many years.
Here is how I see single thread throughput rates for various cores based on the data I have collected.
This is what we refer to as IPC, or throughput per clock cycle.


View attachment 137631

Single thread, not single core? Nice graph (saw the update as well). I still wonder how things look with real world application benchmarks vs synthetic. It just seems more and more that the synthetic doesn't translate into performance in every day use (of course, one could argue that few would notice anyway).

No, according to reviewers a laptop is the more than just raw performance. It's a laptop, not a desktop. And so plenty of other factors are considered, like battery life, chassis temperatures, and cooling noise. All of which saw improvements.

And this is true. This is how a large portion of the laptop buying public judges a laptop, myself included.

I recently bought a new laptop for my Wife for Christmas. Got a Lenovo Yoga 7 AMD Ryzen 7 350 with an OLED screen.

This was purchased based on 3 selling points. 1) The OLED is BEAUTIFUL (hard to get your wife past this one guys. Go give it a try some time!). 2) Battery life, 3) (your gona love this) NPU "Copilot" rated.

Seriously, performance didn't really come into the picture much. Figured this was more than she would ever need, it had pretty good battery life, a gorgeous OLED screen and she is now best buddies with Copilot (go figure).

​

What’s the size of Intels P core in PTL?

Excluding cache.

This is a great question really. In theory 18A has ever advantage in density over everything AMD is selling on N4P right now. This time size really does matter when it comes to how "good" the core design is. I already suspect that 18A is a very expensive process for Intel to make products on. A larger design would make matters even worse in this regard.

It's not that I pick on Intel for this either. I wonder the same about the Apple processors. The products that Apple makes with them seem to be pretty expensive (and right now, so is PTL products).

 

[Hulk]

Here is some of the data I've collected over the years to create that bar graph.

 

[MoistOintment]

Really? So I guess you didn’t buy any Coffelake, Ice lake, Tiger Lake, Arrow Lake, Raptor lake, Meteor Lake laptops right?

Cause those above architectures sucked at battery life, temps etc and were horrible to use in laptops.

I know they sucked for the most part. So skipped them and went from Ice Lake -> Lunar Lake.

This was purchased based on 3 selling points. 1) The OLED is BEAUTIFUL (hard to get your wife past this one guys. Go give it a try some time!). 2) Battery life, 3) (your gona love this) NPU "Copilot" rated.

Seriously, performance didn't really come into the picture much. Figured this was more than she would ever need, it had pretty good battery life, a gorgeous OLED screen and she is now best buddies with Copilot (go figure).

It's like this in every tech forum: People are just completely disconnected from the general consumer market to a questionable level. Give the typical laptop buyer a Macbook Air and they're going to gush over how silent and cool it is. In a world where smart phones have all day battery and amazing screens, giving the average person a chunky laptop that has audible fan noise and gets warm to the touch from just browsing the web, and needs to constantly be plugged in to charge feels like they went back in time 10 years. Most could care less about benchmark scores - they probably never even heard of cinebench or geekbench.

 

[Hulk]

It's like this in every tech forum: People are just completely disconnected from the general consumer market to a questionable level. Give the typical laptop buyer a Macbook Air and they're going to gush over how silent and cool it is. In a world where smart phones have all day battery and amazing screens, giving the average person a chunky laptop that has audible fan noise and gets warm to the touch from just browsing the web, and needs to constantly be plugged in to charge feels like they went back in time 10 years. Most could care less about benchmark scores - they probably never even heard of cinebench or geekbench.

Most people have no idea what so ever what is going on under the hood. They turn it on, browse the web, send e-mails, and maybe do some Word. Done.

We are enthusiasts because we are enthused about the technology and tend to do a lot more than the average user with our systems. But yes it is easy to conflate our opinions with people that just "need a vacuum to suck."

 

[RTX]

https://twitter.com/x/status/2017345200562311617

Is Cougar cove not area efficient compared to Apple M5?

 

[Thibsie]

lol SiliconFly BS 🤣

 

[511]

Intel P cores are not that good in terms of Area they are the worst of the P cores available.

 

[Saylick]

lol SiliconFly BS 🤣

https://twitter.com/x/status/2017367523516358873

How convenient! The goal post moves.

 

[RTX]

https://twitter.com/x/status/2017332916435750933

https://twitter.com/x/status/2017364714876014991

Is he coping or is the comparison valid?

 

[511]

Apple using less cache what? Apple uses tons of cache if anything lol

 

[DavidC1]

Apple using less cache what? Apple uses tons of cache if anything lol

Cache is also very power and area efficient, while also being very repetitive, meaning high yield. There's nothing else more repetitive than cache. That's why the die pictures are so clean for the cache part.

So I read a good rule of thumb when adding new uarch features is: "Does it beat cache power/area wise?" Otherwise just increase cache sizes instead.

Nova Lake is expected to the performance of both the CPU & iGPU by a wide margin.

Yea, based on the Novalake desktop leak of what being little over 10% faster than Arrowlake is not convincing.

This was purchased based on 3 selling points. 1) The OLED is BEAUTIFUL (hard to get your wife past this one guys. Go give it a try some time!). 2) Battery life, 3) (your gona love this) NPU "Copilot" rated.

The average people can also be quite ignorant. There needs to be more of a middle ground. OLEDs are ticking time bomb consumables with their burn in.

I also realized how vapid we can be. My Dell XPS 12 screen was BEAUTIFUL. It all lasted 1 month. After that I didn't think it was anything special at all. While you will notice your battery life, and the fan noise.

 

[Geddagod]

Apple using less cache what? Apple uses tons of cache if anything lol

Intel spends 1.5x more cache, capacity wise, counting from L0 to L3 than Apple does counting from L1 to SL2 in a 4x P-core cluster.

Cache is also very power and area efficient,

I think it's buns for area ngl

 

[poke01]

Here is a comparison 358H P core vs M5 P core:

Panther lake
L0D: 48KBx4
L0I: 64KBx4
L1: 192KBx4
L2: 3MBx4
L3S: 18MB

Also there is that 8MB Memory side cache.

M5
L1D: 128KBx4
L1I: 192KBx4
L2 shared: 8MB per cluster.
L3S: 8MB? M4 was 8MB

Intel spends 1.5x more cache, capacity wise, counting from L0 to L3 than Apple does counting from L1 to SL2 in a 4x P-core cluster.

So yeah Intel does use more cache over all.

 

[Geddagod]

Here is a comparison 358H P core vs M5 P core:

Panther lake
L0D: 48KBx4
L0I: 64KBx4
L1: 192KBx4
L2: 3MBx4
L3S: 18MB

Also there is that 8MB Memory side cache.

M5
L1D: 128KBx4
L1I: 192KBx4
L2 shared: 8MB per cluster.
L3S: 8MB? M4 was 8MB


So yeah Intel does use more cache over all.

I was equalizing it from ARL iso core count (4x vs 4x), but this works too.
Also SLC is slow as fk and pretty much every company that has a SLC (Qcomm + Intel) says it's not mainly about the CPU.

 

[fastandfurious6]

Re: The Strix Halo vs Panther Lake pricing debate, I will say that at least for gaming, SKUs like below blow both of them out of the water for the money. This is a real time availability ad from Best Buy US. PL and Halo both need to be available at significantly sub $2k to even be taken seriously in the gaming conversation. Also, from Amazon, it appears that available Halo laptops are basically non-existent, so theres that too. Halo is indeed a non-competitor unless AMD does something that drastically changes availability.

View attachment 137624

100% been calling it for a whole year now long before release 😉 9955HX laptops are the holy grail

it's a literal 9950X on laptop runs everything AAA silent and cool 1080p 120fps but still many say "gaming laptops are loud hot expensive" 😂

Even better combo, same price, 5070ti (12GB) + 7945HX: https://www.bestbuy.com/product/msi...tx-5070ti32gb-ram1tb-ssdwin11-gray/J3P7T266XV In China it's around $1500. US more expensive at $2100

PTL is just way too expensive and too weak for its price. Same for Strix/Gorgon

 

[511]

Here is a comparison 358H P core vs M5 P core:

Panther lake
L0D: 48KBx4
L0I: 64KBx4
L1: 192KBx4
L2: 3MBx4
L3S: 18MB

Also there is that 8MB Memory side cache.

M5
L1D: 128KBx4
L1I: 192KBx4
L2 shared: 8MB per cluster.
L3S: 8MB? M4 was 8MB


So yeah Intel does use more cache over all.

You are missing E cores as well cause the 8E also have access to 18M L3 it's not for P core alone for 4 P core it would be 3 MB/core