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.

 

[dullard]

What are you talking about?

ARL-U 2+8 Intel 3 - rumored MTL refresh
ARL-H 6+8 N3B real Arrow Lake

If only there were a series of pictures in the OP that answered the question about what else is in 2024. Something like big blue photos that are on the top of every page of this thread forcing people to scroll down a lot to read content so they are hard to miss. Then if only there were tables in that same post listing core counts. But alas...

 

[dullard]

I'm not even sure Lunar lake excitement is entirely sensible , What is the major drawcard for it? What is intels solution for above 4+4c in 2024?

To answer your question directly, Lunar Lake is NOT for everyone. It is a niche product. But for that niche, it should be quite powerful. Anything that needs strong graphics, good NPU, but not a number crunching workstation. Its intended market is the ultralight laptop / tablet / handheld areas. It'll likely go into 2-in-1s similar to the Lenovo X12 or Dell Latitude 7350. You can read their marketing blurb under the Ultralight Upgrade and sections below it here (so I don't have to repeat drawcards that have already been written): https://www.dell.com/en-us/shop/del...-or-2-in-1/spd/latitude-13-7350-2-in-1-laptop It will also go into gaming handhelds like the MSI Claw.

Lunar Lake would also be great for HTPC, kiosks, or fanless designs (industrial uses like defect detection on assembly lines and conveyor belts). But, it is possible that may be more of a Panther Lake thing since Lunar Lake is just a temporary blip on the way to the much longer lasting Panther Lake.

Beyond 4+4, you get Arrow Lake in October 2024 (up to 8+16). Then if you extend your timeframe a bit, it looks like Bartlett Lake is Jan 2025 (also 8+16).

 

[coercitiv]

Something like big blue photos that are on the top of every page of this thread forcing people to scroll down a lot to read content so they are hard to miss.

Not my problem if you haven't yet figured out how to avoid seeing the OP every time you enter the thread. Not that the OP contains the actual info that was asked, but anyway. I'm surprised how tense both you and @mikk must be in order to immediately berate someone for writing an inaccurate post that clearly began with "IIRC". I've been doing my best to stay out of both your and @mikk's way, but apparently it ain't enough. No time for manners from the knights of the holy order.

 

[dullard]

Not my problem if you haven't yet figured out how to avoid seeing the OP every time you enter the thread. Not that the OP contains the actual info that was asked, but anyway. I'm surprised how tense both you and @mikk must be in order to immediately berate someone for writing an inaccurate post that clearly began with "IIRC". I've been doing my best to stay out of both your and @mikk's way, but apparently it ain't enough. No time for manners from the knights of the holy order.

The question that was asked includes: "What is intels solution for above 4+4c in 2024?"

Lets see what the OP includes: Oh yeah, information on CPUs in 2024, including the core counts. So, you think that does not contain actual information that was asked?



No need to go into personal attacks here. Facts only please. Fact is that you were incorrect (now twice in two consecutive posts).

 

[ondma]

To answer your question directly, Lunar Lake is NOT for everyone. It is a niche product. But for that niche, it should be quite powerful. Anything that needs strong graphics, good NPU, but not a number crunching workstation. Its intended market is the ultralight laptop / tablet / handheld areas. It'll likely go into 2-in-1s similar to the Lenovo X12 or Dell Latitude 7350. You can read their marketing blurb under the Ultralight Upgrade and sections below it here (so I don't have to repeat drawcards that have already been written): https://www.dell.com/en-us/shop/del...-or-2-in-1/spd/latitude-13-7350-2-in-1-laptop It will also go into gaming handhelds like the MSI Claw.

Lunar Lake would also be great for HTPC, kiosks, or fanless designs (industrial uses like defect detection on assembly lines and conveyor belts). But, it is possible that may be more of a Panther Lake thing since Lunar Lake is just a temporary blip on the way to the much longer lasting Panther Lake.

Beyond 4+4, you get Arrow Lake in October 2024 (up to 8+16). Then if you extend your timeframe a bit, it looks like Bartlett Lake is Jan 2025 (also 8+16).

Yea, desktop looks great.
1. ARL in October that barely moves the performance meter, especially in ST (and this after 2 refreshes of Alder Lake, one decent, and one a joke)
2. Current flagship products from the last 2 generations that have serious stability issues that Intel seems unable to solve.
3. A "new" product with 2 year old architecture and even older process tech in 2025, that may or may not still have the stability issues
4. A refresh of already underwhelming ARL also in 2025.
5. Meteor Lake and Panther Lake both failed to make desktop at all.
6. Still tied to TSMC silicon, despite all the supposed foundry progress.
What a cluster you know what. Absolutely terrible execution. But wait, there is still one more mirage to chase: 18A is supposed to solve everything.

 

[dullard]

Yea, desktop looks great.
1. ARL in October that barely moves the performance meter, especially in ST (and this after 2 refreshes of Alder Lake, one decent, and one a joke)
2. Current flagship products from the last 2 generations that have serious stability issues that Intel seems unable to solve.
3. A "new" product with 2 year old architecture and even older process tech in 2025, that may or may not still have the stability issues
4. A refresh of already underwhelming ARL also in 2025.
5. Meteor Lake and Panther Lake both failed to make desktop at all.
6. Still tied to TSMC silicon, despite all the supposed foundry progress.
What a cluster you know what. Absolutely terrible execution. But wait, there is still one more mirage to chase: 18A is supposed to solve everything.

Nice list of grievances. Mind if I modify it?

1. ARL in October that barely moves the performance meter, especially in ST (and this after 2 refreshes of Alder Lake, one decent, and one a joke)
1a. Lets be complete in the description. Doubles iGPU, adds NPU, and addresses power efficiency that was everyone's biggest complaint about Intel's products.

2. Current flagship products from the last 2 generations that have serious stability issues that Intel seems unable to solve.
Yes, this is a serious problem that must be addressed. Intel must find all root problems and solve them.

3. A "new" product with 2 year old architecture and even older process tech in 2025, that may or may not still have the stability issues
All comes down to performance / price. Intel is going to a new model where the lower end chips are not produced on the same node as the high end chips. Will they adjust the price to compensate? Or not?

4. A refresh of already underwhelming ARL also in 2025.
Not enough information for us to speculate yet. What has been rumored has also been denied by rumors.

5. Meteor Lake and Panther Lake both failed to make desktop at all.
Fairly irrelevant. No need for one marketing name to be applied to both mobile and desktop. But, if that is really what keeps you up at night, go ahead and worry.

6. Still tied to TSMC silicon, despite all the supposed foundry progress.
Also irrelevant. Using TSMC silicon does not make a CPU good or bad. Intel does not have the capacity. No one but TSMC has the capacity.

 

[ondma]

Nice list of grievances. Mind if I modify it?

1. ARL in October that barely moves the performance meter, especially in ST (and this after 2 refreshes of Alder Lake, one decent, and one a joke)
1a. Lets be complete in the description. Doubles iGPU, adds NPU, and addresses power efficiency that was everyone's biggest complaint about Intel's products.

2. Current flagship products from the last 2 generations that have serious stability issues that Intel seems unable to solve.
Yes, this is a serious problem that must be addressed. Intel must find all root problems and solve them.

3. A "new" product with 2 year old architecture and even older process tech in 2025, that may or may not still have the stability issues
All comes down to performance / price. Intel is going to a new model where the lower end chips are not produced on the same node as the high end chips. Will they adjust the price to compensate? Or not?

4. A refresh of already underwhelming ARL also in 2025.
Not enough information for us to speculate yet. What has been rumored has also been denied by rumors.

5. Meteor Lake and Panther Lake both failed to make desktop at all.
Fairly irrelevant. No need for one marketing name to be applied to both mobile and desktop. But, if that is really what keeps you up at night, go ahead and worry.

6. Still tied to TSMC silicon, despite all the supposed foundry progress.
Also irrelevant. Using TSMC silicon does not make a CPU good or bad. Intel does not have the capacity. No one but TSMC has the capacity.

1. Yes, and how many really care about igpu and npu on the desktop. I certainly dont, and I bet a large majority of those in this forum dont either. Igpu is pointless for high end gaming or productivity; just add a discrete gpu. NPU *might* be useful at some point, but right now, I dont care about that on the desktop either, and again, a discrete gpu can offer much superior AI processing ability. Power efficiency is still an unknown, but hopefully there will be significant improvements. However, I still expect Intel to be significantly behind AMD.
2. agree
3. Maybe Bartlett Lake will be priced appropriately, but will still have the power issues.
4. I will with serious reservation, grant you we dont for sure know what will come after ARL. But everything I have read recently says it will be a refresh. Since they seem to be branding it as a "refresh" I dont hold out hope for any significant improvement.
5. Agreed, the name doesnt matter, but IMO, they need a quick follow up to ARL with serious performance improvements.
6. Agreed the final performance is what matters. But you are arguing against yourself in that n3b or whatever is on ARL doesnt seem to give much performance gain; power use is yet to be determined, but also doesnt seem outstanding. As for capacity, that excuse maybe was ok a year or two ago, but again, it is always the mirage that always seems to disappear when you get there that Intel is going to have equal or superior process tech. And even if they do, for crying out loud, Intel is supposed to become a foundry company. What does it say about that if you cant even produce enough product to satisfy your own demand?

 

[dullard]

1. Yes, and how many really care about igpu and npu on the desktop. I certainly dont, and I bet a large majority of those in this forum dont either. Igpu is pointless for high end gaming or productivity; just add a discrete gpu. NPU *might* be useful at some point, but right now, I dont care about that on the desktop either, and again, a discrete gpu can offer much superior AI processing ability. Power efficiency is still an unknown, but hopefully there will be significant improvements. However, I still expect Intel to be significantly behind AMD.

About 2/3rds of computers use iGPU. I don't think we should just exclude the majority of computers because you personally don't care. I personally care since the company I work for will only buy SFF computers with no ability to install a discrete GPU. And Solidworks chugs on iGPUs--I'll take any boost that I can get. Yes, discrete GPUs are more capable. But there goes any money savings of one CPU over most others, and the few watts of power we bicker about on these forums between CPUs are destroyed once you have to power a separate GPU.

6. Agreed the final performance is what matters. But you are arguing against yourself in that n3b or whatever is on ARL doesnt seem to give much performance gain; power use is yet to be determined, but also doesnt seem outstanding. As for capacity, that excuse maybe was ok a year or two ago, but again, it is always the mirage that always seems to disappear when you get there that Intel is going to have equal or superior process tech. And even if they do, for crying out loud, Intel is supposed to become a foundry company. What does it say about that if you cant even produce enough product to satisfy your own demand?

It says Intel didn't buy enough equipment. Plain and simple. That says nothing about how good or bad the nodes are. It says nothing about how good or bad the desktop CPUs are (which is what I thought you were wanting to talk about). And that says nothing about the future when Intel really wants to be a foundry (which from the announcements really seems to be 18A and beyond). All of Intel's eggs are in that basket. 2026 is going to be a make or break year for their foundry ambitions. If it doesn't work out, they might have to be like Apple, AMD, or ARM: design of chips only.

 

[DavidC1]

It did seem like he was doing a lot of forward looking to Panther lake during Computex

I'm not even sure Lunar lake excitement is entirely sensible , What is the major drawcard for it? What is intels solution for above 4+4c in 2024?

What's not exciting about a x86 laptop platform that has a chance to meet, no even exceed ARM platforms in the *dare I say it!* battery life department!

It is undoubtedly the #1 reason why Apple Mx laptops gained market share despite the compatibility issues you have to go through(no matter how good the translation may be). It's undoubtedly the reason why MS and other manufacturers are jumping blindly into the Snapdragon Elite "AIPC" bandwagon. Recent surveys and reviews have said it isn't "AIPC" that people care about for Elite, but battery life.

@dullard Please. The inclusion of iGPU on Arrowlake is enough. If they deemed iGPUs so important, we'd have had multiple successors to 5775C, yet we do not. They only care in the context of "do we get a better $0 iGPU"? Arrowlake does have a better iGPU, but it's still Alchemist-based 64EUs, or a generation behind times 1/2, nevermind a 5775C-class. This is basically the CPU equivalent of Intel throwing a bone. "Here you go!"

AMD also knows this with the much cut down GPU on Zen 4.

Most of us are talking about bleeding edge for desktops, not a $150 chip. Cause it's not exciting. I'm happy that you are excited for it, but you'll find most don't. Stick a 5 year old RX 480 and it beats the 2025 Arrowlake iGPU by multiples, without needing to spend hundreds of dollars on a new platform. That's how you actually save money not the "oh I'll save WATTS with a new iGPU in this new platform which I had to swap out the CPU, mobo, and RAM for!"

I wish they would get a chip out where it's a combo of $150 CPU with a $100 iGPU, but they never will, because they figure putting the top end iGPU with the fastest CPU is the way they want to upsell.

 

[DavidC1]

It says Intel didn't buy enough equipment. Plain and simple. That says nothing about how good or bad the nodes are.

The 5N4Y wasn't going to come for free. Intel 4 is only used in one product, and 20A even less so. 18A will solve it for most of the markets, but not entirely so. Their greatest weakness has been on servers, and products like Clearwater Forest will put an end to that. But it took time.

The desktops have been strength for them since Alderlake. The year of ADL intro was when they gained marketshare in desktops, while losing significantly on mobile.

We'll see whether this "Arrowlake Refresh" comes true. Who knows, it might actually be decent. Raptorlake was decent was it not? It could also suck but comparatively it might not be. How long have people believed Zen 5 would absolutely dominate and now it seems at least performance wise they'll be similar? If it takes AMD 22 months for Zen 6 then it'll basically be Zen 5 vs ARL-R.

 

[jpiniero]

It is undoubtedly the #1 reason why Apple Mx laptops gained market share

I'd say the #1 reason is what Microsoft has done to Windows. Not enough to get Corpos to switch but certainly a lot of consumers.

 

[ondma]

About 2/3rds of computers use iGPU. I don't think we should just exclude the majority of computers because you personally don't care. I personally care since the company I work for will only buy SFF computers with no ability to install a discrete GPU. And Solidworks chugs on iGPUs--I'll take any boost that I can get. Yes, discrete GPUs are more capable. But there goes any money savings of one CPU over most others, and the few watts of power we bicker about on these forums between CPUs are destroyed once you have to power a separate GPU.

It says Intel didn't buy enough equipment. Plain and simple. That says nothing about how good or bad the nodes are. It says nothing about how good or bad the desktop CPUs are (which is what I thought you were wanting to talk about). And that says nothing about the future when Intel really wants to be a foundry (which from the announcements really seems to be 18A and beyond). All of Intel's eggs are in that basket. 2026 is going to be a make or break year for their foundry ambitions. If it doesn't work out, they might have to be like Apple, AMD, or ARM: design of chips only.

It doesn't matter how "good" the node is if you cant produce it in sufficient quantity to satisfy the demand.
And I bet nobody would be talking about the igpu or the npu for ARL if the cpu itself delivered 20% ST and 30% nT uplift.

 

[Goop_reformed]

It doesn't matter how "good" the node is if you cant produce it in sufficient quantity to satisfy the demand.
And I bet nobody would be talking about the igpu or the npu for ARL if the cpu itself delivered 20% ST and 30% nT uplift.

Yep because it's about the single/multi threaded performance at reasonable power. Everything else is just extra.

 

[majord]

What's not exciting about a x86 laptop platform that has a chance to meet, no even exceed ARM platforms in the *dare I say it!* battery life department!

It's mostly hype anyways. i.e the only thing x86 is struggling with competitiveness with is ST perf/watt , which doesn't directly translate into battery life anyway.

 

[coercitiv]

Arrow Lake i7 made an appearance on the GB 5 score database. Max clocks are 5.5 Ghz, so the it will have very little clock regression vs. 14700K.

Maxsun MS-iCraft Z890 Pacific - Geekbench

Benchmark results for a Maxsun MS-iCraft Z890 Pacific with an Intel Core Ultra 7 265KF processor.

browser.geekbench.com

 

[ErichZannMusician]

Can't tell much from that score as it is using a single 24 GB stick of RAM.

 

[Henry swagger]

Intel Core Ultra 7 265KF 20-Core "Arrow Lake" CPU Benchmarks Leak, Up To 5.5 GHz Clocks

The first benchmarks of Intel's Core Ultra 7 265KF "Arrow Lake" CPU with 20 cores have been spotted within Geekbench.

wccftech.com

20A seems to be clocking high.. wonder if it can reach 5ghz with a oc 🤔🙂

 

[jdubs03]

Arrow Lake i7 made an appearance on the GB 5 score database. Max clocks are 5.5 Ghz, so the it will have very little clock regression vs. 14700K.

Maxsun MS-iCraft Z890 Pacific - Geekbench

Benchmark results for a Maxsun MS-iCraft Z890 Pacific with an Intel Core Ultra 7 265KF processor.

browser.geekbench.com

View attachment 103984

Based on that about a 6% uplift in ST. And a 23% drop in MT (17722 pts). But indeed that single-channel is problematic. If only we could do a proper GB5 search.

I7-14700k in GB 5.4 (perhaps not directly comparable, but close) - notebookcheck
ST: 2121 pts
MT: 23159 pts

 

[MarkPost]

Arrow Lake i7 made an appearance on the GB 5 score database. Max clocks are 5.5 Ghz, so the it will have very little clock regression vs. 14700K.

Maxsun MS-iCraft Z890 Pacific - Geekbench

Benchmark results for a Maxsun MS-iCraft Z890 Pacific with an Intel Core Ultra 7 265KF processor.

browser.geekbench.com

View attachment 103984

Not good. This is my 13900K (with Intel default settings). First, dual channel (2249 / 24512):

Gigabyte Technology Co., Ltd. Z690 AORUS ELITE AX - Geekbench

Benchmark results for a Gigabyte Technology Co., Ltd. Z690 AORUS ELITE AX with an Intel Core i9-13900K processor.

browser.geekbench.com

And single channel (2258 / 19392):

Gigabyte Technology Co., Ltd. Z690 AORUS ELITE AX - Geekbench

Benchmark results for a Gigabyte Technology Co., Ltd. Z690 AORUS ELITE AX with an Intel Core i9-13900K processor.

browser.geekbench.com

So, single channel has only impact in MT score.

 

[Goop_reformed]

Not good.

Yep. Intel's back to single digit gains, again.

 

[jdubs03]

Yep. Intel's back to single digit gains, again.

If there is a gain at all. Using the percentage difference from the multi vs single channel scores. We’re looking at around 22400 points.

 

[511]

Yep. Intel's back to single digit gains, again.

We will get double digit MT gains though

 

[coercitiv]

Not good.

For now I'm not interested in the score, just the boost blocks. If the i7 is indeed 5.5Ghz, then the i5 has a very good change of matching the 14600K in clocks, maybe even gain 100Mhz.

 

[Goop_reformed]

We will get double digit MT gains though

Double digits could be 15% or 50%. Would you say a single digit 1t and 20% nt are disappointing?

 

[FlameTail]

20A seems to be clocking high.. wonder if it can reach 5ghz with a oc 🤔🙂

is it not N3B?