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.

 

[poke01]

Run SIMD code on Apple vs AMD and you will find it gets beaten in both power and perf/wattthe gap is not that big

Not bad for such an INT/FP focused CPU.

Math fun

The 9950X is just better here for peak performance. Now if Apple ever includes proper SVE2.1 then we can talk about DC talk. But have to say not bad especially the M4 Max with NEON it put up a good fight.

A Brief Study of Memory Bound Application Performance on M3 Ultra and M4 Max in Julia

Hello everybody! I wanted to share my Linear Algebra performance findings in Julia on the M4 Max and M3 Ultra in double precision floating point. As a Julia novice, I was hugely impressed how easy it was for me to brainstorm ideas in Julia and run quick performance tests. I subsequently sanity...

discourse.julialang.org

 

[poke01]

View attachment 134984
Not bad for such an INT/FP focused CPU.

View attachment 134985Math fun

The 9950X is just better here for peak performance. Now if Apple ever includes proper SVE2.1 then we can talk about DC talk. But have to say not bad especially the M4 Max with NEON it put up a good fight.

A Brief Study of Memory Bound Application Performance on M3 Ultra and M4 Max in Julia

Hello everybody! I wanted to share my Linear Algebra performance findings in Julia on the M4 Max and M3 Ultra in double precision floating point. As a Julia novice, I was hugely impressed how easy it was for me to brainstorm ideas in Julia and run quick performance tests. I subsequently sanity...

discourse.julialang.org

Now we stop saying Apple cores are better in everything. They are tailor made for extracting ST perf and for compiling your code but NEON is just too slow for SIMD tasks.

Now unless Apple changes direction or mandated by ARM, you won’t ever see SVE2 in their cores

 

[Thunder 57]

Now we stop saying Apple cores are better in everything. They are tailor made for extracting ST perf and for compiling your code but NEON is just too slow for SIMD tasks.

Now unless Apple changes direction or mandated by ARM, you won’t ever see SVE2 in their cores

I'm surprised no one has mentioned the massive L1 caches in particular for the Apple cores. Of course that is excellent for ST performance.

Oh and BTW, there is a thread for x86 vs ARM. Why can't these posts be there instead? Seems OT.

 

[Magio]

◾ “WilCat Lake” is going to be a ENORMOUS DISAPPOINTMENT in the range of chips under 10W (for possible fanless use).
➖ On the one hand, the “Qualcomm Snapdragon 8 Elite Gen 5” with an estimated TDP of 8W and an NPU with 80 TOPS, ideal for “Windows 11 Copilot Plus.”
➖ On the other hand, the AMD “Krackan Point” with a configurable TDP of 9W and an NPU of 50 TOPS, which exceeds 40 TOPS and is compatible with “Copilot Plus.”
➖ And finally, the “Intel WildCat Lake” with an estimated TDP of “less than 10W” and an NPU of 18 TOPS, INSUFFICIENT for “Copilot Plus.”
➖ I think Intel should implement its latest “NPU 6” design in “WilCat Lake” NOW, so that it can at least compete with an NPU that will be compatible with “Copilot Plus.”

I don't know which part of this I find more comical, using NPU TOPS as the barometer on which to judge a chip, citing the max theoretical TOPS in conjunction with the lowest possible TDPs, acting like anyone in the world cares about "Copilot + PC" branding when everything actual people do with AI relies on cloud compute or suggesting Intel swap NPU6 into WCL now as if that's how chip design works.

... Or maybe it's the idea that the fanless Windows laptop market will be a significant segment within WCL's lifespan.

 

[dullard]

...

Please don't repost the same exact block of text in multiple threads. https://forums.anandtech.com/threads/intel’s-unified-core-there-is-hope.2630962/post-41545096

1) You misspelled Wildcat multiple times.
2) Qualcomm products don't really run x86 code well. For example, last I saw was only 90% compatibility in new games--let alone the hundreds of thousands of older games it can't play. https://overclock3d.net/news/gpu-di...ame-compatibility-with-its-new-adreno-x2-gpu/
3) You misspelled Krakan.
4) You don't need to put everything in quotation marks.
5) The rumors for Wildcat Lake is 40 TOPS total which is just enough for Copilot+.
6) Who is exactly going to be disappointed? Wildcat Lake is for the ultra low-end price computer tasks. These people aren't really the ones clamoring for Copilot+.

 

[poke01]

@511 , do you know the size of the N3E GPU in panther lake?

 

[511]

@511 , do you know the size of the N3E GPU in panther lake?

54mm2 N3E for 12Xe3+16MB L2 LNL was 39mm2 for 8Xe2 Cores+8MB L2

 

[regen1]

He got in at early 2021. You can't say it isn't mostly under his leadership. Design by itself takes a year. Rest are basically making it physical and implementation plus testing.

Core u-arch design targets would have been over by then. But yeah he still would have decent time on other parts of LNL/ARL, GNR execution to an extent. With NVL he would have had better say in execution(more so if he could've stayed) than say MTL/ADL or ARL.

CEOs aren't responsible for low level decisions. You can't anyway, there's just not enough time. If he was a chief architect it would be different. Making sure the company runs smoothly so products are made smoothly, those are his responsibilities.

Yeah, mostly they aren't but they are involved in some level for setting design/performance targets esp. wrt new u-arch based products, node selections, choosing(and cancelling like Royal Core) between stuff, etc.

Raptorlake they claimed 3 years from design to availability on shelves.

2.5 years. ADL was 3 years. RKL, IIRC, was 2 years? Could also have been 2.5.

The time period(2? 2.5? 3?) referred here are likely a part of it but not the full period from conception of u-arch to product availability.
In last 10 or so years for a Tock-level or new cutting-edge design for Intel(P-core) or AMD, from ZERO to products on sale, it has been 4-5(+) years of time roughly if not more, there has been a lot of times it's been towards 5 or more.

Spoiler: Timelines

ADL isn't done in 3 years(from GLC conception to ADL availability) when SPR is known to be delayed for so long. RPC(RPL) and RWC(MTL) are small changes over GLC. "Alder Lake" name was known in 2017, so it would have started somewhat earlier.
Cypress Cove(RKL, 14nm family) is a derivative of SNC used in ICL which took quite long to come out(some of it down to 10nm delays). So it isn't just 2 or 2.5 years. Haswell started in 2007, launched in 2013. Mike Clark claimed in 2018 about working on Zen 5 and first Zen 5 based products released in July 2024(Strix Point). I am not completely sure about Lion Cove(might be 2019 or 2020). Zen 8's initial work started quite some time(years) ago but it isn't coming anytime soon.

 

[regen1]

54mm2 N3E for 12Xe3+16MB L2 LNL was 39mm2 for 8Xe2 Cores+8MB L2

Image from High Yield's video

Xe3P should provide another decent PPA improvement.

 

[511]

View attachment 135081
Image from High Yield's video

Xe3P should provide another decent PPA improvement.

Why use them when you have even better stuff
54.73mm2

 

[regen1]

Why use them when you have even better stuff
54.73mm2

View attachment 135118

Yeah, that was just all 3 configs side by side.

 

[DavidC1]

Please don't repost the same exact block of text in multiple threads. https://forums.anandtech.com/threads/intel’s-unified-core-there-is-hope.2630962/post-41545096

1) You misspelled Wildcat multiple times.
2) Qualcomm products don't really run x86 code well. For example, last I saw was only 90% compatibility in new games--let alone the hundreds of thousands of older games it can't play. https://overclock3d.net/news/gpu-di...ame-compatibility-with-its-new-adreno-x2-gpu/
3) You misspelled Krakan.
4) You don't need to put everything in quotation marks.
5) The rumors for Wildcat Lake is 40 TOPS total which is just enough for Copilot+.
6) Who is exactly going to be disappointed? Wildcat Lake is for the ultra low-end price computer tasks. These people aren't really the ones clamoring for Copilot+.

Dead internet theory.

AI promoting AI, because it's only promoting AI related features.

 

[Hulk]

Just read at Tom's that the 290K will have boost frequencies of 5.6 for P and 4.8 for E. Thinking about how area performant Skymont is I'm wondering if Intel considered holding the P core count 8 and using the rest of the die space for the E's?

 

[511]

Just read at Tom's that the 290K will have boost frequencies of 5.6 for P and 4.8 for E. Thinking about how area performant Skymont is I'm wondering if Intel considered holding the P core count 8 and using the rest of the die space for the E's?

it's like 6.83mm2 for the Skymont Cluster with 4MB L2 and 4.68mm2 for Lion Cove and 3MB L2 you can see how much area LNC Hogs and Intel only got 12 Stops on the Ring

 

[regen1]

Exclusive: This is Lenovo's ThinkPad lineup for 2026, including ThinkBook Plus with rotatable display and ThinkCentre X AIO

Lenovo’s ThinkPad lineup for CES 2026 includes a Rollable concept, Auto Twist, Aura Editions, new Intel CPUs, and refined flagships.

www.windowslatest.com

Exclusive: Intel Panther Lake SKUs for Lenovo's 2026 refresh (Ultra 7 356H, Ultra X9 388H, Ultra X7 358H)

Windows Latest has obtained details on some unannounced Intel SKUs for 2026, through a slew of exclusive information about some upcoming Lenovo laptops, slated to be announced at CES 2026. According to our reliable sources, Lenovo will use Intel’s next CPU lineup under multiple labels, including...

www.windowslatest.com

Some specs-preview of some of the upcoming models.

Just read at Tom's that the 290K will have boost frequencies of 5.6 for P and 4.8 for E.

Small addition, U9 290K Plus will have TVB of max 5.8GHz for P cores.

 

[Hulk]

Intel has a long way to go. Just a year ago, after 30 years of building Intel only inside computers, I built a 9950X. It has been a flawless computer.

I was an ardent Intel supporter. Now I have kind of written them off and am much more interested in Zen 6 rather than Nova Lake. I don't believe Intel can do it. All of the news about Intel is "plans." No working silicon, no proof of 18A delivering and yielding. Could they do it? Of course. But I won't believe it until I see it.

On the otherhand I'm feeling a quiet confidence from AMD. I'm thinking that AMD will be a solid upgrade from Zen 5 and won't have to fill any massive performance holes (like ARL gaming) from the last generation.

I'm pulling for Intel but am highly suspect of them delivering.

 

[511]

Intel has a long way to go. Just a year ago, after 30 years of building Intel only inside computers, I built a 9950X. It has been a flawless computer.

Not that long to go

I was an ardent Intel supporter. Now I have kind of written them off and am much more interested in Zen 6 rather than Nova Lake. I don't believe Intel can do it. All of the news about Intel is "plans." No working silicon, no proof of 18A delivering and yielding. Could they do it? Of course. But I won't believe it until I see it.

Half the media is against Intel tbh at this point as for No working Silicon on 18A they are ramping the node in Arizona and are shipping PTL to their Customers.

On the otherhand I'm feeling a quiet confidence from AMD. I'm thinking that AMD will be a solid upgrade from Zen 5 and won't have to fill any massive performance holes (like ARL gaming) from the last generation.

I'm pulling for Intel but am highly suspect of them delivering.

Well NVL is something they have thrown the kitchen sink at the problem

 

[jdubs03]

Intel has a long way to go. Just a year ago, after 30 years of building Intel only inside computers, I built a 9950X. It has been a flawless computer.

I was an ardent Intel supporter. Now I have kind of written them off and am much more interested in Zen 6 rather than Nova Lake. I don't believe Intel can do it. All of the news about Intel is "plans." No working silicon, no proof of 18A delivering and yielding. Could they do it? Of course. But I won't believe it until I see it.

On the otherhand I'm feeling a quiet confidence from AMD. I'm thinking that AMD will be a solid upgrade from Zen 5 and won't have to fill any massive performance holes (like ARL gaming) from the last generation.

I'm pulling for Intel but am highly suspect of them delivering.

I think sentiment will really driven by how Panther Lake does in the mobile segments compared to Zen 5. If it’s a meaningful improvement, that will inspire confidence for Nova Lake.

But Nova Lake was always* supposed to be a big step. Don’t be surprised when it trades blows with Zen 6 on overall performance and efficiency.

 

[Khato]

Heh, whereas I happily 'downgraded' from an i9-14900k to an Ultra 7 265k and have only noticed performance improvements. Which matches up with my expectations from reviews. I know the narrative of those reviews was 'ARL gaming performance regressions', which is quite true if you're gaming with an RTX 5090 at 720p. In the real world of sub $1k graphics cards running at 1440p or 4k there are only a few games where there's a meaningful performance difference.

 

[Geddagod]

Half the media is against Intel tbh

If half the media is against intel and the other half is for Intel, isn't it balanced?

I think sentiment will really driven by how Panther Lake does in the mobile segments compared to Zen 5.

I honestly don't think mobile has that much of an effect mindshare, maybe with the exception of LNL, and that's strictly against ARM, not gaining mindshare against AMD.

Don’t be surprised when it trades blows with Zen 6 on overall performance and efficiency.

Why shouldn't people be surprised if this is the case?
The current scenario is that AMD has better design teams. If they are trading blows iso node, that means Intel must have at least made some progress in that front. Or maybe they just throw silicon at it to catch up to AMD.

 

[jdubs03]

If half the media is against intel and the other half is for Intel, isn't it balanced?

I honestly don't think mobile has that much of an effect mindshare, maybe with the exception of LNL, and that's strictly against ARM, not gaining mindshare against AMD.

Why shouldn't people be surprised if this is the case?
The current scenario is that AMD has better design teams. If they are trading blows iso node, that means Intel must have at least made some progress in that front. Or maybe they just throw silicon at it to catch up to AMD.

Well looking at the 285K vs the 9950X-variants, it’s relatively close. Like I said Nova Lake is supposed to be a big step. I get the pessimism, Intel has disappointed on the design and process side for quite some time. That doesn’t mean it has to continue.

 

[Joe NYC]

Heh, whereas I happily 'downgraded' from an i9-14900k to an Ultra 7 265k and have only noticed performance improvements. Which matches up with my expectations from reviews. I know the narrative of those reviews was 'ARL gaming performance regressions', which is quite true if you're gaming with an RTX 5090 at 720p. In the real world of sub $1k graphics cards running at 1440p or 4k there are only a few games where there's a meaningful performance difference.

What prompted you to make such a "sideways" move?

They only reason I can think of someone moving from 14900k to 265k is if the 14900k is experiencing silicon degradation, or the performance has been degraded by various mitigation patches Intel released.

My goal, when upgrading is to get 2x performance, or +100% increase. Not +/- 5%

 

[Hulk]

Not that long to go

Half the media is against Intel tbh at this point as for No working Silicon on 18A they are ramping the node in Arizona and are shipping PTL to their Customers.

Well NVL is something they have thrown the kitchen sink at the problem

This is all great news and I truly hope to be proven wrong and would be happy to see a state-of-the-art node from Intel with good yields and solid architecture on display for both Panther and Nova Lake. It would be a great comeback story after quite a few tough years, especially on the fab side of things. I'm not nearly as "in the know" about fabs in particular but I always felt like Intel fabbing at TMSC was a "waving the white" flag moment from them. I still haven't forgotten "Five nodes in four years." Trust has to be earned back and it's gonna take more than PR this time. I'm not "buying it" like I used to.

I should also mention that my move to AMD was complete when I ditched my Surface 2 laptop for a Asus ProArtPX13 with an AMD HX370 CPU. Another amazing piece of engineering from AMD. I remained loyal to Intel until they forced me to AMD.

Eagerly waiting for Panther Lake reviews and the news that "Intel is Back!"

 

[regen1]

So many things can go right and yet very few things that don't go that well could put you behind the competition. Eg. (hypothetical)everything including IPC looks good or better vs competition but chips don't clock high enough bringing overall ST below the competition. Or your line-up is competitive in ST, MT but don't have a proper answer to X3D in gaming.
Even GNR vs Turin, apart from a gen old core(RWC), mesh issues leave out a lot of performance on the table despite being otherwise decent in many areas. Also 2P scaling issues on GNR.
ARL is pretty decent in notebooks(much more volume and design choices vs competition). Even in desktop the main issue is not having an answer for Zen 5 X3D(in gaming hence the mind-share) and some others but it is competitive in ST, MT.
NVL should be decent upgrade over ARL in many areas including performance but where it lands against the competition will be see then. NVL will very likely do really well in notebooks.

 

[Khato]

What prompted you to make such a "sideways" move?

They only reason I can think of someone moving from 14900k to 265k is if the 14900k is experiencing silicon degradation, or the performance has been degraded by various mitigation patches Intel released.

My goal, when upgrading is to get 2x performance, or +100% increase. Not +/- 5%

In terms of performance the 265k has been equal or better since I limited the 14900k to 200w. (Prefer reasonably quiet heatpipe based cooling over liquid.) I'm sure that raises an equally valid question of why I bothered with the 14900k in the first place. Answer there is I've been upgrading every generation and re-selling the used CPU+motherboard on eBay for roughly the original price paid since core 2 duo. Perk of having started my career at Intel and still having friends who work there is access to their employee purchase program. The at-cost pricing is far, far better than the 15-20% reimbursement AMD offers, or the 10% discount with NVIDIA. I had thought about skipping ARL, but the abnormally cheap pricing of the 265k combined with higher end z890 boards being re-sold at 1/2 to 1/3 of original price earlier this year made it a bit too tempting. (Thanks Newegg for your forced bundling of motherboards no one wanted with graphics cards far too many people wanted.)

I'd certainly like it if there were reasonable 2x performance opportunities to be had. It was fun getting those kinds of improvements from graphics cards for a bit without having to pay an obscene amount.