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.

 

[coercitiv]

Oh please let's not have another month of "wait for new BIOS/power limit/windows update".

Nah, this time we're going straight for "wait for CAMM2".

 

[511]

The worst ending for Intel has happened, the reviewer mob has forced them to pre nerf their chips. Add to it the fiasco with Raptor Lake and the enforced power limits on a overclocker + unlocked chip and we get this bad ending.... pathetic.

A K unlocked chip that will be reviewed as if it has no overclocking and the "K" is just a marketing artifact. It is sadly over....

If there is some OC potential to exceed the 14900K, how many of these AMD curve optimizer tuning, command line running and windows reinstalling folks will do the same for Intel, the marketing machine has worked, I'm saddened to know that a focused and timed mob can indeed push a company towards the wrong path.

A core i9 should be the most unleashed chip, with the highest power consumption and performance, the lower SKUs are for everything else. It should be a halo product like the RTX4090, no one complains about the power consumption under load, the same will be the case for the 5090.

The only way for Intel to survive now is to embrace the power consumption, I don't care about that on a high performance desktop BUT they also must deliver a multi generational leap in ST and MT performance to warrant that power draw, in essence they need a RTX-X090 class product with performance so high, no serious person will talk about the power draw.

Like this though, I'm just disappointed and will return in maybe 4 years to see if there is any positive movement.

Conclusion: Stagnation since Alder Lake

Stagnation since Raptor Lake*

 

[MS_AT]

The chart is with APO enabled so without APO it's gonna tank hard

Sorry, what is APO? I am getting lost in all those acronyms...

 

[511]

Canned Arrow lake refresh makes sense

 

[511]

Sorry, what is APO? I am getting lost in all those acronyms...

Intel® Application Optimization Overview

Explains Intel® Application Optimization for gaming optimization.

www.intel.com

Basically a way to increase FPS in games

 

[coercitiv]

Sorry, what is APO? I am getting lost in all those acronyms...

Software gaming optimization, they're tuning E core utilization (example - in some games disabling a number of E cores allows the rest to make better use of the shared L2, which lowers L3 hits, which allows better L3 utilization for P cores, and you can see where this is going)

Optimizations are "manual" for each title though, so the software has limited real-world support.

 

[Det0x]

View attachment 108882

If this is the best Intel can show in their own benchmarks, the 285k is DOA for gamers

 

[AcrosTinus]

https://videocardz.com/newz/intel-admits-core-ultra-9-285k-will-be-slower-than-i9-14900k-in-gaming#disqus_thread

Increasing energy efficiency is great, but the performance itself disappointing.

But what if all these performance numbers are for the conservative default power settings??? It could increase considerably with the 250W power setting.

That might be true but it is all about mindshare right now. Intel might have the higher performing chip at 250 to 300W again but who will test this unlocked chip in that conditions.

The likes of hardware unboxed will put it against AMD, leave the locks in place, throw in some in spec RAM and call the 7800X3D a winner, I can see the headline.

Any benefit this chip has, be it in hardware encoding, chipset, tuning will be smoothed over for the gaming folks. The moment Intel does well, it will be countered with "at an increased power draw, at a higher cost, a the cost of a new platform".

Sadly there is no winning, going for power draw even though you had to impress people with performance FIRST, is intel's downfall.
Most people I know game on an i5 or Ryzen 7 the upper chips should be halo products.

This all started with idiots reviewing the 14900K, a huge chip against a 7800x3D a lower tier chip with half its guts missing. They call it faster but forget to add in the constraint of "just in a particular task". Had they used a overclocked i5 14600K for comparison which would be maybe 3 to 5% slower than the i9, the power draw nonsense would not have happened... It would be more fair, the tradeoff between higher MT and gaming performance against cost would be on Intel's side with older Mainboards and even DDR4 but no one bothered to do this but niche channels...

The media machine worked....

 

[coercitiv]

If this is the best Intel can show in their own benchmarks, the 285k is DOA for gamers

Pretty much, yes. The only good news is Ultra 7 and Ultra 5 will fare a bit better versus their respective previous gen counterparts because there's less or no frequency regression. Still a tough sell.

 

[511]

RPL Intel 7 making Zen4/Zen5/ARL on N4/N3B look weak in ST how the hell did we get here P core team needs to deliver something big like 30-40% with NVL or it's meh

 

[adroc_thurston]

RPL Intel 7 making Zen4/Zen5/ARL on N4/N3B look weak in ST

One little detail: it goes kaboom.
Hope that helps!

 

[coercitiv]

RPL Intel 7 making Zen4/Zen5/ARL on N4/N3B look weak in ST how the hell did we get here

We got here by pushing RPL to the point of failure.

 

[511]

One little detail: it goes kaboom.
Hope that helps!

Let us wait for the 0x12B fix but I don't think it was nodes fault more like a design/insanse settings issue

 

[AcrosTinus]

RPL Intel 7 making Zen4/Zen5/ARL on N4/N3B look weak in ST how the hell did we get here P core team needs to deliver something big like 30-40% with ARL or it's meh

They have to go big or go home, Arrow Lake might break their arms, the next Alder Lake+++ will break their neck

 

[igor_kavinski]

Therefore anyone who's already on RPL has very little incentives to upgrade. I expect the prices won't help either.

But if suddenly RPL prices mysteriously start seeing a downward trend, people may panic and jump ship to either Ryzen or ARL. We just need one more security related performance degrading patch or an advisory from Intel saying that HT must be turned off on Raptor Lake to protect your digital life while on internet

 

[511]

They will definitely reduce RPL price after Fiasco if RPL hadn't happened it would not have been this good 🤣

 

[igor_kavinski]

Also, we don't know how well ARL overclocks yet. The lower power should allow eeking out at least 200 MHz extra on good samples. That's not a nothing.

 

[Kocicak]

I expect the reviews to have the new CPUs tested in both default and performance settings.

If intel chose to show just one power setting in their presentation, it should be the default setting.

 

[tsamolotoff]

Also, we don't know how well ARL overclocks yet. The lower power should allow eeking out at least 200 MHz extra on good samples. That's not a nothing.

Considering that they've got rid of HT to eke out more Mhz in ST and MT this is not very likely. Probably the same OC potential as with AMDs offering on TSMC processes

 

[511]

I don't think N3B can compete with 6Ghz clocking Intel 7 that's something no one can do outside of Intel Fabs Ghz node skymont is the true chad basically more IPC than Zen 5+ Arl combined and still has few percent left

 

[CouncilorIrissa]

I've said this before, but this outcome shouldn’t be a surprise. RPL was essentially a swan song of Intel's approach to desktop CPUs: a monolithic chip that does not suffer from penalties typical to disaggregated architectures (higher mem and C2C latencies), setup with no care for power consumption on a node that clocks sky high like no other, pushed to the absolute limits and beyond of its silicon. It’s a perfect setup for gaming.

 

[poke01]

I don't think N3B can compete with 6Ghz clocking Intel 7 that's something no one can do outside of Intel Fabs Ghz node skymont is the true chad basically more IPC than Zen 5+ Arl combined and still has few percent left

Its a real funny sort of situation be in, after finally moving to a new node, Intel got stuck with N3B the worst 3nm node from TSMC. If it was N3E, it would allowed them to clock higher.

 

[CouncilorIrissa]

Its a real funny sort of situation be in, after finally moving to a new node, Intel got stuck with N3B the worst 3nm node from TSMC. If it was N3E, it would allowed them to clock higher.

I don't think so, N3E's entire shtick is that it's cheaper and yields better.

 

[majord]

So Zen 5% out and "ARL - 2% " in ?

 

[511]

So Zen 5% out and "ARL - 2% " in ?

Zen 5% = Arl -2%