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]

Ah, Asus had to do something special

 

[DavidC1]

Not necessarily wrong design. AMD has been dealing with this challenge (successfully) since the dawn of the chiplets. Maybe a skill issue.

It was reported Meteorlake's development time went through hell of sorts. And 7nm delays are related to that.

If the leaked results are anywhere near true, then we're still seeing the ramifications of various layoffs and development hell during that period. When you have disgruntled employees, it doesn't merely end up in standstill of development, but even regressions.

How do fancy tech like Foveros, EMIB matter when you have a team incapable of implementing it properly? They will simply fail to execute on the vision of the technology.

People>>>Tech.

 

[igor_kavinski]

It was reported Meteorlake's development time went through hell of sorts. And 7nm delays are related to that.

They had no plan B this time. Went all in on tiles and paid the price with their inexperience. They must've known this at least a year back. Still did not take any corrective measures to try to get the performance up to snuff. They must've thought that the Intel brand will just sell itself. Maybe it will. To an extent. For gamers and most rational enthusiasts, they almost do not exist anymore, unless they release Bartlett Lake next year to save face.

 

[Thunder 57]

It was reported Meteorlake's development time went through hell of sorts. And 7nm delays are related to that.

If the leaked results are anywhere near true, then we're still seeing the ramifications of various layoffs and development hell during that period. When you have disgruntled employees, it doesn't merely end up in standstill of development, but even regressions.

How do fancy tech like Foveros, EMIB matter when you have a team incapable of implementing it properly? They will simply fail to execute on the vision of the technology.

People>>>Tech.

I guess we know why MTL on desktop was cancelled.

 

[Hulk]

It's all fired up in here like when the Harvesters showed up in Independence Day and all the doodads started turning on!

 

[Thunder 57]

It's all fired up in here like when the Harvesters showed up in Independence Day and all the doodads started turning on!

I had no idea what Harvesters were or that Independence Day had such a, what would it be called, fan fiction following? Like the Expanded Universe in Star Wars? There's probably a word for it. I know The Matrix has one as well.

Anyway, about 12 hours to go. I wonder how many new pages this topic gets in the next 24 hours. Over/Under 8? I'd take the over.

 

[Wolverine2349]

@Wolverine2349 , looks like Bartlett Lake is a go!

Any source on that?

 

[LightningZ71]

What I'm trying to wrap my head around is this: how is arrow lake a better product for Intel than just a shrink of Raptor Lake on a 3nm process such as N3 or i3? Or, better yet, a mild refresh of Raptor Lake on a 3nm node with an updated iGPU, slightly larger L2 caches and 8 more e cores, even if they were still Crestmont? You'd likely have slightly improved ST, slightly improved MT, better power numbers, and wouldn't have to fight with packaging for emib/foveros.

Heck, just shrinking Bartlett Lake on I3/N3 with mild tweaks to L2 and the IMC and you still get better ST with perhaps flat MT.

This just strikes me as poor decision making at this point if these bench numbers prove out.

 

[gdansk]

Golden Cove on N3B could be a disaster too. Lion Cove looks like Golden Cove made more process portable instead of bolted to Intel fabrication. They made some changes (8 wide, buffers, cache hierarchy) to offset the clock regression and some removals (HT) to simplify validation.

Why did it end up dis-aggregated? To re-use some components or technology from Meteor Lake.
It seems like decent enough decision making with perhaps not the greatest execution. But apparently good enough to keep up with Zen 5.

 

[Thunder 57]

What I'm trying to wrap my head around is this: how is arrow lake a better product for Intel than just a shrink of Raptor Lake on a 3nm process such as N3 or i3? Or, better yet, a mild refresh of Raptor Lake on a 3nm node with an updated iGPU, slightly larger L2 caches and 8 more e cores, even if they were still Crestmont? You'd likely have slightly improved ST, slightly improved MT, better power numbers, and wouldn't have to fight with packaging for emib/foveros.

Heck, just shrinking Bartlett Lake on I3/N3 with mild tweaks to L2 and the IMC and you still get better ST with perhaps flat MT.

This just strikes me as poor decision making at this point if these bench numbers prove out.

My guess is that they realized they couldn't stay monolithic forever. They had to move to tiles even if it meant a harsh learning curve. I also think they really wanted to get away from RPL. I don't think a shrunk RPL would've been more efficient. From the first page of the leaked review it is clear they were pushing efficiency (probably because they couldn't market performance).

The press-briefing was all about efficiency. All of the literature mentions it. Less power equals less heat. Less heat is what the Intel Core range is crying out for.

Honestly, if was equal to RPL but way more efficient, I think a lot of people (myself included) would've been fine with that. The leaked benchs today didn't show that, though there were some anomolies. Best to wait for more reviews before judging.

 

[Hulk]

I had no idea what Harvesters were or that Independence Day had such a, what would it be called, fan fiction following? Like the Expanded Universe in Star Wars? There's probably a word for it. I know The Matrix has one as well.

Anyway, about 12 hours to go. I wonder how many new pages this topic gets in the next 24 hours. Over/Under 8? I'd take the over.

I'm a sci-fi nerd, no doubt about it. Loved that scene with Brent Spiner in Independence Day.

 

[Josh128]

Some whispers at WCCFTECH comments section that everything leaked today was a smoke screen, and what we will witness tomorrow will revive the #BlackArrow mantra.

 

[jdubs03]

Some whispers at WCCFTECH comments section that everything leaked today was a smoke screen, and what we will witness tomorrow will revive the #BlackArrow mantra.

Tbh after the whole Zen 5 thing, I’ve just decided I’ll come to a conclusion after all of the reviews are in. Hopefully no required Windows updates/new firmware versions are needed to have an established understanding.

 

[Hulk]

Remind me again, please? What was the final concenus regarding Lunar Lake? Total fail? Amazing? Good battery life/ST performance, not so good MT performance? Where Intel's performance/efficiency claims substantiated by reviews?

I just want to know where we are before we dive into ARL.

 

[poke01]

What was the final concenus regarding Lunar Lake? Total fail

Excellent battery life for video playback/web browsing and good for teams.

Some whispers at WCCFTECH comments section that everything leaked today was a smoke screen, and what we will witness tomorrow will revive the #BlackArrow mantra.

I can’t tell if this sarcasm or not?

 

[Thunder 57]

Some whispers at WCCFTECH comments section that everything leaked today was a smoke screen, and what we will witness tomorrow will revive the #BlackArrow mantra.

As if they are in the know. That would be one hell of a smokescreen though. Intel itself didn't exactly have much to saw about the gaming performance in particular. If there slides are accurate though then kudos to them compared to the Zen 5 slides.

 

[controlflow]

Tbh after the whole Zen 5 thing, I’ve just decided I’ll come to a conclusion after all of the reviews are in. Hopefully no required Windows updates/new firmware versions are needed to have an established understanding.

Windows updates do seem to matter for ARL

Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 - Geekbench

Benchmark results for a Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 with an Intel Core Ultra 7 265K processor.

browser.geekbench.com

Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 - Geekbench

Benchmark results for a Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 with an Intel Core Ultra 7 265K processor.

browser.geekbench.com

Same machine but 2nd result is with Windows build 26100.2152. This is from a member over at overclockers forum. Probably going to take a few OS, driver and FW updates to get these things working right.

 

[poke01]

What longhorn is talking about is LNL performance in when it’s unplugged and windows mode set to balanced. Then the performance sucks.

Compare that with M3 it doesn’t drop performance on battery. It’s just Intel going aggressive to get goood battery life. This is something Qualcomm took advantage of.

 

[jdubs03]

Windows updates do seem to matter for ARL

Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 - Geekbench

Benchmark results for a Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 with an Intel Core Ultra 7 265K processor.

browser.geekbench.com

Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 - Geekbench

Benchmark results for a Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 with an Intel Core Ultra 7 265K processor.

browser.geekbench.com

Same machine but 2nd result is with Windows build 26100.2152. This is from a member over at overclockers forum. Probably going to take a few OS, driver and FW updates to get these things working right.

That should be directly disclosed in the reviews then. At a minimum tested on the most recent build, which I would assume it would be.
26100.2152 the release preview. 26100.2033 is the current WW GA release. Was that first test on .2033?

 

[Saylick]

Windows updates do seem to matter for ARL

Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 - Geekbench

Benchmark results for a Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 with an Intel Core Ultra 7 265K processor.

browser.geekbench.com

Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 - Geekbench

Benchmark results for a Gigabyte Technology Co., Ltd. Z890 AORUS ELITE WIFI7 with an Intel Core Ultra 7 265K processor.

browser.geekbench.com

Same machine but 2nd result is with Windows build 26100.2152. This is from a member over at overclockers forum. Probably going to take a few OS, driver and FW updates to get these things working right.

Right. Seems like performance is all over the place, with a lot of it depending on the Windows version.

Also from OCUK: https://forums.overclockers.co.uk/t...s-15th-gen-on-lga-1851.18980945/post-37426875



Apparently the 13950HX was pulling 160W average in that CP2077 scene, while the 265K was pulling somewhere in the 60W range on average. Sounds like the 265K is unintentionally not clocking as high as it should.

 

[Saylick]

I feel like the launch reviews tomorrow will either be all over the place and we'll have to wait a few more weeks or even months for the situation to settle (similar to Zen 5) OR all of this was just a smokescreen and 1st party reviewers will be in alignment because they had the launch BIOS and Windows updates from the get-go. There won't be an in-between. My money is on the former.

 

[Kepler_L2]

Another review: https://www.coolaler.com/forums/thr...te-z890-aorus-master-ckd.386911/#post-3463822

ARL seems to mostly match Zen4 non-X3D in gaming.

 

[Saylick]

Another review: https://www.coolaler.com/forums/thr...te-z890-aorus-master-ckd.386911/#post-3463822

ARL seems to mostly match Zen4 non-X3D in gaming.

Arrowlake is pretty disappointing. Not going to lie. Performance seems to have regressed in more situations than not compared to 14th gen. What’s really killing it is the high latency to DRAM. It’s over 100ms in AIDA. Even with 9300 MT/s memory, it’s still in the mid-90s range.

Edit: At least Intel got the power down, even though this is a CPU stress test.

 

[Hitman928]

Another review: https://www.coolaler.com/forums/thr...te-z890-aorus-master-ckd.386911/#post-3463822

ARL seems to mostly match Zen4 non-X3D in gaming.

All tests were at 4k. With that in mind, those are pretty bad results. Most of these tests show a regression. However, the Geekbench single core result suggests there is an issue with the setup, possibly Bios or OS version.

 

[jdubs03]

Another review: https://www.coolaler.com/forums/thr...te-z890-aorus-master-ckd.386911/#post-3463822

ARL seems to mostly match Zen4 non-X3D in gaming.

I know it’s one review (edit: now deleted but likely to be back soon lul) and I said what I said a little bit ago.
But this isn’t a good look:

The Geekbench results are funky (mixes 6.2.2 for the 285K and 9950X and 5.1.0 with the others).

But almost every game is a regression (oddly CP2077 and Wukong aren’t).
I suspect we will see a lot of the same once the embargo lifts.