Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads

Tigerick · Aug 22, 2022

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.

Comparison of upcoming Intel's U-series CPU: Core Ultra 100U, Lunar Lake and Panther Lake

Model	Code-Name	Date	TDP	Node	Tiles	Main Tile	CPU	LP E-Core	LLC	GPU	Xe-cores
Core Ultra 100U	Meteor Lake	Q4 2023	15 - 57 W	Intel 4 + N5 + N6	4	tCPU	2P + 8E	2	12 MB	Intel Graphics	4
?	Lunar Lake	Q4 2024	17 - 30 W	N3B + N6	2	CPU + GPU & IMC	4P + 4E	0	8 MB	Arc	8
?	Panther Lake	Q1 2026 ?	?	Intel 18A + N3E	3	CPU + MC	4P + 8E	4	?	Arc	12

Comparison of die size of Each Tile of Meteor Lake, Arrow Lake, Lunar Lake and Panther Lake

	Meteor Lake	Arrow Lake (20A)	Arrow Lake (N3B)	Arrow Lake Refresh (N3B)	Lunar Lake	Panther Lake
Platform	Mobile H/U Only	Desktop Only	Desktop & Mobile H&HX	Desktop Only	Mobile U Only	Mobile H
Process Node	Intel 4	Intel 20A	TSMC N3B	TSMC N3B	TSMC N3B	Intel 18A
Date	Q4 2023	Q1 2025 ?	Desktop-Q4-2024 H&HX-Q1-2025	Q4 2025 ?	Q4 2024	Q1 2026 ?
Full Die	6P + 8P	6P + 8E ?	8P + 16E	8P + 32E	4P + 4E	4P + 8E
LLC	24 MB	24 MB ?	36 MB ?	?	8 MB	?
tCPU	66.48
tGPU	44.45
SoC	96.77
IOE	44.45
Total	252.15

Intel Core Ultra 100 - Meteor Lake

As mentioned by Tomshardware, TSMC will manufacture the I/O, SoC, and GPU tiles. That means Intel will manufacture only the CPU and Foveros tiles. (Notably, Intel calls the I/O tile an 'I/O Expander,' hence the IOE moniker.)

Ghostsonplanets · Apr 9, 2024

uzzi38 said:
If everybody loves and wants LNL, what do you think happens to ARL-U?

Why even bother having a successor to the cheaper (but not cheaper than RPL-U) mobile die that nobody really wants?

I dunno, to me it makes perfect sense for LNL style designs to cover the entire -U laptop range, excluding cheaper designs.

That's actually what I'm expecting. That LNL replaces the U series. I'm guessing Intel will use LNL sales performance and users/media impression as a way to gauge how they will move forward.

igor_kavinski · Apr 9, 2024

If that's <=47% ST, that would be a WOW.

<=47% MT compared to MTL-U, hmm ok not bad.

<=47% perf/w? Still good enough I guess.

Ghostsonplanets · Apr 9, 2024

igor_kavinski said:
If that's <=47% ST, that would be a WOW.

<=47% MT compared to MTL-U, hmm ok not bad.

<=47% perf/w? Still good enough I guess.

They're comparing current Intel Core Ultra (MTL) vs a 3 year old PC (So TGL or Cezanne) and claiming 47% improvement on unspecified workloads (ST?MT? AI? GPU? Perf/W? Battery Life? All of them combined on some weird test set?). Don't read much on it tbh.

DavidC1 · Apr 9, 2024

igor_kavinski said:
If that's <=47% ST, that would be a WOW.

<=47% MT compared to MTL-U, hmm ok not bad.

<=47% perf/w? Still good enough I guess.

Still using the stupid oddly-specific number rather than the general one they used to give. Damn marketers.

SpudLobby · Apr 9, 2024

uzzi38 said:
If everybody loves and wants LNL, what do you think happens to ARL-U?

Why even bother having a successor to the cheaper (but not cheaper than RPL-U) mobile die that nobody really wants?

I dunno, to me it makes perfect sense for LNL style designs to cover the entire -U laptop range, excluding cheaper designs.

Agree, Uzzi. If LNL is even a B-, but A for Intel’s mainstream lineup, which is where it seems we’re headed — then to me they might as well “simplify” and just have one low-power optimized (see system cache etc) part succeeding LNL they follow up on for the entire 5-28W range.

Hopefully it does alright and this is what happens

SiliconFly · Apr 9, 2024

dullard said:
Intel is extremely limited in each node's production quantity. So, Intel has a choice: (A) extreme shortages of processors or (B) processors from multiple different nodes at the same time. They chose route (B) which can work if they are priced correctly.

Makes sense.

SiliconFly · Apr 9, 2024

FlameTail said:
Intel's strategy is to flood the market with options, even if it makes no sense.

Actually igor (I think) recently used the term "SKU spamming" to accurately describe what Intel does.

igor_kavinski · Apr 9, 2024

SiliconFly said:
Actually igor (I think) recently used the term "SKU spamming" to accurately describe what Intel does.

Not this Igor

But SKU spamming is indeed an apt description of what Intel does.

Ghostsonplanets · Apr 9, 2024

SKU spamming is one of the things Gelsinger wants to stop with the new reorganization and separation of Intel Design Teams and Intel Foundry/External. Because now IDT will need to be much more conscious of their designs to reduce costs.

Ghostsonplanets · Apr 9, 2024

https://twitter.com/x/status/1777808830836998426

"With Lunar Lake, the client division expects to get back leadership across most/all workload categories, with AI performance being the most critical. This product is an important step to stave off competition from
@amd and @Snapdragon. I know that @MJHolthaus is driving the team to build a killer product here!"

It seems Intel has some lofty goals with Lunar Lake. Let's see if they pan out.

Markfw · Apr 9, 2024

Ghostsonplanets said:
It seems Intel has some lofty goals with Lunar Lake. Let's see if they pan out.

Intel says lofty things every time they says something. And for the last 7 years or so, they have been wrong.

Not holding my breath.

gdansk · Apr 9, 2024

Wow, LNL is going to have leadership over Apple's M4 in most/all workload categories. It will make MTL look totally obsolete then. Or maybe Apple is not relevant whenever they don't want them to be.

Hulk · Apr 9, 2024

igor_kavinski said:
If that's <=47% ST, that would be a WOW.

<=47% MT compared to MTL-U, hmm ok not bad.

<=47% perf/w? Still good enough I guess.

I was thinking the same thing. Probably 47% over MTL's NPU or some other nonsense that matters not.

Hitman928 · Apr 9, 2024

Hulk said:
I was thinking the same thing. Probably 47% over MTL's NPU or some other nonsense that matters not.

47% more hopes and dreams.

gdansk · Apr 9, 2024

Hulk said:
I was thinking the same thing. Probably 47% over MTL's NPU or some other nonsense that matters not.

It's 165H compared to 1185G7 in SYSmark 30 Overall score.
🤣

Ghostsonplanets · Apr 9, 2024

gdansk said:
Wow, LNL is going to have leadership over Apple's M4 in most/all workload categories. It will make MTL look totally obsolete then. Or maybe Apple is not relevant whenever they don't want them to be.

That's very unlikely to happen. But it will indeed make MTL obsolete at LNL intended power range (7 - 28W). Bionic has said LNL 17W > 1.5x MTL 15W MT.

Hulk said:
I was thinking the same thing. Probably 47% over MTL's NPU or some other nonsense that matters not.

They're actually comparing MTL with a 3 year old PC and quoting a non-specified 47% uplift.

poke01 · Apr 9, 2024

gdansk said:
Or maybe Apple is not relevant whenever they don't want them to be.

Apple is relevant to the market Intel is targeting with LNL. Tablets, thin and light laptops are Apples forte.

Intel only compares with Apple when Apple is behind. Because right now Apple is ahead of Intel in every metric on mobile.

The last time when Intel compared to Apple was with HX 13th gen CPUs against the M2 Max. Intel had a huge lead in MT and it made slides to market that.

With the M3 Max being better than Intel 14th Gen i9 HX CPUs this time they skipped comparing to the M3 Max.

Ghostsonplanets · Apr 9, 2024

poke01 said:
Apple is relevant to the market Intel is targeting with LNL. Tablets, thin and light laptops are Apples forte.

Yup. The entire motivation of LNL is basically to get back leadership in perf/W and showcase that Intel can also build an impressive PPC/PPW SoC to be employed on T&L fanless designs. Except that LNL is coming 4 years later than M1, is more expensive to manufacture and is also using a newer node process😂.

Jest aside, I do believe LNL is a good endeavor for Intel. It's basically what they should have aimed since eons for mainstream mobile T&L. Had they continued the Core M/Y series and pushed them to be the main series for 15W and below, Intel wouldn't be known and memed as the company you choose when you don't care about battery life. Nor would have been Apple M1 as impactful. But alas, better later than never.

poke01 · Apr 9, 2024

Ghostsonplanets said:
Yup. The entire motivation of LNL is basically to get back leadership in perf/W and showcase that Intel can also build an impressive PPC/PPW SoC to be employed on T&L fanless designs. Except that LNL is coming 4 years later than M1, is more expensive to manufacture and is also using a newer node process😂.

Jest aside, I do believe LNL is a good endeavor for Intel. It's basically what they should have aimed since eons for mainstream mobile T&L. Had they continued the Core M/Y series and pushed them to be the main series for 15W and below, Intel wouldn't be known and memed as the company you choose when you don't care about battery life. Nor would have been Apple M1 as impactful. But alas, better later than never.

Honestly a lot big breakthroughs made a big impact due to Intel being behind or ignorant.

Zen, M1 and rise of ARM in smartphones are just a few. Both Intels design teams and its foundries are at fault.

Truly a mismanaged company before.

Ghostsonplanets · Apr 9, 2024

poke01 said:
Honestly a lot big breakthroughs made a big impact due to Intel being behind or ignorant.

Zen, M1 and rise of ARM in smartphones are just a few. Both Intels design teams and its foundries are at fault.

Truly a mismanaged company before.

Indeed. Apple cores started to truly approach (and soon surpass) Intel x86 with A9. Everyone, from foundry (TSMC) to design (Apple, AMD, Arm) started to execute and execute non-stop from 2015 onwards. While Intel stalled. Now they pay the price of it: Their P Core design teams still follow the Desktop first approach of "Performance at any price" and Foundry is behind industry best (TSMC). It will take time for them to get their feet on the ground and running.

But the fact Intel is still somewhat competitive is a testament of how solid their foundation was. Just that they need to act as the foundation doesn't turn into mud.

FlameTail · Apr 9, 2024

Ghostsonplanets said:
"With Lunar Lake, the client division expects to get back leadership across most/all workload categories, with AI performance being the most critical. This product is an important step to stave off competition from
@amd and @Snapdragon. I know that @MJHolthaus is driving the team to build a killer product here!"

LOL. Did they actually ping AMD and Snapdragon?

FlameTail · Apr 9, 2024

Ghostsonplanets said:
Their P Core design teams still follow the Desktop first approach of "Performance at any price"

Which also explains their bloat and addiction to using HP process libraries.

Hulk · Apr 9, 2024

Ghostsonplanets said:
That's very unlikely to happen. But it will indeed make MTL obsolete at LNL intended power range (7 - 28W). Bionic has said LNL 17W > 1.5x MTL 15W MT.

They're actually comparing MTL with a 3 year old PC and quoting a non-specified 47% uplift.

Why go 3 year? Why not compare to a Pentium and show us 5000% increase!

Ghostsonplanets · Apr 9, 2024

FlameTail said:
LOL. Did they actually ping AMD and Snapdragon?

It was actually Ryan Shrout (I don't think he's at Intel anymore), but indeed he did.

FlameTail said:
Which also explains their bloat and addiction to using HP process libraries.

Yup.

Hulk said:
Why go 3 year? Why not compare to a Pentium and show us 5000% increase!

They actually compared Xeon 6 (GNR/SRF) to a Xeon Scalable 2nd Gen (Cascade Lake/Skylake Server ++ on 14nm) and claimed 1MW power savings. So you're not that far off hahaha

mikk · Apr 9, 2024

Tigerick said:
Panther Lake is new design with 2 tiles only. 18A is for CPU+SoC, tGPU is available for 2 version: Intel 3 for 4 XE and N3E for 12 XE

And what about the PCD tile?

Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads

Senior member

Attachments

Senior member

Lifer

Senior member

Senior member

Senior member

Golden Member

Golden Member

Lifer

Senior member

Senior member

Moderator Emeritus, Elite Member

Platinum Member

Diamond Member

Diamond Member

Platinum Member

Senior member

Senior member

Senior member

Senior member

Senior member

Platinum Member

Platinum Member

Diamond Member

Senior member

Diamond Member