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

[Tigerick]

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.)

 

[igor_kavinski]

I hope the memory part is outdated by now and LNL-MX actually debuts with 24/48GB options.

 

[Goop_reformed]

Uhh not looking good for intel foundry. Made on N3B lol

 

[rtxtwt]

This LNL leak is a catastrophe to Intel according to some leaker's statement. It's even more disasterous than ARL performance slides few months ago. Let's hope the guy who leak this won't get fired.

I hope the memory part is outdated by now and LNL-MX actually debuts with 24/48GB options.

These slides are just few months old.

 

[tamz_msc]

This LNL leak is a catastrophe to Intel according to some leaker's statement. It's even more disasterous than ARL performance slides few months ago. Let's hope the guy who leak this won't get fired.

Uhh not looking good for intel foundry. Made on N3B lol

LOL what, why?

I hope the memory part is outdated by now and LNL-MX actually debuts with 24/48GB options.

It cannot, because LPDDR5x density is either 64 Gb or 128 Gb, so only those configs are possible.

 

[tamz_msc]

More BW and capacity on the higher end SKUs than Apple M3. Same CPU config - 4P+4E as the M3. If Apple stagnates on P-core gen on gen performance improvements, and Intel gains momentum with Lunar Lake MX and its followers, Apple is gonna regret going the ARM route.

 

[Joe NYC]

I hope the memory part is outdated by now and LNL-MX actually debuts with 24/48GB options.

I don't think it is going to be a huge problem. 32 GB will be the sweet spot, 16 GB for entry level system.

Apple is selling 8 GB laptops...

 

[Joe NYC]

This LNL leak is a catastrophe to Intel according to some leaker's statement. It's even more disasterous than ARL performance slides few months ago. Let's hope the guy who leak this won't get fired.


These slides are just few months old.

Why do you think this is catastrophic? It is only catastrophic for Intel foundry, but otherwise, without knowing the performance, it looks fine to me.

 

[mikk]

So Lunar Lake MX will integrate the CPU and GPU on one tile?

Yes it will.

40% scenario power reduction over MTL-U T4 sounds big whatever it exactly means. VVC HW decoding for the first time.

16GB is fine, for the majority of people this will be enough. This is not a high end gaming or workstation device at the end of the day. For those people who needs more there is a 32GB option.

 

[Joe NYC]

So Lunar Lake MX will integrate the CPU and GPU on one tile?

It seems like a more optimum breakdown between N3 and (likely) N6 for SoIC. At least compared to Strix Halo, which has the SoC with I/O + graphics on N3 (apparently) together with analog stuff and the MALL SRAM...

 

[coercitiv]

I was about to make a joke, but then I decided to check Videocardz and the joke had become reality:

 

[Goop_reformed]

LOL what, why?


It cannot, because LPDDR5x density is either 64 Gb or 128 Gb, so only those configs are possible.

I mean intel had to fab their lead chips 2 gens in a row on n3b I don't know what to tell you.

 

[tamz_msc]

I mean intel had to fab their lead chips 2 gens in a row on n3b I don't know what to tell you.

Eh, first define what you mean by "lead chip". They are pursuing the strategy of different nodes for different products. Arrow Lake is 20A, Granite Rapids is 3nm, MTL 4nm etc.

Only Intel has the luxury of not having to be wholly dependent on TSMC or Samsung.

 

[Joe NYC]

Eh, first define what you mean by "lead chip". They are pursuing the strategy of different nodes for different products. Arrow Lake is 20A, Granite Rapids is 3nm, MTL 4nm etc.

Only Intel has the luxury of not having to be wholly dependent on TSMC or Samsung.

Arrow Lake is N3B, some 20A, and other chiplets are all TSMC.
Lunar Lake is all TSMC.

So leading desktop and leading client product are going to be mostly on TSMC precisely when Intel should be completing its "5 nodes in 4 years".

 

[tamz_msc]

Arrow Lake is N3B, some 20A, and other chiplets are all TSMC.
Lunar Lake is all TSMC.

So leading desktop and leading client product are going to be mostly on TSMC precisely when Intel should be completing its "5 nodes in 4 years".

Arrow Lake CPU tile is 20A. Granite Rapids is 3nm. Lunar Lake is obviously catered to premium ultrabooks that compete with Macs, so it will be low volume. So the most important products are still on Intel nodes.

And I don't think that 5 nodes in 4 years means Intel will have lead products in segments for each of those nodes. The plan is to regain process leadership.

 

[Joe NYC]

Arrow Lake CPU tile is 20A.

But also TSMC N3. Where else do you think Intel's big order for TSMC N3 is going to?

According to the press report, it was a sizeable order that Intel placed for TSMC N3 capacity.

 

[tamz_msc]

But also TSMC N3. Where else do you think Intel's big order for TSMC N3 is going to?

According to the press report, it was a sizeable order that Intel placed for TSMC N3 capacity.

That still doesn't imply "two generations of lead products" on non-Intel nodes, as the other commenter was saying.

 

[Kinshasa]

OK, WOAH!!! The thing that jumps out at me is real time raytracing! What the heck is that??? Does that mean zero cost raytracing in any scene????

No, it will have hardware accelerated ray tracing, but while this is a step in the right direction, it is probably not enough to do anything too interesting. It is going to still be a while before hardware allows for that kind of ray tracing.

 

[Goop_reformed]

Eh, first define what you mean by "lead chip". They are pursuing the strategy of different nodes for different products. Arrow Lake is 20A, Granite Rapids is 3nm, MTL 4nm etc.

Only Intel has the luxury of not having to be wholly dependent on TSMC or Samsung.

So now the question points to what happened to intel foundry that they INCREASED reliance on TSMC. I mean at some point they gotta fab their own lead chips. Granite Rapids aren't coming out soon to be honest.

 

[tamz_msc]

So now the question points to what happened to intel foundry that they INCREASED reliance on TSMC. I mean at some point they gotta fab their own lead chips. Granite Rapids aren't coming out soon to be honest.

Granite Rapids is H2 2024. Unless AMD announces Turin at the upcoming December 6 event, Intel and AMD are going to release their server lineup close to one another.

 

[Joe NYC]

That still doesn't imply "two generations of lead products" on non-Intel nodes, as the other commenter was saying.

Well, not "two generations", but I would say two leading products (Arrow Lake and Lunar Lake).

At the same time, Raptor Lake and Meteor lake will be trailing / legacy products.

 

[igor_kavinski]

At the same time, Raptor Lake and Meteor lake will be trailing / legacy products.

MTL a legacy product even before it's launched

 

[FlameTail]

"MTL is a legacy product". How so?

 

[Joe NYC]

MTL a legacy product even before it's launched

We will see. It should be an improvement over Raptor Lake for notebooks.

 

[cebri1]

"MTL is a legacy product". How so?

Simple: “Trolling”

 

[Joe NYC]

"MTL is a legacy product". How so?

Not when it is launched, but when Lunar Lake is launched.