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	12 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)	Lunar Lake	Panther Lake
Platform	Mobile H/U Only	Desktop Only	Desktop & Mobile H&HX	Mobile U Only	Mobile H
Process Node	Intel 4	Intel 20A	TSMC N3B	TSMC N3B	Intel 18A
Date	Q4 2023	Q1 2025 ?	Desktop-Q4-2024 H&HX-Q1-2025	Q4 2024	Q1 2026 ?
Full Die	6P + 8P	6P + 8E ?	8P + 16E	4P + 4E	4P + 8E
LLC	24 MB	24 MB ?	36 MB ?	12 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.)

 

[511]

Th Zen 6 32 core variant is the even weaker C cores. I remember reading it somewhere on how they would get 32 onto one CCD with upcoming Zen 6. But maybe things change do not know. But if Zen 6 is coming 2025, hard to imagine they could get 32 into one CCD without making them even weaker C cores/

Zen 6 is not coming before 2026 Typical AMD schedule anyway LNL NDA is 24 days away

 

[OneEng2]

It is a updated core on entirely new node with GAA and BSPD so i would call it new otherwise crestmont wouldn't be a new core ?

Agree. The point I was trying to make is 18A is a very high risk node.

Referring to glue? Like how are things glued and how intel does it.

I thought a process node was all an SOC and no glue?

I mean technically didn't Intel glue two Pentium 4 chips into one package to make the Pentium D and thus they were not a true dual core but rather a dual socket Pentium 4s in one package. Also kind of like how AMD currently glues 2 5800X/7700X/5600X/7600X/9700X/9600X CCDs into one paclkage chip to make the current Ryzen 9 parts??

I specifically meant digital glue . Not actual glue.

Also, seems like AMD have released a new bios update that greatly lowers the CCD to CCD latency many have mentioned.

You expect people to find this position acceptable? That's the core of assessing whether a company is moral, immoral, or simply amoral.

No. I personally find it unacceptable. I am a high level manager and have been in several companies. This is how it is done. Investors don't care about the relative morality of a company, only how it performs financially. It is brutal, but it is true. I do find it amusing that some of the same people that detest company morals are also decrying federal regulation. It's like people expect companies to behave like a loving parent .

 

[511]

Agree. The point I was trying to make is 18A is a very high risk node.

Well they derisked BSPDN GAA is the only main risk alongside yields

Intel Details PowerVia Chipmaking Tech: Backside Power Performing Well, On Schedule For 2024

www.anandtech.com

No. I personally find it unacceptable. I am a high level manager and have been in several companies. This is how it is done. Investors don't care about the relative morality of a company, only how it performs financially. It is brutal, but it is true. I do find it amusing that some of the same people that detest company morals are also decrying federal regulation. It's like people expect companies to behave like a loving parent .

Exactly lawsuit doesn't matter it's a matter of by doing this or that how much did you gain in profit

 

[cebri1]

They had an Intel 4 hybrid to de-risk 20A and 18A.

 

[SiliconFly]

They had an Intel 4 hybrid to de-risk 20A and 18A.
View attachment 107950

It's basically a variant of Intel 4 with PowerVia to de-risk their BSPDN. Not RibbonFET (GAAFET). Hence, only partially de-risks 20A/18A.

 

[SiliconFly]

Last time i9-14900HX same core count. HX is desktop transferred to BGA, this is not a real mobile CPU.

Doesn't matter you feel that way. I have a HX laptop & the gaming perf is pure awesome (and mine is just a Raptor 8+16 with RTX). I'm sure the newer ARL-HX will just kill it, leaving the competition in the dust.

Oh edit that they do with the 13900HX Raptor Lake after research.

The HX series is very power hungry. And for best gaming perf, it needs to plugged in. But it does what it does best... eXtreme Performance. The regular H series won't even come near it.

 

[SiliconFly]

Referring to glue? Like how are things glued and how intel does it.

Zen 5 uses the old Infinity Fabric. Basically an outdated point-to-point interconnect that resides directly on the substrate itself.

Intel Lunar Lake & Arrow Lake both use the cutting-edge Coherent Fabric Interconnect (similar to MTL's NOC Fabric) that sits directly on top the base tile and is connected using Foveros. Basically an IP network-like architecture (where traffic is switched). It's light years ahead of Infinity Fabric.

RWC is much better than most people think because launch microcode was horrible. Perf/mhz increased after several updates.

Maybe so. But RWC didn't bring in any IPC increase, hence it's a lost cause. And worse, at launch, it had regression. I still wish RWC didn't exist.

 

[cebri1]

HUB saying the ARL announcement is expected to happen the 10th of October and the release will follow the 24th. They also said there hasn't been any delay, the 24th date has been more or less known for some time.

 

[DavidC1]

Maybe so. But RWC didn't bring in any IPC increase, hence it's a lost cause. And worse, at launch, it had regression. I still wish RWC didn't exist.

You keep saying this over and over but oftentimes a refresh of a core on a new process is necessary which is what Redwood Cove is.

 

[DavidC1]

I do find it amusing that some of the same people that detest company morals are also decrying federal regulation. It's like people expect companies to behave like a loving parent .

Because government is the penultimate monopoly with near zero consequences to it's actions.

Rather than deregulation things would be better solved if certain crimes couldn't be *cough* bribed *cough* lobbied and/or you can pay your way out of jail and/or you fine the company rather than fining/jailing the individuals responsible.

Volkswagen was fined a massive amount and most of the individuals responsible basically got off with a mere slap in the wrist. Let's say instead they were put in prison regardless of position or fined relative to their earnings, meaning a millionaire would be fined 10x compared to someone having 10x less.

 

[511]

LNL Blender benchmark

 

[jdubs03]

Figured I’d attach too as this is a related 258V posting (Blender CPU).
Closest analogy would be the 155U (14 threads) at 15W.

Seems like the 165U is an extreme outlier, considering it should be faster than the 155U, and somewhere between the 155U and 155H.