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.)

 

[cebri1]

Not sure...
Is "superior per core performance" something you would claim for a 128 core CPU? Sounds more like an Argument for Desktop CPUs to me.

Turin goes up to 192C, that could explain the reference to “per Core performance”. Lunar Lake review embargo is also lifted the same day.

 

[cannedlake240]

https://twitter.com/x/status/1836267219032187387

MTL also crushes local playback

 

[511]

MTL also crushes local playback

Not by this much meteor lake is still way better than RPL-H and on par with zen 4 in terms of Battery life btw this is official lenovo YouTube 🤣 they broke the NDA 🤣🤣

 

[cebri1]

https://videocardz.com/newz/intel-core-ultra-200-arrow-lake-s-to-support-10000-mt-s-cudimm-memory

 

[SteinFG]

So, cudimm or camm2. Seems like intel's pushing for cudimm, since it's compatible with what we have now. Does camm2 have its own clock gen like cudimm or no?

 

[bakyt115]

cb 23 multy

 

[SiliconFly]

They lied about 20A and ARL, why would anyone trust Intel.

No. They never lied. Plans changed due to financials. People still trust Intel more than AMD.

They kept saying everything was on track until the last minute, obviously they lied.

Not worse than AMD Zen 5% lies!!!

Except one thing. When was the last time Intel told the truth ?

Like again, Intel has more truth than Zen 5%

When was the last time they did something good for customers ?

Past 4 decades they've done more good than AMD can ever dream of.

 

[poke01]

View attachment 107761

cb 23 multy

using cb r23 and not cb 2024 and using the 15" Air is something as the M3 scores much higher in 14" Pro. cb 2024 is more optimised for ARM.

 

[511]

if we are talking about good there is no company who has done as good as intel minus the last decade ofc they literally put silicon in Silicon valley

 

[poke01]

Also they ran that test plugged in. I wonder what the unplugged results are

 

[511]

Same

 

[OneEng2]

In fairness, companies are not moral or imoral, they are amoral. They all strive to maximize profit. Nothing wrong with that at all in my book. It is a very good argument for strong competition. No one should wish for one or the other company to encounter rough roads leading to company failure. That would definitely not be good for consumers.

To me, Intel dropping 20A signals a lack of time and resources. My guess is that they are concentrating on 18A; however, 20A was a de-risking node to make 18A more certain to succeed. Dropping 20A seems like a red flag for 18A in my thinking.

18A includes so many new things. It reminds me of Intel's 10nm, and we all know how that turned out.

It is my understanding that Samsung's GAA has suffered from yield issues. Not sure if this is still true or not; however, Samsung's yield issues also raise the red flag in my book. Sure, they were first to make it, but it seems like it is still pretty tricky.

So, I still believe that the forward processor designs at Intel are relying on the 18A advanced processing node to achieve their design goals and that these designs are likely not possible on TSMC N3E. If true, this doesn't just place the Intel foundry business at risk, it further impacts the competitive landscape of their future server and client processors as well.

I am very much looking forward to the Arrow Lake reviews. Keeping in mind that Arrow Lake enjoys a pretty decent bump in process technology over Zen 5 (N3B vs N4P?), It should enjoy either a performance, or power (or both) advantage over Zen 5. If it does not, I would say it is not a good sign for Intel.

 

[inquiss]

In fairness, companies are not moral or imoral, they are amoral. They all strive to maximize profit. Nothing wrong with that at all in my book. It is a very good argument for strong competition. No one should wish for one or the other company to encounter rough roads leading to company failure. That would definitely not be good for consumers.

To me, Intel dropping 20A signals a lack of time and resources. My guess is that they are concentrating on 18A; however, 20A was a de-risking node to make 18A more certain to succeed. Dropping 20A seems like a red flag for 18A in my thinking.

18A includes so many new things. It reminds me of Intel's 10nm, and we all know how that turned out.

It is my understanding that Samsung's GAA has suffered from yield issues. Not sure if this is still true or not; however, Samsung's yield issues also raise the red flag in my book. Sure, they were first to make it, but it seems like it is still pretty tricky.

So, I still believe that the forward processor designs at Intel are relying on the 18A advanced processing node to achieve their design goals and that these designs are likely not possible on TSMC N3E. If true, this doesn't just place the Intel foundry business at risk, it further impacts the competitive landscape of their future server and client processors as well.

I am very much looking forward to the Arrow Lake reviews. Keeping in mind that Arrow Lake enjoys a pretty decent bump in process technology over Zen 5 (N3B vs N4P?), It should enjoy either a performance, or power (or both) advantage over Zen 5. If it does not, I would say it is not a good sign for Intel.

It was derisking. The more interesting point for me is that the goals of 18a are much more watered down than they were a while ago. 18a has benefits much more in like with what 20a was supposed to offer Vs intel 3.

 

[Magio]

View attachment 107761

cb 23 multy

I'm curious about average package power during this benchmark. It's the 258v so if it's doing it mainly at the 17W PL1 that's great, if it average 30W that'd be a bit underwhelming.

 

[511]

It will be around 25-30W i guess Lunar lake Stop ls scaling past 21-22W in terms of CPU even from Intel's PPT.

 

[9949asd]

Looks like ARL will launch in 6 days anyway, the old Intel Innovation Slot.

That’s Lunar lake release day.

 

[511]

That’s Lunar lake release day.

Lunar lake already released it is Xeon 6 as it was q3 they can maybe announce arl at best

Intel Xeon 6 6700E Sierra Forest Shatters Xeon Expectations

The Intel Xeon 6 "Sierra Forest" is here shattering expectations on what it means to be a Xeon with 144 cores and very low power consumption

www.servethehome.com

 

[9949asd]

Lunar lake already released it is Xeon 6 as it was q3 they can maybe announce arl at best

Intel Xeon 6 6700E Sierra Forest Shatters Xeon Expectations

The Intel Xeon 6 "Sierra Forest" is here shattering expectations on what it means to be a Xeon with 144 cores and very low power consumption

www.servethehome.com

lunar lake was only announced, 24th is release to market and allow to publish testing data.

 

[511]

lunar lake was only announced, 24th is release to market and allow to publish testing data.

Nope it is already released 24 is date of shipping to customer you can already pre order also the NDA date launch/release doesn't guarantee product in your day the same day also it is already released to makrket

 

[Philste]

That’s Lunar lake release day

It says announcement, so it's not related to LNL.

Lunar lake already released it is Xeon 6 as it was q3 they can maybe announce arl at best

Probably both. I mean why should they switch launch date because of Innovation Event not Happening? ARL and GNR would've both be presented there so why shouldn't they keep the date the same?

 

[9949asd]

View attachment 107761

cb 23 multy

What was the leaked score? Only 9k?

 

[Joe NYC]

No this is false, they simply ignored talking about 20A for months. It was on track maybe a year ago or something like that. It's not like they claimed on track shortly before they cancelled it you are implying.

So, we need to watch for the dog that doesn't bark.

 

[Magio]

What was the leaked score? Only 9k?

It was 8.2K at 17W PL1/17W PL2 and low 10k/high 9k 30W PL1/30W PL2.

This is 258v so presumably 17W PL1/30W PL2 but it's a higher score than we've seen in any prior leaked score. Average package power over the benchmark run still tbd.

 

[9949asd]

It was 8.2K at 17W PL1/17W PL2 and low 10k/high 9k 30W PL1/30W PL2.

This is 258v so presumably 17W PL1/30W PL2 but it's a higher score than we've seen in any prior leaked score. Average package power over the benchmark run still tbd.

Yea way higher than the leak.

 

[mikk]

It was 8.2K at 17W PL1/17W PL2 and low 10k/high 9k 30W PL1/30W PL2.

This is 258v so presumably 17W PL1/30W PL2 but it's a higher score than we've seen in any prior leaked score. Average package power over the benchmark run still tbd.

This is not 17W PL1, score too high. Maybe 30W PL1 with 37W PL2.