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.)

 

[Schmide]

100 Celcius = 212 Fahrenheit.

So you don't understand metric. American right?

That's the joke you gleamed from that ?

 

[SiliconFly]

Its amazing that people think Arrow lake S is a dud. I don't think it will be and it will be very efficient.

ARL desktop is getting a massive node jump. From Intel 7 to 20A. Thats more than a 2X node jump. ARL is gonna be a lot more power efficient than the RPL generations. 50% to 60% efficiency gains just from the node jump alone clock-for-clock. Architectural efficiency gains are extra.

As far as performance is concerned, it squarely rests on LNC's performance gains and 20A's Fmax. Both are unknowns as of now. So, it's too soon to say ARL is a dud or ARL is a performance king, etc. It can be anything.

 

[dullard]

ARL desktop is getting a massive node jump. From Intel 7 to 20A. Thats more than a 2X node jump. ARL is gonna be a lot more power efficient than the RPL generations. 50% to 60% efficiency gains just from the node jump alone clock-for-clock. Architectural efficiency gains are extra.

Seems like you didn't get the memo. When Intel is on an older, less efficient node, we are to ignore single thread performance and complain about power & efficiency. When Intel moves to a newer, more efficient node, we are to ignore power & efficiency and complain that it might not match previous single thread performance.

 

[SiliconFly]

Seems like you didn't get the memo. When Intel is on an older, less efficient node, we are to ignore single thread performance and complain about power & efficiency. When Intel moves to a newer, more efficient node, we are to ignore power & efficiency and complain that it might not match previous single thread performance.

If LNC turns out to be good, ST perf may be a lot better than previous gen (even after clock regression). Just sayin'.

 

[Elfear]

Seems like you didn't get the memo. When Intel is on an older, less efficient node, we are to ignore single thread performance and complain about power & efficiency. When Intel moves to a newer, more efficient node, we are to ignore power & efficiency and complain that it might not match previous single thread performance.

I mean I get your sarcasm, some people will complain no matter what. For the past couple gens, Intel has had great performance (at least on desktop) but efficiency has been rather subpar. So if ARL-S brings major efficiency gains, that will be excellent.

I think the sentiments about ARL-S being a dud or "meh" stem from the rumors of overall performance not going up much. Kind of like the reaction to RDNA3. It brought efficiency but not much on the performance front. Enthusiasts want to see longer bar graphs in reviews!

 

[FlameTail]

Perhaps we should be using curves instead of bar graphs

 

[dullard]

I mean I get your sarcasm, some people will complain no matter what. For the past couple gens, Intel has had great performance (at least on desktop) but efficiency has been rather subpar. So if ARL-S brings major efficiency gains, that will be excellent.

I think the sentiments about ARL-S being a dud or "meh" stem from the rumors of overall performance not going up much. Kind of like the reaction to RDNA3. It brought efficiency but not much on the performance front. Enthusiasts want to see longer bar graphs in reviews!

Yes it was sarcasm, pointed at the recent discussion (split between this thread and the foundry node thread) about a 2% drop in ST performance on one 7-Zip benchmark comparing a laptop Meteor Lake to a SFF desktop Raptor Lake.

Both efficiency and performance are important. It does seem like Arrow Lake (at least at first) is focusing on efficiency more than pure performance. Nothing wrong with that. Intel is addressing their biggest weakness.

I'm personally disappointed that they aren't (yet) taking the area savings and adding more cores, so we would get significant efficiency and MT performance gains. I wouldn't call it a dud. I personally think most chips released in the past few years (from all companies) are good chips.

 

[igor_kavinski]

I'm personally disappointed that they aren't (yet) taking the area savings and adding more cores, so we would get significant efficiency and MT performance gains.

I don't think Intel would ever do that unless AMD forces their hand. We would still be getting max 10C/20T CPUs for consumer platforms and having to pay the HEDT tax for more cores if it weren't for AM4/AM5. They just want to churn out as many small dies as they can out of a wafer and increase their bottom line. And they keep shifting the goal posts. Developers were just getting used to the idea of an individual having access to 32 threads but then they threw a kink in that with high perf threads and efficency threads so it's an additional headache for developers to squeeze the maximum performance out of Intel CPUs but they have to do that coz that's where the volume is. AMD finally brought AVX-512 to the average person but then Intel bailed out and they are gonna have their new AVX10 spec which is surely going to try to sabotage AMD's AVX512 performance in one way or the other.

 

[SiliconFly]

Perhaps we should be using curves instead of bar graphs

Actually, a simple line graph is usually sufficient. Intel client *still* has 82% market share (desktop + laptop) in spite of Zen4 competition as of today. Amd only has 18% market share and I'd actually like to know which is the best selling Ryzen cpu in that remaining 18%... cos, the rest don't seem to be worth their development cost. Is amd wasting money on parts that aren't selling? Perhaps they should invest more on stuff that sells.

 

[SiliconFly]

...is focusing on efficiency more than pure performance.

I don't think those rumors are true. ARL performance might actually be better than previous gen. We just have to wait for an ES for more info which is just a few months away.

 

[igor_kavinski]

I hope that this year, both Zen 5 and ARL go for baseline DDR5-6400 stock RAM config. Absolutely hated the reviews where they decided to test stock performance with Intel@DDR5-5600 and AMD@DDR5-5200. That's like comparing two race cars with one of them having an inferior quality tire brand.

 

[H433x0n]

Seems like you didn't get the memo. When Intel is on an older, less efficient node, we are to ignore single thread performance and complain about power & efficiency. When Intel moves to a newer, more efficient node, we are to ignore power & efficiency and complain that it might not match previous single thread performance.

You’re being sarcastic but I’m 99% sure this is what’s going to happen in 6 months. Everybody will determine that 1T performance is the most important metric for a client CPU (which to me it already is, IMO). Being power efficient will fall out of style once they’re at relative parity. Techtubers will give lip service to power efficiency and then dedicate their time to talk about how ARL is unusable because it’s 10% behind in 1T performance.

 

[gdansk]

1T is all that matters for client since 8 cores became ubiquitous. But AMD has magic cache that makes -10% 1T not matter for the most common case where you'd complain about being behind (gaming).
There will be no change in priorities but if Zen 5 does put AMD ahead in 1T even without stacked cache then what hope is there for Intel?

Anyway...

 

[igor_kavinski]

what hope is there for Intel?

No hope in tech circles but the common man basically worships Intel coz that's what 9 out of 10 people around him gush about. Intel will still sell millions of CPUs. The secretary or the clerk isn't suddenly gonna go to their IT and demand AMD because of some MS OFFICE benchmark they saw online.

 

[Markfw]

where i

Actually, a simple line graph is usually sufficient. Intel client *still* has 82% market share (desktop + laptop) in spite of Zen4 competition as of today. Amd only has 18% market share and I'd actually like to know which is the best selling Ryzen cpu in that remaining 18%... cos, the rest don't seem to be worth their development cost. Is amd wasting money on parts that aren't selling? Perhaps they should invest more on stuff that sells.

Where is that graph sourced ? Considering performance of both parties over this timeline, that insane shift in one month seems wrong. I think the reporting or something changed, and is wrong.

 

[igor_kavinski]

Where is that graph sourced ? Considering performance of both parties over this timeline, that insane shift in one month seems wrong. I think the reporting or something changed, and is wrong.

Yeah. And it can't be accurate. Possibly from a limited group of retailers in one geographic region or from some analyst firm who might actually be paid under the table by Intel. Gotta have the fine print enlarged to understand how they arrived at those figures.

 

[poke01]

No hope in tech circles but the common man basically worships Intel coz that's what 9 out of 10 people around him gush about. Intel will still sell millions of CPUs. The secretary or the clerk isn't suddenly gonna go to their IT and demand AMD because of some MS OFFICE benchmark they saw online.

This isn’t true anymore. AMD is king in CPUs and mindshare is now better than Intels.

 

[FlameTail]

This isn’t true anymore. AMD is king in CPUs and mindshare is now better than Intels.

How will things shape up for Qualcomm? Can we get a third wheel?

Or a fourth wheel too? (Nvidia)

 

[igor_kavinski]

AMD is king in CPUs and mindshare is now better than Intels.

Not in a lot of places outside the US.

My current strategy to making people believe that AMD is not some small company is to make them do a google search for AMD market cap and then compare with Intel's. They are usually taken aback but I can see behind their faces that it still didn't fully convince them. And these are usually people who don't give much credence to benchmarks.

On a side note, buy INTEL! You might get rich when their stock price starts climbing

 

[FlameTail]

Hmm

 

[Hitman928]

where i

Where is that graph sourced ? Considering performance of both parties over this timeline, that insane shift in one month seems wrong. I think the reporting or something changed, and is wrong.

It's Puget Systems customer orders. Basically market share from a boutique builder of workstation systems. It is an extremely small part of the market overall.

 

[SiliconFly]

where i

Where is that graph sourced ? Considering performance of both parties over this timeline, that insane shift in one month seems wrong. I think the reporting or something changed, and is wrong.

Why? If Intel has 82% market share, it has to be wrong, right? See it for yourself... link

 

[SiliconFly]

This isn’t true anymore. AMD is king in CPUs and mindshare is now better than Intels.

Nope. AMD mind share in only among personal PC builders, gamers and enthusiasts that too in very specific parts of the world. A very tiny market. Majority of the office and homes around the world still prefer Intel.

 

[Schmide]

The real girth of those numbers is edge computing. intel rules this space

 

[DAPUNISHER]

Nope. AMD mind share in only among personal PC builders, gamers and enthusiasts that too in very specific parts of the world. A very tiny market. Majority of the office and homes around the world still prefer Intel.

Correct. Intel still dominates the mainstream PC pre-built markets. Especially laptops and notebooks.