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

LightningZ71 · Oct 26, 2024

The enlarged L2 on Arrow lake P cores as opposed to Lunar Lake likely wasn't so much about boosting core performance as it was about trying to PRESERVE core performance by hiding more of the L3 latency.

SolidQ · Oct 26, 2024

tsamolotoff said:
in some more GPU-heavy game segments ARL looks nice, but when the game starts to put more pressure on the CPU, situation is drastically reversed.

When i had FX8350, same situation was, in GPU scenario was fine, in CPU bad

Meteor Late · Oct 26, 2024

LightningZ71 said:
The enlarged L2 on Arrow lake P cores as opposed to Lunar Lake likely wasn't so much about boosting core performance as it was about trying to PRESERVE core performance by hiding more of the L3 latency.

Yeah good luck with that in games lol

Bouowmx · Oct 26, 2024

Kocicak said:
I tested a power limited Arrow lake here:

Testing ARROW lake - power limited

I got myself a testing PC with 265K, Strix E board and ordinary RAM (32GB, 6000 MHz, CL36) and tested the CPU with two power limits 100 and 160W, and compared it to 13900KS with 4 E cores disabled, both with HT on and off. As you can see, the results are pretty consistent, both apps in both...

www.techpowerup.com

HT off at the silicon level is such a bizarre move.
If possible, at the system level, can you test whether DLVR Power Gate mode with static core voltage is more efficient?

511 · Oct 26, 2024

https://twitter.com/x/status/1850060440887570884

E core won boys

Kocicak · Oct 26, 2024

Bouowmx said:
HT off at the silicon level is such a bizarre move.

It is not. HT threads were the weakest ones and in the spectrum of all possible loads that CPUs can experience in home PCs they got almost never used. Only the heaviest multithreaded applications used those threads and people running such apps would be better off running a threadripper system anyway. HT is useless today, when you have up to 24 core CPUs in home PCs.

Magio · Oct 26, 2024

511 said:
https://twitter.com/x/status/1850060440887570884
E core won boys

So if Darkmont is a rather minimal upgrade as has been said, it's fair to expect Arctic Wolf and Golden Eagle (is that the first time this is mentioned?) to be big ones if they have to pave the way for the E core to take over the P core's duties after them.

Saylick · Oct 26, 2024

511 said:
https://twitter.com/x/status/1850060440887570884
E core won boys

Will be interesting to see how the convergence plays out. I have to imagine the size of the E core slowly grows until it basically becomes roughly the size of the competing Zen core, mostly because Intel’s P cores are bloated to begin with (AMD has been able to get similar levels of performance in a smaller die area) and E cores will need a lot more xtors to have feature parity with Zen, namely AVX512, but also clock parity.

cannedlake240 · Oct 26, 2024

Magio said:
So if Darkmont is a rather minimal upgrade as has been said, it's fair to expect Arctic Wolf and Golden Eagle (is that the first time this is mentioned?) to be big ones if they have to pave the way for the E core to take over the P core's duties after them.

Maybe the e core will gradually evolve to become larger and larger until Intels final hybrid architecture is just two different P cores but one of them is a lower clocked version of the Unified core, and the next gen could then feature a full fledged high frequency variant

igor_kavinski · Oct 26, 2024

511 said:
https://twitter.com/x/status/1850060440887570884
E core won boys

I really hope that's true. Means they came to their senses finally.

OneEng2 · Oct 26, 2024

desrever said:
Intel would have to lose for many many years for AMD to get to 50% market share then.

They are certainly working hard in that direction though, don't you think?

Thunder 57 said:
Intel seems to be doing enough damage on its own to make that happen. Their 18A better be as promised. The other threat is ARM. I like x86 because of its backwards compatibility. Also the fact that it isn't a "walled garden".

I believe ARL is good enough to minimize the damage, but another 14+++++++++ situation might actually do Intel in to the point of being bought out.

TESKATLIPOKA said:
265K and 245K are the same chip as 285K If I am not mistaken.
If there is a supply issue, then only because there are many faulty chips.

I saw that this morning. Smells like a yield issue.

SiliconFly said:
Intel's on a different level when it comes outdoing competition. Wonder what happened to the old Intel. But all is not lost yet. For that we need to wait for (ARL-H) mobile parts.

I agree. A good showing here would be much more important than the flip in the desktop.

511 · Oct 26, 2024

igor_kavinski said:
I really hope that's true. Means they came to their senses finally.

Apparently ARL was pre pat

https://twitter.com/x/status/1850049858373288124

Saylick · Oct 26, 2024

igor_kavinski said:
I really hope that's true. Means they came to their senses finally.

Intel is a slow learner, but I’ll give them credit for learning. Will take them a 5+ year detour to do so, but they’ll eventually get back on the right path.

igor_kavinski · Oct 26, 2024

511 said:
Apparently ARL was pre pat

No. I won't accept the excuse that Pat couldn't do anything with a design that was already in progress. His job was to steer the ship RIGHT.

Arrow Lake and specifically Lion Cove performance tells me:

1) He doesn't care about CPU compute performance which is really weird since he's a CPU architect himself.

2) He's been out of the CPU loop so long (especially due to his previous job at EMC) that he had a talk with the existing CPU architects at Intel, didn't dive technically too deep in those conversations and foolishly trusted them. If that's true, that's on HIM. It is one thing for a fool like BK to run the company into the ground but a senior CPU Architect letting Intel release Arrow Lake with such unbalanced performance is simply inexcusable.

3) The Raptor Lake and Meteor Lake issues speak volumes about his "let the professionals do their work" approach to being a CEO. That approach would be fine for a company that wasn't in trouble but this is not what Intel needed. Intel needed someone like Jensen (yes sad but true) or Lisa Su.

cannedlake240 · Oct 26, 2024

igor_kavinski said:
His job was to steer the ship RIGHT

Not much he can do a year or 2 before product launch. Intel can't keep refreshing raptor lake or not launch a product for 4 years. But overall, yes Intel design core business isn't in a much better state due to constant layoffs and cancellations and they have zero presence in growth segments like AI.

igor_kavinski · Oct 26, 2024

Saylick said:
Intel is a slow learner, but I’ll give them credit for learning.

Unfortunately, that's "toddler sticks finger in electrical socket" type of learning

mpumalanga · Oct 26, 2024

cannedlake240 said:
Not much he can do a year or 2 before product launch. Intel can't keep refreshing raptor lake or not launch a product for 4 years. But overall, yes Intel design core business isn't in a much better state due to constant layoffs and cancellations and they have zero presence in growth segments like AI.

"Not much he can do a year or 2 before product launch"

Gelsinger started his return to Intel beginning of February 2021. That is 3.5 years before arrow lake launch.
Might still be too late to radically change arrow lake, but certainly enough time to influence it.

maddie · Oct 26, 2024

Kocicak said:
It is not. HT threads were the weakest ones and in the spectrum of all possible loads that CPUs can experience in home PCs they got almost never used. Only the heaviest multithreaded applications used those threads and people running such apps would be better off running a threadripper system anyway. HT is useless today, when you have up to 24 core CPUs in home PCs.

This isn't true. Both threads in a HT core shares the resources. There isn't 1 strong and 1 weak one, or at least not continually. The threads are dynamic.

HT adding, say 30% to throughput does not mean 1 thread is at 100% and the other stays at 30%. It could be 65% + 65%, or any mix within the limits.

igor_kavinski · Oct 26, 2024

cannedlake240 said:
Not much he can do a year or 2 before product launch.

Sorry. Disagree.

Things he could've done:

1) Release Bartlett Lake type of CPU on latest refinement of Intel 7 instead of 14900KS (which was useless because it couldn't sustain high clocks without special cooling due to thermal throttling) with AVX-512 enabled.

2) If Meteor Lake was too much in demand (like he said), he should've tried to improve and fix Intel 4 yields before churning those out instead of incurring a huge loss with hot lots.

3) Intel had working Lunar Lake prototypes in October 2023 yet the laptops were ready for sale only a year later. Being a CPU architect, he should've understood the importance of Lunar Lake over Meteor Lake and sped up the development of the former no matter what. He probably didn't do that because MTL was on Intel 4 and he wanted Intel fabs to keep being utilized. That stupid plan backfired spectacularly on him.

4) He probably cut the funding of the GPU team and slowed down ARC's development to focus on other stuff he thought was more profitable. That was a bad move. We could've had Battlemage six months ago had he not messed with that team by firing a lot of them.

5) Gaudi 3? Pfffttt. How many billions did Intel make from THAT?

Pat's been too slow for Intel. Intel needs someone ambitious with a great sense of direction. Not some old has-been who can't figure out what Intel is good at.

511 · Oct 26, 2024

igor_kavinski said:
No. I won't accept the excuse that Pat couldn't do anything with a design that was already in progress. His job was to steer the ship RIGHT.

He influenced Granite Rappids and the Nodes and LNL was on him lol also iirc he also started development of Sierra Forest I don't like him that much but he is a lot better CEOs than 3 people before him

igor_kavinski said:
Arrow Lake and specifically Lion Cove performance tells me:

1) He doesn't care about CPU compute performance which is really weird since he's a CPU architect himself.

2) He's been out of the CPU loop so long (especially due to his previous job at EMC) that he had a talk with the existing CPU architects at Intel, didn't dive technically too deep in those conversations and foolishly trusted them. If that's true, that's on HIM. It is one thing for a fool like BK to run the company into the ground but a senior CPU Architect letting Intel release Arrow Lake with such unbalanced performance is simply inexcusable.

I agree with ARL/LNC BK and Pat are in two different leagues in terms of knowledge about CPUs and to think he was the chief Architect of i486

igor_kavinski said:
3) The Raptor Lake and Meteor Lake issues speak volumes about his "let the professionals do their work" approach to being a CEO. That approach would be fine for a company that wasn't in trouble but this is not what Intel needed. Intel needed someone like Jensen (yes sad but true) or Lisa Su.

Intel needs someone more like Andy Grove than these guys 🙂

gdansk · Oct 26, 2024

igor_kavinski said:
He probably cut the funding of the GPU team and slowed down ARC's development to focus on other stuff he thought was more profitable

Do you know anything about Pat? The only reason there are still plans for discrete GPUs is because he is obsessed with Intel having their own graphics that they can eventually scale into DC. Any other CEO would have killed all those plans by now.

He has said in multiple interviews that not having GPGPU in the 2010s was a huge miss by Intel, even before he was CEO.

igor_kavinski · Oct 26, 2024

maddie said:
HT adding, say 30% to throughput does not mean 1 thread is at 100% and the other stays at 30%. It could be 65% + 65%, or any mix within the limits.

That's true. However, ST performance of Raptor Lake improving a bit when HT is turned off (and quite a few gamers have been doing that since Alder Lake launch) means that some resources get statically partitioned so it's not dynamic HT. Turning on HT starves the primary thread of some resources it could've had if HT hadn't been turned on.

511 · Oct 26, 2024

igor_kavinski said:
Sorry. Disagree.

Things he could've done:

1) Release Bartlett Lake type of CPU on latest refinement of Intel 7 instead of 14900KS (which was useless because it couldn't sustain high clocks without special cooling due to thermal throttling) with AVX-512 enabled.

I can't disagree on this this is idiocity Bartlett lake should be out for consumers as well

igor_kavinski said:
2) If Meteor Lake was too much in demand (like he said), he should've tried to improve and fix Intel 4 yields before churning those out instead of incurring a huge loss with hot lots.

I am pretty sure you don't know much more than Intel on how to run fabs

igor_kavinski said:
3) Intel had working Lunar Lake prototypes in October 2023 yet the laptops were ready for sale only a year later. Being a CPU architect, he should've understood the importance of Lunar Lake over Meteor Lake and sped up the development of the former no matter what. He probably didn't do that because MTL was on Intel 4 and he wanted Intel fabs to keep being utilized. That stupid plan backfired spectacularly on him.

He fixed Intel 7nm(OG) something both the other CEOs were unable to do what is the point of your fabs if they are empty

igor_kavinski said:
4) He probably cut the funding of the GPU team and slowed down ARC's development to focus on other stuff he thought was more profitable. That was a bad move. We could've had Battlemage six months ago had he not messed with that team by firing a lot of them.

Show me the proof for this

igor_kavinski said:
5) Gaudi 3? Pfffttt. How many billions did Intel make from THAT?

Yes Gaudi is the worst bet

igor_kavinski said:
Pat's been too slow for Intel. Intel needs someone ambitious with a great sense of direction. Not some old has-been who can't figure out what Intel is good at.

And who would that be Intel is an IDM seperation of either design or fab will kill the company no one has that experience outside of few people in running IDMs everyone will try AMD/GF Moves

511 · Oct 26, 2024

gdansk said:
Do you know anything about Pat? The only reason there are still plans for discrete GPUs is because he is obsessed with Intel having their own graphics that they can eventually scale into DC. Any other CEO would have killed all those plans by now.

He has said in multiple interviews that not having GPGPU in the 2010s was a huge miss by Intel, even before he was CEO.

That also Not getting the CEO role was the reasons he left if Pat was CEO i am pretty sure Intel would have been on a much better state than rn he kept saying Intel Needed GPUs but MBAs
He need go lay off some upper management

cannedlake240 · Oct 26, 2024

mpumalanga said:
"Not much he can do a year or 2 before product launch"

Gelsinger started his return to Intel beginning of February 2021. That is 3.5 years before arrow lake launch.
Might still be too late to radically change arrow lake, but certainly enough time to influence it.

He can't do performance simulations ahead of time. Arrow lake is tied to Meteor lake and by the time problems in it's design were apparent they weren't going to start designing a whole new soc in between generations just for the Desktop market. Arrow lakes problems aren't caused by chiplets only. There's multiple things at play. Its already delayed and trying to redefine it would result in further delays. Pats main mistake was cancelling PTL-S which might have been a decent upgrade for gaming by early 2026

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

Senior member

Attachments

Golden Member

Senior member

Member

Golden Member

Golden Member

Golden Member

Member

Diamond Member

Senior member

Lifer

Senior member

Golden Member

Diamond Member

Lifer

Senior member

Lifer

Junior Member

Diamond Member

Lifer

Golden Member

Diamond Member

Lifer

Golden Member

Golden Member

Senior member