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)	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 ?	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.)

 

[mikk]

Bad sign. 14th gen (RPL-Refresh and MTL)seems to be a longer peroid product than previous speculation, it include RPL-S, HX, and MTL which are low power SKUs only. This would be another 'Coffeelake+Cannonlake' situation until full-lineup Arrowlake arrive.

https://twitter.com/x/status/1645654155028221952

Cannonlake didn't exist, the better comparison would be Icelake-U which also was low volume (wasn't really low volume later on) and Comet-U, they co-existed. Icelake-U launched 2 years before ADL-S. At the moment it looks like Meteor-M/P is launching only 1 year before MTL-S/ARL-S.


Something about Lunar:

https://twitter.com/x/status/1646049654952321024

Arrow Lake is the successor of Lunar Lake, so we should see it before Arrow Lake (P?). Not sure it they refer explicitly to Arrow Lake-P because Lunar is mobile only or Arrow Lake in general including Arrow Lake-S. That's why they talk more about Lunar Lake. They are pushing Lunar Lake.

 

[eek2121]

TSMC 3nm is what rumors have been saying.
Intel 20A would be working at the same time as Intel 3 would be working (For GNR/SRF) but I get your point.
Who knows maybe it will be Intel 3 : )

Nothing has changed as far as Intel’s plans. TSMC is not slated for use with Intel’s core client IP at any point in the near future. Unless you have evidence otherwise, please do your part to stop spreading this nonsense.

Intel will be using TSMC for the GPU and other “less important stuff” just like they do today, this is unchanged from previously as well.

Intel is currently on track to deliver chips and processes on schedule.

 

[A///]

Arrow Lake is the successor of Lunar Lake, so we should see it before Arrow Lake (P?). Not sure it they refer explicitly to Arrow Lake-P because Lunar is mobile only or Arrow Lake in general including Arrow Lake-S. That's why they talk more about Lunar Lake. They are pushing Lunar Lake.

hmm? I'm certain lunar comes after arrow and for mobile only as you pont out. after arrow is panther lake wich is for both desktop and mobile if not mistaken.

 

[A///]

Nothing has changed as far as Intel’s plans. TSMC is not slated for use with Intel’s core client IP at any point in the near future. Unless you have evidence otherwise, please do your part to stop spreading this nonsense.

Intel will be using TSMC for the GPU and other “less important stuff” just like they do today, this is unchanged from previously as well.

Intel is currently on track to deliver chips and processes on schedule.

I think mr. gedda is confusing the GPU and CPU parts with each other based on a stupid rumor peddled around by the usual sort who peddle bs.

 

[mikk]

Nothing has changed as far as Intel’s plans. TSMC is not slated for use with Intel’s core client IP at any point in the near future.

You should prove it. Can you? Intel only told they are using 20A and TSMC 3nm for Arrow Lake, this is all we have. They didn't exactly specified CPU or GPU process nodes. People assumed 20A will be CPU tile and TSMC 3nm will be GPU tile but that was was never confirmed.

We even got a leaked roadmap from Arrow Lake-P some time ago where TSMC 3nm was the node used for Lion Cove and Skymont, it isn't far fetched to assume TSMC will be used for Arrow Lake (S) and/or Lunar Lake. The old reddit leak claimed TSMC 3nm for Lunar Lake.

 

[A///]

you lot do know you can search through the thread right?

 

[FangBLade]

As far as I understand, MTL iGPU will have a whole chiplet to itself? This should be a big boost in performance, but the chip area alone doesn't mean much, we know how Arc dGPU scales, their chip area is already larger than Nvidia's and AMD's competitors, yet they have less performance.

 

[A///]

As far as I understand, MTL iGPU will have a whole chiplet to itself? This should be a big boost in performance, but the chip area alone doesn't mean much, we know how Arc dGPU scales, their chip area is already larger than Nvidia's and AMD's competitors, yet they have less performance.

Intel architecture day presentations will lay that out, i believe. if not it's been discussed for a very long time.

 

[Geddagod]

what intel is doing?

No, the idea that Intel can't get AVX-512 working on E-cores so they switch over to multiple compute tiles. I don't think it makes any sense.
AVX-512 isn't exactly the next big step of CPU evolution, both AMD and Intel are able to successful implement AVX-512 in their processors now. There's nothing really inherent to the little cores that make it hard for them to implement AVX-512. Maybe their area/performance goals might not get hit due to them making the core getting bigger to implement it, but that's the worst case scenario I see.
IMO Intel is going to try to share FPU clusters like Bulldozer to save space on the E-cores, but who knows.
Plus getting AVX-512 not working is such a high level architectural design goal that if they can't get it to work, it sounds like they could get it caught very early on the core design process and tweak the design to get it fixed.

 

[Geddagod]

hmm? I'm certain lunar comes after arrow and for mobile only as you pont out. after arrow is panther lake wich is for both desktop and mobile if not mistaken.

I think there's a decent chance LNL and ARL mobile release at the same time. Extremely late 2024 or early 2025.
Plus since they are apparently going to use the same architecture, it makes sense to class them into the same generation.

 

[A///]

No, the idea that Intel can't get AVX-512 working on E-cores so they switch over to multiple compute tiles. I don't think it makes any sense.

?????????????????

reread what I said. Intel disabled avx512 initially then began fusing them off during production because the e cores wouldn't work with it. What I said is that intel may in the future drop the e cores and go to p cores or normal cores and going tile based their yields will improve while riskfactor will reduce. Make sense now?

 

[Geddagod]

?????????????????

reread what I said. Intel disabled avx512 initially then began fusing them off during production because the e cores wouldn't work with it. What I said is that intel may in the future drop the e cores and go to p cores or normal cores and going tile based their yields will improve while riskfactor will reduce. Make sense now?

E-cores wouldn't work with it because they didn't support AVX-512, but future E-cores might change that if the cores themselves are able to support AVX-512.
I doubt Intel drops E-cores anytime soon before their next architectural rehaul. Why do you think they will?
Also which generation do you suspect multiple compute chiplets will appear? Doesn't look like ARL or MTL will feature them.

 

[Exist50]

Nothing has changed as far as Intel’s plans. TSMC is not slated for use with Intel’s core client IP at any point in the near future. Unless you have evidence otherwise, please do your part to stop spreading this nonsense.

Current rumors heavily suggest N3 for Lunar Lake and some Arrow Lake SKUs. If nothing else, nothing Intel's said has eliminated that possibility.

hmm? I'm certain lunar comes after arrow and for mobile only as you pont out. after arrow is panther lake wich is for both desktop and mobile if not mistaken.

LNL and ARL are more likely to be contemporaries than anything else. And yes, Panther Lake is the rumored name of the gen after both.

here's nothing really inherent to the little cores that make it hard for them to implement AVX-512.

128b datapath certainly complicates things a lot.

 

[A///]

E-cores wouldn't work with it because they didn't support AVX-512, but future E-cores might change that if the cores themselves are able to support AVX-512.
I doubt Intel drops E-cores anytime soon before their next architectural rehaul. Why do you think they will?
Also which generation do you suspect multiple compute chiplets will appear? Doesn't look like ARL or MTL will feature them.

As long as both p and e cores utilise separate uarchs it won't work under the current design mindset. The cost for intel to find a solution to get it to work withstanding from now is greater than being able to pack standard cores while reducing or using the same power. E cores are largely a weight around intel's neck from software that can't properly use them to their fullest to the windows scheduler. the only other easy method is if they copy amd's future attempt. One can sing praise about intel's e cores but at the end of the day regardless of how powerful they are relative to the e cores they're still a headache.
It's going to be a while before multiple compute chiplets become common for intel. I want to say Intel has learned from their lessons and is going slowly. I don't expect Panther to be anything unique but around the time of post panther release looking towards the next gen we may see whispers of a new intel solution. This all hinges on amd continuing to deliver. as long as they make intel run a few km for their morning jelly donut the more advancements we'll see.
A very old rumor was panther lake is a radical new design but i am not buying it. not much is said about what comes after because the line gets fuzzy between it and the so called nova lake.

Intel

 

[Geddagod]

As long as both p and e cores utilise separate uarchs it won't work under the current design mindset. The cost for intel to find a solution to get it to work withstanding from now is greater than being able to pack standard cores while reducing or using the same power. E cores are largely a weight around intel's neck from software that can't properly use them to their fullest to the windows scheduler. the only other easy method is if they copy amd's future attempt. One can sing praise about intel's e cores but at the end of the day regardless of how powerful they are relative to the e cores they're still a headache.
It's going to be a while before multiple compute chiplets become common for intel. I want to say Intel has learned from their lessons and is going slowly. I don't expect Panther to be anything unique but around the time of post panther release looking towards the next gen we may see whispers of a new intel solution. This all hinges on amd continuing to deliver. as long as they make intel run a few km for their morning jelly donut the more advancements we'll see.
A very old rumor was panther lake is a radical new design but i am not buying it. not much is said about what comes after because the line gets fuzzy between it and the so called nova lake.

Intel

Just because they have different architectures doesn't mean they aren't compatible. As long as they have the same level of support, there should be no issues. From what I have seen, literarily everything except programs that use AVX-512 is fully compatible with the P-cores, and that's only because the E-cores don't support AVX-512 (for this generation, nothing technically really is stopping them from adding AVX-512 to future generations of E-cores).

 

[A///]

LNL and ARL are more likely to be contemporaries than anything else. And yes, Panther Lake is the rumored name of the gen after both.

Going back to something Pat "Guns A'Blazin Gunslinger" Gelsinger said a long while back was espousing good things about Meteor, but as I see it Arrow Lake will be the refined child for them.

Panther should be interesting! IDK if you saw the memo of ARM and Intel today but they hinted at that their fabs are doing well at least my take on it from a 2c view.

 

[A///]

Just because they have different architectures doesn't mean they aren't compatible. As long as they have the same level of support, there should be no issues. From what I have seen, literarily everything except programs that use AVX-512 is fully compatible with the P-cores, and that's only because the E-cores don't support AVX-512 (for this generation, nothing technically really is stopping them from adding AVX-512 to future generations of E-cores).

You'll have to address the engineers @ intel with your knowledge because they couldn't get it to work and ended up disabling it then physically fusing processors. Clearly you know far more than them.

 

[Geddagod]

You'll have to address the engineers @ intel with your knowledge because they couldn't get it to work and ended up disabling it then physically fusing processors. Clearly you know far more than them.

I don't know if you read the stuff in parenthesis, but I said for future generations
Idk how many times I stated that this generation of E-cores don't support AVX-512, but there's nothing stopping future generations of E-cores from adding AVX-512. It's not an inherent challenge to the design of E-cores, adding AVX-512 adds space yes, but there are design techniques that can reduce space in other places as well, or even just sharing FPU units, etc etc.
Do you really think Intel can't change the design of E-cores in the future?

 

[igor_kavinski]

It's not an inherent challenge to the design of E-cores, adding AVX-512 adds space yes, but there are design techniques that can reduce space in other places as well, or even just sharing FPU units, etc etc.

Seems plausible. Knowing Intel, they may put a single AVX-512 unit in the E-core cluster so they can unlock the dormant AVX-512 units in their P-cores. Two birds with one stone. If they hadn't rushed ADL/RPL due to the AMD threat, we would have gotten that design on day one of ADL. Granite Rapids is rumored to feature AVX-1024 so maybe Arrow Lake will be the design done right without needing to disable units or wasting die space on stuff no one can use.

 

[A///]

I don't know if you read the stuff in parenthesis, but I said for future generations
Idk how many times I stated that this generation of E-cores don't support AVX-512, but there's nothing stopping future generations of E-cores from adding AVX-512. It's not an inherent challenge to the design of E-cores, adding AVX-512 adds space yes, but there are design techniques that can reduce space in other places as well, or even just sharing FPU units, etc etc.
Do you really think Intel can't change the design of E-cores in the future?

It's not that I don't think they can it's that I don't see the cost vs benefit basis of it. Crestmont, which comes after Gracemont, is not expected to have AVX512 either. For Intel the time and energy spent implementing AVX512 on both p and e cores and making sure they communicate with the OS without hiccup under Linux or Windows is a large money and time glut when they can not include AVX512. If I had to pull a number out of my rear I'd say less than 2% of general consumers benefit from AVX512. My understanding is the benefit general consumers will see from AVX512 is in retro game emulation which can't be discussed at length due to it being gray area piracy.

 

[igor_kavinski]

My understanding is the benefit general consumers will see from AVX512 is in retro game emulation which can't be discussed at length due to it being gray area piracy.

I hope Microsoft Excel or LibreCalc is able to spice things up with AVX-512. Suddenly, offices everywhere start ordering Zen 4 PCs. That would get Intel really worried

 

[A///]

Seems plausible. Knowing Intel, they may put a single AVX-512 unit in the E-core cluster so they can unlock the dormant AVX-512 units in their P-cores. Two birds with one stone. If they hadn't rushed ADL/RPL due to the AMD threat, we would have gotten that design on day one of ADL. Granite Rapids is rumored to feature AVX-1024 so maybe Arrow Lake will be the design done right without needing to disable units or wasting die space on stuff no one can use.

neither Alder nor Raptor were rushed. Intel didn't feel a threat by amd, recall amd's sales had been poor for months and only am4 was selling better. intel had repeatedly tried to get avx512 working on both platforms but kept running into performance problems.to give you some insight here, alder lake was not intel's first heterogeneous cpu design. lakefield was, and lakefield also had its avx512 fused off because of limitations imposed by a hybrid architecture. kakefield was Intel's first modern foray into hybrid designs but alder was their first real swim in it with mass adoption. kakefield began production in 2019 and I can't tell you when development began, but the issue of Intel not being able to get avx512 running on a hybrid design is an old problem. From the rumors Intel's main competitor looks to use the same core in a future generation but lacking cache for their alleged hybrid design. this should allow them to omplement avx512 without too much trouble at least on paper.

what about power you say? have you seen current gen? intel's competitor smokes them off the line in lower power consumption thruout the load range.

 

[igor_kavinski]

neither Alder nor Raptor were rushed.

I would believe that, except Intel launched Alder Lake with AVX-512 enabling option in BIOS. Then backtracked due to some order from a high ranking executive (no idea who). Dr. Ian reported that the engineers who had worked on AVX-512 on ADL were taken aback and pissed when they learned that AVX-512 was getting disabled post launch. It makes no sense to take AVX-512 away, just to make your puny cores work. It feels like they forced their decision down everyone's throats and took an option away from enthusiasts while also making their CPU look lacking compared to Zen 4.

 

[A///]

I hope Microsoft Excel or LibreCalc is able to spice things up with AVX-512. Suddenly, offices everywhere start ordering Zen 4 PCs. That would get Intel really worried

I had my taxes done 2 weeks ago and I was chatting with my accountant and the discussion fell to the new excel and how it used "ai" instruction to automate large tasks and speed work flow up. Idk what generation processor his laptop has but it wasn't new. i myself saw what used to take a very long time to do manually even with software aids be finished in minutes. intel and their main competitor are planning on incorporating more ai instruction sets whatever that means and accelerators into future hardware, like z6 and beyond and maybe panther lake for intel. i couldn't tell you how much benefit slapping on a neural engine to a modern x86 processor that runs at significantly higher frequencies than non x86 competitors will give end users. it would speed up work for certain people such as creatives whose livelyhood hinges on being able to churn out more and more junk to upload onto youtube.

 

[igor_kavinski]

I had my taxes done 2 weeks ago and I was chatting with my accountant and the discussion fell to the new excel and how it used "ai" instruction to automate large tasks and speed work flow up. Idk what generation processor his laptop has but it wasn't new. i myself saw what used to take a very long time to do manually even with software aids be finished in minutes.

Very interesting. What kind of calculation was that? Would you be able to share a sample sheet without any confidential data?