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.)

 

[majord]

ST is around expected I think.. MT score is bad , it would want to be a bad run or something for their sake.

Nothing else really stands out.. Probably most interesting thing now will be IGP performance. Since that's the other thing they promised it would excel in

 

[poke01]

Right but Intel’s own slide is saying they get 1.165x the ST score of the HX 370, which has a score ranging from 112-116. That comes out to 130-135. The scores also match somewhere within the 1.0526x the X80-X84: 124-128.
I don’t think it’ll be that high (135) but they’ve put it out there that it could be. So a 127 is plenty reasonable for the 268V.

Spotted this. Not sure what to think of it, but kinda jives with what I was pointing out (could just be throttling, but that still seems too low).

https://twitter.com/x/status/1832999226751467745

Does this XPS have a fan?

 

[cebri1]

From the same video


I wouldn’t draw too many conclusions.

 

[poke01]

From the same video
View attachment 107126

I wouldn’t draw too many conclusions.

Probably running pre-release firmware

 

[jdubs03]

Does this XPS have a fan?

It does as far as I know.

 

[naukkis]

ST is around expected I think.. MT score is bad , it would want to be a bad run or something for their sake.

No it's just what to be expected from max 17w soc. Those 700-800 results would have only possible without powerlimits consuming about 100w.

 

[jdubs03]

No it's just what to be expected from max 17w soc. Those 700-800 results would have only possible without powerlimits consuming about 100w.

It can go up to 30W though. It’s obviously going to be less than the HX 370 in MT, because of core count. But at 28W if the HX 370 can get 919, I don’t see why the 268V can’t get in the high 600s.

From https://www.computerbase.de/2024-07/amd-strix-point-zenbook-s-16-test/

 

[Abwx]

It can go up to 30W though. It’s obviously going to be less than the HX 370 in MT, because of core count. But at 28W if the HX 370 can get 919, I don’t see why the 268V can’t get in the high 600s.
View attachment 107127

All HX 370 12 cores have SMT, that s a big difference.
50% more cores + 30% SMT amount to 2 x the core count, so it can score roughly 70% better at a same 33W.

Beside the 8945HS@37W score 852 pts, so Hawk Point should have no trouble scoring 600 pts at 17W while the 286V require 30W to do so, if not with a turbo at 37W.

 

[naukkis]

It can go up to 30W though. It’s obviously going to be less than the HX 370 in MT, because of core count. But at 28W if the HX 370 can get 919, I don’t see why the 268V can’t get in the high 600s.
View attachment 107127
From https://www.computerbase.de/2024-07/amd-strix-point-zenbook-s-16-test/

No it can't max tdp for 268v is 17w.

 

[jdubs03]

All HX 370 12 cores have SMT, that s a big difference.
50% more cores + 30% SMT amount to almost 2 x the core count, so it can score roughly 70% better at a same 33W.

Beside the 8945HS@37W score 852 pts, so Hawk Point should have no trouble scoring 600 pts at 17W while the 286V require 30W to do so, if not with a turbo at 37W.

You’re being very generous about that 30%. It’s obviously workload dependent but I’ve seen half that number quoted as well. But that’s also the whole purpose of Skymont. It is to negate the loss of 15-30% performance. Hence why you’ll see the 285K still do quite well in MT tasks.

No it can't max tdp for 268v is 17w.

See timestamp 05:26

 

[511]

We need to manually set Power limit rn it is driven by Dynamic Tuning driver we want the values to be static to properly examine the performance

 

[naukkis]

View attachment 107128

See timestamp 05:26

See yourself. Even 17w is a lot for proper thin&light laptop and configuratoon limits for LL is 8-17w. 288v is outlier for non thin & light use cases.

 

[jdubs03]

See yourself. Even 17w is a lot for proper thin&light laptop and configuratoon limits for LL is 8-17w. 288v is outlier for non thin & light use cases.

What do you mean see for yourself? I watched it that’s why I attached it. You said the 268V can’t go up to 30W. I said it can. The video says that it can.

 

[naukkis]

What do you mean see for yourself? I watched it that’s why I attached it. You said the 268V can’t go up to 30W. I said it can. The video says that it can.

Video is mixing ultra9 to 5 and 7. U9 can go up to 30w, others up to 17w. See that table you posted.

 

[jdubs03]

Video is mixing ultra9 to 5 and 7. U9 can go up tp 30w, other up to 17w.

The guy literally says it’s a Core Ultra 7 running at 17W and can go up to 30W. Nothing is mixed up.

Here’s more evidence:

Intel® Core™ Ultra 7 Processor 268V (12M Cache, up to 5.00 GHz) - Product Specifications | Intel

Intel® Core™ Ultra 7 Processor 268V (12M Cache, up to 5.00 GHz) quick reference with specifications, features, and technologies.

www.intel.com

That says max turbo power is 37W!

 

[naukkis]

The guy literally says it’s a Core Ultra 7 running at 17W and can go up to 30W. Nothing is mixed up.

Meaning can boost short periods to 30w.

 

[jdubs03]

Meaning can boost short periods to 30w.

Yes. But that’s not what you were saying. You were saying 17W was a hard limit. Despite me showing it wasn’t numerous times.

 

[naukkis]

Yes. But that’s not what you were saying. You were saying 17W was a hard limit. Despite me showing it wasn’t numerous times.

For cinebench mt result only tdp matters. You expected something other than 17w tdp mt results and were disapointed. 17w tdp is limiting those results.

 

[cebri1]

Power envelopes are listed in the official specs

 

[jdubs03]

For cinebench mt result only tdp matters. You expected something other than 17w tdp mt results and were disapointed. 17w tdp is limiting those results.

I get that the TDP is 17W. But that’s not the point of contention here. There are likely some points in those tests that there are spikes to 30W if configured for it. My whole argument was that it can hit 30W that’s it. I don’t see why we’ve had to go back and forth like 4 times about that.

To me yes 479 is a very low score considering people are framing LNL as an M3 class chip. An M3 does 650 in MT. Perhaps because of throttling, and the sustained 17W what I was saying about scoring high-600s is too much. I’ll admit that.

 

[naukkis]

I get that the TDP is 17W. But that’s not the point of contention here. There are likely some points in those tests that there are spikes to 30W if configured for it. My whole argument was that it can hit 30W that’s it. I don’t see why we’ve had to go back and forth like 4 times about that.

To me yes 479 is a very low score considering people are framing LNL as an M3 class chip. An M3 does 650 in MT. Perhaps because of throttling, and the sustained 17W what I was saying about scoring high-600s is too much. I’ll admit that.

People are expecting way too much. That 17w tdp is probably also lowering st score so that explains why results are bit lower than just scaling 288v down. Apple cpu cores are still miles away for efficiency.

 

[jdubs03]

People are expecting way too much. That 17w tdp is probably also lowering st score so that explains why results are bit lower than just scaling 288v down. Apple cpu cores are still miles away for efficiency.

No argument from me on that one!

 

[poke01]

Are we expecting 288V to get close to M3 in MT?
It uses 30 watts as well in Cinebench R23. Intel's Skymont E-cores are more powerful than Apple's so it might.

 

[cebri1]

M4 is going to massacre competition. Apple vs x86 right now is a settled debate.

 

[majord]

People are expecting way too much. That 17w tdp is probably also lowering st score so that explains why results are bit lower than just scaling 288v down. Apple cpu cores are still miles away for efficienE

Expecting too much in reference to what?