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 (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.)

Josh128 · 2024-09-27T13:20:45-0400

9949asd said:
Because 370 is using almost double the power.

Come again? The tests were at 28W power limit for all laptops.

9949asd · 2024-09-27T13:32:35-0400

Josh128 said:
Come again? The tests were at 28W power limit for all laptops.

View attachment 108331

You should watch the full video.
What’s wrong with fanboys?

TESKATLIPOKA · 2024-09-27T13:37:34-0400

9949asd said:
It’s kind of life style 😎, you don’t want to carry a brick of charger hanging around in a cafe or a classroom in university.

Not just being lazy?

Hitman928 · 2024-09-27T13:43:34-0400

9949asd said:
You should watch the full video.
What’s wrong with fanboys?
View attachment 108332 View attachment 108333 View attachment 108334

If you watch the full video as you suggest, you’ll see that he set all three processors to a 28W power limit. The graphs you posted are for full system power with different systems and different screens, as has been pointed out multiple times now.

9949asd · 2024-09-27T13:44:37-0400

Josh128 said:
Come again? The tests were at 28W power limit for all laptops.

View attachment 108331

Can you understand What he’s saying?

Henry swagger · 2024-09-27T13:44:40-0400

Lunar lake and new xeons will help intel consolidate its 80% market share

TESKATLIPOKA · 2024-09-27T13:47:37-0400

adroc_thurston said:
No lmao.

Ryzen AI 7/9 says Hi.

TESKATLIPOKA · 2024-09-27T13:53:18-0400

Josh128 said:
This, this, this. I dont know how many times it needs to be said. Screen size and type are much more significant in regards to battery life and system power draw when you get to these <28W CPUs. But people rather stick to hyperbole without context for their favorite brand.

An what's the difference in W?
I also did a comparison LNL vs Strix, both Zenbooks, even the screens the same from the same manufacturer, the only difference was 2 inches.
I know in some cases 2 inches make a big difference, but are yo saying this is also such a case?

9949asd · 2024-09-27T13:53:34-0400

Hitman928 said:
If you watch the full video as you suggest, you’ll see that he set all three processors to a 28W power limit. The graphs you posted are for full system power with different systems and different screens, as has been pointed out multiple times now.

The actual power draw always different to the power limit.

cebri1 · 2024-09-27T14:00:37-0400

9949asd said:
Can you understand What he’s saying?View attachment 108335 View attachment 108336

The 258V is clearly way more efficient. People just can get their head around to accept it.

TESKATLIPOKA · 2024-09-27T14:01:06-0400

9949asd said:
The actual problem draw always different to the power limit.

If power limit is set to 28W, then the CPU(APU, SoC) simply won't go over It.
Unless this measurement is when turbo is still active or he is lying about the power limit.

Hitman928 · 2024-09-27T14:01:34-0400

9949asd said:
The actual problem draw always different to the power limit.

Not true. Additionally, it’s not going to go over the power limit but being at only 28W, the SoCs should be really close in actual power consumption, even if they don’t quite hit the limit. There’s no way the MTL and STX SoCs are drawing 20 extra watts over LNL when they are limited to 28W power.

Hitman928 · 2024-09-27T14:03:11-0400

TESKATLIPOKA said:
If power limit is set to 28W, then the CPU(APU, SoC) simply won't go over It.
Unless this measurement is when turbo is still active or he is lying about the power limit.

He set the max limit to 28W, so turbo power doesn’t play a role.

TESKATLIPOKA · 2024-09-27T14:04:10-0400

Hitman928 said:
He set the max limit to 28W, so turbo power doesn’t play a role.

Then I don't understand the difference in power consumption. Was that some spike or what?

Hitman928 · 2024-09-27T14:15:37-0400

TESKATLIPOKA said:
Then I don't understand the difference in power consumption. Was that some spike or what?

LNL systems are built for power efficiency, not just the processor. It also has a smaller screen, lower refresh screen. There could be some power consumption difference in gaming loads between the SoCs, but with them limited to 28W, it’s not going to be that big and no where near the 20W shown for the full system power difference.

Khato · 2024-09-27T14:15:42-0400

TESKATLIPOKA said:
Then I don't understand the difference in power consumption. Was that some spike or what?

One reason for the difference is that LNL includes DRAM in the 28W power limit. Many like to simply brush that detail aside, but even if you go with a low figure of 1.5W for DRAM that still results in 26.5W for LNL and 29.5W for MTL/STX.

511 · 2024-09-27T14:15:46-0400

I wonder what did he set exactly for 28W as setting P1<PL2 will result in slightly higher average power draw but in AMD it's usually higher than PL they have some like 1.35X on top of tdp PPT i pressume what it's called so maybe he forgot to set PPT to 28 ?

511 · 2024-09-27T14:16:44-0400

Hitman928 said:
LNL systems are built for power efficiency, not just the processor. It also has a smaller screen, lower refresh screen. There could be some power consumption difference in gaming loads between the SoCs, but with them limited to 28W, it’s not going to be that big and no where near the 20W shown for the full system power difference.

Yes it should be like 5-7W extra at max not 20W

rainy · 2024-09-27T14:32:12-0400

SiliconFly said:
Turin is based on Zen5. People generally tend to over inflate expectations, but it looks like Turing is not gonna get that much of an performance uplift this gen. It's Zen 5% after all. GNR is not under any major threat anymore.

What Zen 5 IPC advantage? Zen 5% advantage??

Sadly, no more "massive lead" for Zen products this gen due to disappointing Zen 5 perf uplift.

You definitely shouldn't translate gaming performance of Granite Ridge to server/data center type of tasks because they are completely different.
Phoronix have tested 9950x/9900x in such scenario and first was circa 18 percent faster than 7950x and latter 21.5 percent than 7900x.
https://www.phoronix.com/review/amd-ryzen-9950x-9900x

Btw, I expect that Turin/Turin Dense will perform even better because those CPUs are optimized for that kind of workload.

9949asd · 2024-09-27T14:32:34-0400

511 said:
I wonder what did he set exactly for 28W as setting P1<PL2 will result in slightly higher average power draw but in AMD it's usually higher than PL they have some like 1.35X on top of tdp PPT i pressume what it's called so maybe he forgot to set PPT to 28 ?

For sure the 28w power limit was not working.

Hitman928 · 2024-09-27T15:00:37-0400

511 said:
I wonder what did he set exactly for 28W as setting P1<PL2 will result in slightly higher average power draw but in AMD it's usually higher than PL they have some like 1.35X on top of tdp PPT i pressume what it's called so maybe he forgot to set PPT to 28 ?

PL1=PL2=28W or PL=PPT=28W.

coercitiv · 2024-09-27T15:06:46-0400

Khato said:
One reason for the difference is that LNL includes DRAM in the 28W power limit. Many like to simply brush that detail aside, but even if you go with a low figure of 1.5W for DRAM that still results in 26.5W for LNL and 29.5W for MTL/STX.

Including the DRAM on package is a design choice, it also brings important advantages not just another contributor to package power. You get slightly better memory tuning, you get considerably lower mem controller voltage and Intel also gets to ensure all chips have LPDDR5X 8533. The price they pay by lowering max available package power is arguably more than made up by higher memory performance in combination with around 100mV less voltage needed to drive the memory chips.

Nobody brings up the advantage of having mem on package when comparing battery life, because it would be absurd to penalize LNL for an advantageous design choice. The same applies when scaling power though, it would be absurd to penalize Strix, MTL or even the upcoming ARL-U for their design choices.

SiliconFly · 2024-09-27T15:12:54-0400

Josh128 said:
Come again? The tests were at 28W power limit for all laptops.

View attachment 108331

9949asd said:
You should watch the full video.
What’s wrong with fanboys?

9949asd said:
Can you understand What he’s saying?View attachment 108336

Hilarious! HX 370 sucks! Looks like Lunar Lake is king of performance-per-watt in gaming too!

Henry swagger said:
Lunar lake and new xeons will help intel consolidate its 80% market share

Sweet. At this rate, they're gonna start reclaiming their lost market share as well!

TESKATLIPOKA said:
I know in some cases 2 inches make a big difference, but are yo saying this is also such a case?

We're talking abt displays, right?

cebri1 said:
The 258V is clearly way more efficient. People just can get their head around to accept it.

Never surrender! The fun should never stop.

Khato said:
One reason for the difference is that LNL includes DRAM in the 28W power limit. Many like to simply brush that detail aside, but even if you go with a low figure of 1.5W for DRAM that still results in 26.5W for LNL and 29.5W for MTL/STX.

Wow. I didn't take that into account.

coercitiv said:
Including the DRAM on package is a design choice, it also brings important advantages not just another contributor to package power. You get slightly better memory tuning, you get considerably lower mem controller voltage and Intel also gets to ensure all chips have LPDDR5X 8533. The price they pay by lowering max available package power is arguably more than made up by higher memory performance in combination with around 100mV less voltage needed to drive the memory chips.

Nobody brings up the advantage of having mem on package when comparing battery life, because it would be absurd to penalize LNL for an advantageous design choice. The same applies when scaling power though, it would be absurd to penalize Strix, MTL or even the upcoming ARL-U for their design choices.

Awsm!

H433x0n · 2024-09-27T15:21:55-0400

TESKATLIPOKA said:
Please tell me why is It a problem to put a small charger in your bag or connect your laptop to a charger If needed after 8 hours instead of 10?
You can easily stress LNL so It won't last 2 hours, so you will need that charger.
So what are we even talking about?
It really looks like people love to go to extremes, just to make their chosen product the best.

I will say It directly, Strix is a good product, LNL is also a good product. You won't really go wrong with either of them.

Are we unironically arguing that battery life doesn’t matter? If we’re using this argument, isn’t RPL a superior mobile product over Phoenix & Hawk Point? It had more cores, better multi threading and a slight edge in single threading.

9949asd · 2024-09-27T15:24:22-0400

SiliconFly said:
Hilarious! HX 370 sucks! Looks like Lunar Lake is king of performance-per-watt in gaming too!

Sweet. At this rate, they're gonna start reclaiming their lost market share as well!

We're talking abt displays, right?

hx370 should not compare with lunar lake, the lunar lake direct competitor is M3 MacBook Air.

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

Senior member

Attachments

Senior member

Member

Platinum Member

Diamond Member

Member

Senior member

Platinum Member

Platinum Member

Member

Senior member

Platinum Member

Diamond Member

Diamond Member

Platinum Member

Diamond Member

Golden Member

Senior member

Senior member

Senior member

Member

Diamond Member

Diamond Member

Golden Member

Golden Member

Member