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	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.)

TESKATLIPOKA · Dec 23, 2023

SiliconFly said:
Still trying to measure power efficiency with only the P core?

Still trying to measure power efficiency with only the LP E-cores?

SiliconFly said:
On the contrary, majority of the so called "power efficiency" test results that we've come across so far has squarely focused on stressing the CPU tile alone while COMPLETELY ignoring the LPE cores, which I believe it completely wrong as it doesn't align with MTL's use cases.

MTL is a general purpose CPU, so you can use It for many things, It's not limited to only light workload.
BTW, I didn't see any confirmation that those LP e-cores are not working during full load.

SiliconFly said:
Just give MTL a chance to prove itself in lighter load tests with newer pcode. Thats all I'm sayin.

And you expect what? Or rather what would It really change?
It will either win or lose against PHX in idle, video, web, but is that really that important?
Does It really matter that Intel will maybe consume 1-2watts less than AMD in those cases?
The only thing It will do is extend your battery life, but how many really need more than 8 hours of battery life, or rather how many wouldn't have electrical outlet to charge their laptop?
If It was about electricity bill, then consuming under load should be more important I think.

H433x0n · Dec 23, 2023

TESKATLIPOKA said:
Nothing new there, It was already mentioned here.
And you even call It as a huge efficiency uplift.
It's not that big, only 7% higher sustained score in CB R23. And you call It as huge.

Relative to expectations set last week it’s huge lol

Tigerick · Dec 23, 2023

	ZenBook 14 Retail	MechRevo 14 Pro 2024 Retail	Lenovo Legion Pro 16 Retail	RedmiBook 16 Retail	MechRevo 14+ Retail	Lenovo Air 14 Retail
	ICU-7 155H	ICU-5 125H	ICU-5 125H	i5-13500H	R7 7840HS	R7 7840U
	6P + 8E	4P + 8E	4P + 8E	4P + 8E	8P	8P
	24 MB	18 MB	18 MB	18 MB	16 MB	16 MB
	32G 7467	32G 5600	32G 7467	16G 6400	16G 6400	16G 6400
	128EU	112 EU	112EU	80EU	768SP	768SP
	1 Fan	2 Fans	2 Fans	2 Fans	2 Fans	2 Fans
	75 WHr	60 WHr	84 WHr	72 WHr	60 WHr	65 WHr
AC Adaptor	65 W	100 W	100 W	100 W	100 W	65 W

BIOS Tested	203	T140MTLTFV1615 Date: 12/23/2023	MECN42WW	0303	2.17	M4CN31WW
TDP	28 W	28 W	28 W	45 W	35 W	28 W
PL1 (Long)	64 W	60 W	85 W	70 W	54 W	30 W
PL2 (Short)	64 W	75 W	115 W	115 W	NA	NA
Base	1.6/1.4 GHz	1.4/1.2 GHz	1.4/1.2 GHz	1.9 GHz	3.8 GHz	3.3 GHz
Turbo	4.8 GHz	4.5 GHz	4.5 GHz	4.7 GHz	5.1 GHz	5.1 GHz
CPU Burn-In
Clock Speed	1.9 / 2.3 / 1.3 GHz	2.8 / 4.0 / 2.2 GHz	2.8 / 4.3 / 2.5 GHz	3.5 / 2.8 GHz	4.325 GHz	3.525 GHz
CPU Power	28 W	59.89 W	68.75 W	46 W	53.55 W	36 W

CineBench R23
10 mins	11306	13745	14567	12044	16058	14825
Single T	1729	1636	1641	1739	1788	1689
AIDA Memory Latency	143.9 ns	145.2 ns	133.7 ns	89.7 ns	105.0 ns	126.7 ns
3DMark TimeSpy	3442	3261	3564	1545	2928	2901
Assassin's Creed FHD LQ	45	43	46	NA	58	50
Tomb Raider FHD LQ	59	56	63	33	66	59
Video Editing CPU Only	164 sec	143 sec	132 sec	159 sec	123 sec	135 sec
Video Editing CPU + GPU	48 sec	35 sec	34 sec	NA	NA	NA

I have compiled numbers from Chinese Youtuber 中正评测. He bought the retail version of ZenBook 14 and Lenovo's Legion 16 Pro; both videos are available at Youtube. For comparison, I have listed AMD 7840HS's numbers as well...

From the numbers we can conclude Meteor Lake is not as power efficient as Intel want you to believe. You need 115W PL2 to really hit the performance level approaching AMD's 780M. So much for Centrino and Intel 4 process node, if anyone still believe in IFS's skill, wait until you see CPU from Intel 20A/18A...

That's explain why Intel delays the launching of Ultra 9 cause Ultra 9 comes with 45W TDP and 2.5/2.3GHz base clock....

Khato · Dec 23, 2023

Tigerick said:
I have compiled numbers from UltraBook Review and Chinese Youtuber 中正评测. He bought the retail version of both ZenBook 14 and Lenovo 16 Pro; both videos are available at Youtube. For comparison, I have listed AMD 7840HS's numbers as well...

From the numbers we can conclude Meteor Lake is not as power efficient as Intel want you to believe. You needs 115W PL2 to really hit the performance level approaching AMD's 780M. So much for Centrino and Intel 4 process node, if anyone still believe in IFS's skill, wait until you see CPU from Intel 20A/18A...

That's explain why Intel delays the launching of Ultra 9 cause Ultra 9 comes with 45W TDP and 2.5/2.3GHz base clock....

And if you watch the full video you'll find that actual power numbers using AIDA64 system stability test on the stress FPU setting are provided. Which show that the ZenBook 14 is actually running 28W while the Legion Pro 16 is around 68W. Meanwhile the referenced MechRevo 14+ shows 54W. Once it goes down to the 35W level the CB23 score apparently drops to 12810 according to the video.

In case others are interested in looking through the source videos which are being referred to I've included below. Note that despite being in Chinese without subtitles there's enough English and numbers in the screenshots to get some information at least.

Zenbook 14:

Legion Pro 16:

MechRevo 14+: https://www.youtube.com/watch?v=y7afgiUYKFI

H433x0n · Dec 23, 2023

SiliconFly said:
Thats while running under full load. Remember, CB benchmarking is a stress test. MTL isn't designed for that and will surely not fare well in it.

The power efficiency improvement under daily usage scenario should be a lot higher with the new pcode. Lets see real world power usage test results with the new pcode.

It will do fine under load with the updated microcode. If GoldenPig’s numbers are to be believed, the efficiency curve is fine. You don’t need to make caveats for it.

Abwx · Dec 23, 2023

SiliconFly said:
So, with Win12 are we getting the "rear-view mirror"?

Only rear-view man know the answer, he ll tell you that hardware is no more relevant, only the software is, albeit he could change his mind...

H433x0n said:
It will do fine under load with the updated microcode. If GoldenPig’s numbers are to be believed, the efficiency curve is fine. You don’t need to make caveats for it.

Actually efficency may well have decreased..

If you set the CPU at say 28W and then increase the time it s turboing at say 50W you ll measure a higher score at the "28W" point, same at the 35W point FI, that s the "efficency improvement" they got, actually increasing the boost time and doing like if the CPU was at 28W and other tested point, just a big farce.

Philste · Dec 23, 2023

Tigerick said:
From the numbers we can conclude Meteor Lake is not as power efficient as Intel want you to believe. You needs 115W PL2 to really hit the performance level approaching AMD's 780M.

A Zenbook 14 as one of the lightest and smallest Ultrabooks definitely can't keep a 64W PL1 cooled, I bet the real power draw is much lower.

SiliconFly · Dec 23, 2023

H433x0n said:
It will do fine under load with the updated microcode. If GoldenPig’s numbers are to be believed, the efficiency curve is fine. You don’t need to make caveats for it.

I don't think we should overestimate RWC. It's just tiled version of RPC (or GLC+). So, when it comes to power efficiency gains, we can't expect much from it except for the node gains from Intel 7 to Intel 4 which is 20% best case (and nothing else significant I guess).

MTL real efficiency comes from it's whole architecture rather than RWC I think.

mikk · Dec 23, 2023

They have a pre installed Windows 10 in China or why don't he use Windows 11?

H433x0n · Dec 23, 2023

Abwx said:
Actually efficency may well have decreased..

If you set the CPU at say 28W and then increase the time it s turboing at say 50W you ll measure a higher score at the "28W" point, same at the 35W point FI, that s the "efficency improvement" they got, actually increasing the boost time and doing like if the CPU was at 28W and other tested point, just a big farce.

I tend to believe GoldenPig & UltraBookReview know how to review products and take accurate measurements. However, it’s possible they both screwed up. I don’t think it’s the most likely scenario but who knows.

We’ll see when more people get their hands on production laptops with the updated Bios.

controlflow · Dec 23, 2023

Tigerick said:
ZenBook 14 Sample ZenBook 14 Sample ZenBook 14 Retail Lenovo Legion Pro 16 Retail MechRevo 14+ Retail
Tested By UBR UBR 中正评测中正评测中正评测
ICU-7 155H ICU-5 125H R7 7840HS
6P + 8E 4P + 8E 8P
24 MB 18 MB 16 MB
32G 7467 32G 7467 16G 6400
128EU 112EU 768SP
1 Fan 2 Fans 2 Fans
75 WHr 84 WHr 60 WHr
BIOS Tested 201 203 203 MECN42WW 2.17
TDP 28 W 28 W 35 W
PL1 (Long) 64 W 85 W 54 W
PL2 (Short) 64 W 115 W NA
Base 1.6/1.4 GHz 1.4/1.2 GHz 3.8 GHz
Turbo 4.8 GHz 4.5 GHz 5.1 GHz
CineBench R23 12048 13873
10 mins 11126 12132 11306 14567 16058
Single T 1751 1761 1729 1641 1788
AIDA Memory Latency 143.9 ns 133.7 ns 105.0 ns
Battery Life (Standard, Video Playback) 5:15 Hours 5:05 Hours NA
Assassin's Creed FHD LQ 45 46 58
Tomb Raider FHD LQ 59 63 66
Video Editing CPU Only 164 sec 132 sec 123 sec
Video Editing CPU + GPU 48 sec 34 sec NA

I have compiled numbers from UltraBook Review and Chinese Youtuber 中正评测. He bought the retail version of both ZenBook 14 and Lenovo's Legion 16 Pro; both videos are available at Youtube. For comparison, I have listed AMD 7840HS's numbers as well...

From the numbers we can conclude Meteor Lake is not as power efficient as Intel want you to believe. You need 115W PL2 to really hit the performance level approaching AMD's 780M. So much for Centrino and Intel 4 process node, if anyone still believe in IFS's skill, wait until you see CPU from Intel 20A/18A...

That's explain why Intel delays the launching of Ultra 9 cause Ultra 9 comes with 45W TDP and 2.5/2.3GHz base clock....

The 115W number is completely irrelevant and not happening in a Zenbook 14. That is a single fan 2.8 lb laptop with a tiny heatsink that can barely sustain 64W for more than a few seconds, forget even hitting 115W.

DaaQ · Dec 23, 2023

Wow the adroc of intel is here, I wouldn't have known without reading last 3 pages.

cebri1 · Dec 24, 2023

controlflow said:
The 115W number is completely irrelevant and not happening in a Zenbook 14. That is a single fan 2.8 lb laptop with a tiny heatsink that can barely sustain 64W for more than a few seconds, forget even hitting 115W.

They are using PL1 and PL2 numbers straight out from the specs sheet. No indication in any of those test it hit those power levels.

It's like posting 13900K gaming numbers and saying it needed 250W to achieve those results. Complete baloney.

Thunder 57 · Dec 24, 2023

controlflow said:
The 115W number is completely irrelevant and not happening in a Zenbook 14. That is a single fan 2.8 lb laptop with a tiny heatsink that can barely sustain 64W for more than a few seconds, forget even hitting 115W.

The real question is since when is 115W acceptable for a laptop? For awhile we seemed to settle on 35W and that seemed a bit much. Soon enough we were in the 15W range.

DAPUNISHER · Dec 24, 2023

Any chance they release a desktop APU?

TESKATLIPOKA · Dec 24, 2023

DAPUNISHER said:
Any chance they release a desktop APU?

Highly doubt It. I don't think they have enough volume for that.

FlameTail · Dec 24, 2023

Thunder 57 said:
The real question is since when is 115W acceptable for a laptop? For awhile we seemed to settle on 35W and that seemed a bit much. Soon enough we were in the 15W range.

Perhaps take a look at Macbooks and what they do - arguably the best laptops in the industry.

Thunder 57 · Dec 24, 2023

FlameTail said:
Perhaps take a look at Macbooks and what they do - arguably the best laptops in the industry.

Nah, when they have to claim that 8GB of RAM is equal to 16GB on a PC I think I've heard enough.

cebri1 · Dec 24, 2023

DAPUNISHER said:
Any chance they release a desktop APU?

Didn't they confirmed All-in-Ones back in September? But that market should be quite small so it may take some time, most corporate environments have laptops + dockers.

TESKATLIPOKA · Dec 24, 2023

cebri1 said:
Didn't they confirmed All-in-Ones back in September? But that market should be quite small so it may take some time, most corporate environments have laptops + dockers.

But that can simply use laptop parts.

mikk · Dec 24, 2023

Core Ultra 5 125H DTT On vs DTT Off

In 3 games DTT off runs faster, 1 game DTT on faster, 1 game equal. GPU clock speeds reporting is wrong or not? Seems low, not sure. CPU clock is wrong for sure.

On MTL devices with LP 7500 the memory can be downclocked to 5600 by the looks of it. SAGV seems to be off by default which is faster but also not as efficient. If it's off in battery mode this might be not so great.

majord · Dec 25, 2023

SiliconFly said:
So, with Win12 are we getting the "rear-view mirror"?

yes , just like how win 8 made Bulldozer bulldoze sandy bridge

CouncilorIrissa · Dec 25, 2023

SiliconFly said:
On the contrary, majority of the so called "power efficiency" test results that we've come across so far has squarely focused on stressing the CPU tile alone while COMPLETELY ignoring the LPE cores, which I believe it completely wrong as it doesn't align with MTL's use cases. Just give MTL a chance to prove itself in lighter load tests with newer pcode. Thats all I'm sayin.

But if all you care is light and idle-level loads, then you'd be better off buying an M1-based Macbook Air, a three year old chip that is still unmatched by every non-Apple manufactured chip.

You also get to access the OS that does not BSOD on you (and spies just as well).

SiliconFly · Dec 25, 2023

CouncilorIrissa said:
But if all you care is light and idle-level loads, then you'd be better off buying an M1-based Macbook Air, a three year old chip that is still unmatched by every non-Apple manufactured chip.

You also get to access the OS that does not BSOD on you (and spies just as well).

M1 is a lot slower.

Also, Windows 11 is more stable than Macs and doesn't have the Spinning Wheel of Death like macOS.

CouncilorIrissa · Dec 25, 2023

SiliconFly said:
M1 is a lot slower.

Also, Windows 11 is more stable than Macs and doesn't have the Spinning Wheel of Death like macOS.

But you yourself suggested measuring MTL in an idle, non-performance sensitive scenario. Why is performance important in scenarios when you're barely touching a keyboard (the only ones that LP-E cores are actually doing anything)?

And regarding Windows' 11 stability: I guess my ROG Ally didn't get the memo about its stability when it crashed into BSOD while playing Genshin Impact, failing to boot afterwards and forcing me to do cloud recovery that resulted in the Windows 11 being reinstalled without a freaking Microsoft Store.

	ZenBook 14 Sample	ZenBook 14 Sample	ZenBook 14 Retail	Lenovo Legion Pro 16 Retail	MechRevo 14+ Retail
Tested By	UBR	UBR	中正评测	中正评测	中正评测
			ICU-7 155H	ICU-5 125H	R7 7840HS
			6P + 8E	4P + 8E	8P
			24 MB	18 MB	16 MB
			32G 7467	32G 7467	16G 6400
			128EU	112EU	768SP
			1 Fan	2 Fans	2 Fans
			75 WHr	84 WHr	60 WHr
BIOS Tested	201	203	203	MECN42WW	2.17
TDP			28 W	28 W	35 W
PL1 (Long)			64 W	85 W	54 W
PL2 (Short)			64 W	115 W	NA
Base			1.6/1.4 GHz	1.4/1.2 GHz	3.8 GHz
Turbo			4.8 GHz	4.5 GHz	5.1 GHz
CineBench R23	12048	13873
10 mins	11126	12132	11306	14567	16058
Single T	1751	1761	1729	1641	1788
AIDA Memory Latency			143.9 ns	133.7 ns	105.0 ns
Battery Life (Standard, Video Playback)			5:15 Hours	5:05 Hours	NA
Assassin's Creed FHD LQ			45	46	58
Tomb Raider FHD LQ			59	63	66
Video Editing CPU Only			164 sec	132 sec	123 sec
Video Editing CPU + GPU			48 sec	34 sec	NA

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

Senior member

Attachments

Platinum Member

Senior member

Senior member

Golden Member

Senior member

Lifer

Member

Golden Member

Diamond Member

Senior member

Member

Golden Member

Member

Platinum Member

Super Moderator CPU Forum Mod and Elite Member

Platinum Member

Platinum Member

Platinum Member

Member

Platinum Member

Diamond Member

Attachments

Senior member

Member

Golden Member

Member