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)	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.)

 

[mikk]

It's nothing new and btw there is a newer bios available with a bump from 2x to 3x. Good Bios support from Asus, the others are slower. For the Acer there is zero bios update on the support page.

BIOS Update for Windows
Version 301
2023/12/22
IMPORTANCE: CRITICAL
Important Information: Optimize system performance

 

[TESKATLIPOKA]

It's nothing new and btw there is a newer bios available with a bump from 2x to 3x. Good Bios support from Asus, the others are slower. For the Acer there is zero bios update on the support page.

BIOS Update for Windows
Version 301
2023/12/22
IMPORTANCE: CRITICAL
Important Information: Optimize system performance

At least we know which brand we should prefer. Lenovo

 

[mikk]

At least we know which brand we should prefer. Lenovo

It's from Asus. Lenovo has a poor support page, I can't check without a Lenovo.

 

[PJVol]

Nothing new there, It was already mentioned here.
And you even call It as a huge efficiency uplift.

That's right, don't let yourself be Greg'ed (© GN)

 

[eek2121]

I think the future of laptop RAM is this:

(1) On-package LPDDR

or

OEMs would flee if they couldn’t charge a premium for memory and AMD/Intel made the money instead. Dell practically invented a standard to avoid this.

So, just for the record, your though Intel was going to dominate mobile with lipstick on a power hog?

Every test I have seen shows MTL-P to be competitive with the 7840. Doesn’t seem like a power hog to me? Can Intel do better? probably. Is the chip better than previous Intel designs in terms of efficiency? Absolutely. Is it competitive with AMD? 100%. Will it outsell Zen 4 mobile? Without a doubt. Do I need a haircut? Yes.

 

[TESKATLIPOKA]

It's from Asus. Lenovo has a poor support page, I can't check without a Lenovo.

I know It's from Asus.

 

[SiliconFly]

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.

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.

 

[TESKATLIPOKA]

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.

Then what is It designed for? Doing nothing or very light work?
For that you need 16 cores? Why do you even need so much cores, If you won't use them?

You really should think It trough before you write something, because what you wrote is BS.

 

[SiliconFly]

Then what is It designed for? Doing nothing or very light work?
For that I need 16 cores? You really should think It trough before you wrote such a colossal BS.

Not really. Running CB to gauge power efficiency is BS as no real users in the world will do such nonsense.

Put it thru real world usage scenarios. Test it yourself if you have to. Only then you'll realize the difference.

 

[TESKATLIPOKA]

Not really. Running CB to gauge power efficiency is BS as no real users in the world will do such nonsense.

Checking out efficiency during full load is certainly not a BS like you claim. How you can write such a BS like this is beyond me. Not everyone is doing nothing on their laptops.

Put it thru real world usage scenarios. Test it yourself if you have to. Only then you'll realize the difference.

If you call watching videos or doing very light work on your laptop as real world usage scenarios, then you clearly don't need 16core MTL for that. You would be better off buying a Mackbook with better efficiency.

BTW, before you tell others to test something out, you should first do It yourself.

 

[Abwx]

Not really. Running CB to gauge power efficiency is BS as no real users in the world will do such nonsense.

Put it thru real world usage scenarios. Test it yourself if you have to. Only then you'll realize the difference.

You forgot that in single thread usage it use 20-25W, depending of the setting, so if even a fully loaded single thread is too much one has to wonder what is daily usage...

 

[coercitiv]

Every test I have seen shows MTL-P to be competitive with the 7840. Doesn’t seem like a power hog to me? Can Intel do better? probably. Is the chip better than previous Intel designs in terms of efficiency? Absolutely. Is it competitive with AMD? 100%. Will it outsell Zen 4 mobile? Without a doubt. Do I need a haircut? Yes.

Follow the entire discussion pls, I'm not the one claiming the P core is a power hog. My question was merely a recall of multiple claims made by the other poster. You know, the kind of quick recap you get at the beginning of a new episode from your favorite show, which is great if you missed the last one while getting a haircut!

 

[SiliconFly]

Checking out efficiency during full load is certainly not a BS like you claim. How you even can write such a BS like this is beyond me. Not everyone is doing nothing on their laptops.

CB represents clearly less than 1% of the real world power usage scenarios. If you want to know MTL's real efficiency, we need to run tests that reflect its design purpose.

BTW, before you tell others to test something out, you should first do It yourself.

Before calling others BS, you should first think thru.

 

[SiliconFly]

It's really hilarious when people assume that MTL's LPE cores are not performant and completely useless!!!

They're pretty powerful in their own right. Their performance is pretty close to that of regular E cores. They are pretty much capable of doing lots of daily workload stuff without ever firing up the CPU tile. Thats MTL's design philosophy.

And testing power efficiency with ALL the P cores up and running simultaneously and continuously with CB is completely useless as the LPE cores never gets their due diligence.

 

[SiliconFly]

You forgot that in single thread usage it use 20-25W, depending of the setting, so if even a fully loaded single thread is too much one has to wonder what is daily usage...

CB?

 

[TESKATLIPOKA]

CB represents clearly less than 1% of the real world power usage scenarios. If you want to know MTL's real efficiency, run tests that reflect its design purpose.

Why don't you say It clearly, MTL is not as good as PHX during load so that's why you want to ignore It and only look where It wins.

Before calling others BS, you should first think thru.

Did you try out a MTL laptop? No
Yet you tell me to check It out. Does that make sense?
If you want to show how great MTL is, then give us charts.

 

[TESKATLIPOKA]

CB?

No, Geekbench.

 

[Abwx]

CB?

Yes, but with Integer code power would be the same, this is documented, contrary to Zen 4 cores RPL and MTL cores use as much power in Integer than in FP.

That being said if you re talking of efficency when doing nothing that s another matter...

 

[SiliconFly]

No, Geekbench.

GB? yikes.

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

 

[SiliconFly]

...you want to ignore It and only look where It wins...

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.

 

[igor_kavinski]

Can't you guys wait for WIN12 before discussing MTL's efficiency? That's what the CPU is built for. Let's go silent at least until Zen 5 laptop chips are launched. Why go around and around in circles, especially with SiliconFly who can't stop thinking how sexy the LP cores are.

 

[SiliconFly]

Can't you guys wait for WIN12 before discussing MTL's efficiency? That's what the CPU is built for. Let's go silent at least until Zen 5 laptop chips are launched. Why go around and around in circles, especially with SiliconFly who can't stop thinking how sexy the LP cores are.

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

 

[igor_kavinski]

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

If it's as AI centric as rumored to be, current Zen 4 mobile chips may not fare as well.

 

[PJVol]

Why go around and around in circles, especially with SiliconFly

Because he is this Greg

 

[igor_kavinski]

Imagine the arguing here when Zen4c parts enter the picture