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

Page 72 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Tigerick

Senior member
Apr 1, 2022
696
602
106
PPT1.jpg
PPT2.jpg
PPT3.jpg



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

ModelCode-NameDateTDPNodeTilesMain TileCPULP E-CoreLLCGPUXe-cores
Core Ultra 100UMeteor LakeQ4 202315 - 57 WIntel 4 + N5 + N64tCPU2P + 8E212 MBIntel Graphics4
?Lunar LakeQ4 202417 - 30 WN3B + N62CPU + GPU & IMC4P + 4E08 MBArc8
?Panther LakeQ1 2026 ??Intel 18A + N3E3CPU + MC4P + 8E4?Arc12



Comparison of die size of Each Tile of Meteor Lake, Arrow Lake, Lunar Lake and Panther Lake

Meteor LakeArrow Lake (20A)Arrow Lake (N3B)Lunar LakePanther Lake
PlatformMobile H/U OnlyDesktop OnlyDesktop & Mobile H&HXMobile U OnlyMobile H
Process NodeIntel 4Intel 20ATSMC N3BTSMC N3BIntel 18A
DateQ4 2023Q1 2025 ?Desktop-Q4-2024
H&HX-Q1-2025
Q4 2024Q1 2026 ?
Full Die6P + 8P6P + 8E ?8P + 16E4P + 4E4P + 8E
LLC24 MB24 MB ?36 MB ?12 MB?
tCPU66.48
tGPU44.45
SoC96.77
IOE44.45
Total252.15

LNL-MX.png

Intel Core Ultra 100 - Meteor Lake

INTEL-CORE-100-ULTRA-METEOR-LAKE-OFFCIAL-SLIDE-2.jpg

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.)



Clockspeed.png
 

Attachments

  • PantherLake.png
    PantherLake.png
    283.5 KB · Views: 24,006
  • LNL.png
    LNL.png
    881.8 KB · Views: 25,490
Last edited:

BorisTheBlade82

Senior member
May 1, 2020
680
1,069
136
Yup. But they only launched that as mobile only because 10nm was still not ready at that point . IIRC Tiger Lake launched as 4 core parts first and then 8 core parts, and never had exceptional volume until a while after launch too.
ICL-U was indeed only and exclusively a 4c part mobile-only.
TGL was firstly launched as a much improved 4c part. Much much later it was launched as an 8c NUC only part.
 
  • Like
Reactions: Geddagod

uzzi38

Platinum Member
Oct 16, 2019
2,705
6,427
146
Why would Intel rename Raptor Cove to Redwood Cove for a 1% improvement? I'm not denying the claim just curious as I would think the one thing that Intel would know with clarity early in the design process through simulations is how one core will compare against another. I remember watching a video with Geller I believe saying simulations are amazingly accurate these days.

1% improvement or on the order of a few percent is usually just tweaking memory things or perhaps OoO logic (Broadwell), no architectural changes. On one hand this makes sense as they generally don't make big changes when going to a node. But on the other hand why would Intel plan a ML desktop released with less cores that are no more performant than the predecessor?

New designs get a new codename, that's all. Doesn't really matter how large the changes are. See: Willow Cove vs Sunny Cove, or Golden Cove vs Raptor Cove.

Redwood Cove does bring changes, but the overall performance improvement per clock will be tiny.
 

DrMrLordX

Lifer
Apr 27, 2000
22,065
11,693
136
It's not possible cos I don't think Intel would work on their cpu tile on two different nodes at the same time! It's unprecedented. Never happened before!

Lots of things that have never happened before are happening (or will happen) at Intel. Like using TSMC for anything.

Plus Intel almost directly ported Nehalem over to 32nm as Gulftown, so while I doubt the work was done simultaneously, there was probably some overlap.
 
  • Like
Reactions: BorisTheBlade82

Hulk

Diamond Member
Oct 9, 1999
4,524
2,515
136
New designs get a new codename, that's all. Doesn't really matter how large the changes are. See: Willow Cove vs Sunny Cove, or Golden Cove vs Raptor Cove.

Redwood Cove does bring changes, but the overall performance improvement per clock will be tiny.

So they basically "ticked" Raptor Lake to create Meteor and Arrow will be the "tock?"
 

SiliconFly

Golden Member
Mar 10, 2023
1,541
897
96
I'm not following. Are you saying ML clocks will be 10% higher than Raptor Lake?
That would be fantastic but I can't see Intel 4 clocks matching Intel 7 for years, if ever.

What I find very surprising is the fact that most people tend to forget the performance gains when an cpu core (logic) is moved from an old node to a more advanced node.

For example, Zen 3 was on TSMC N7P. And Zen 4 is on TSMC N4. Thats a one full node jump (slightly more actually). That gave Zen 4 upto 25% performance increase or upto 50% reduction in power (but not both) over Zen 3 due to the node jump alone. So, AMD chose 10% to 15% of the PPW budget for performance increase for Zen 4 and used the remaining PPW budget for power efficiency.

The increase in Zen 4 performance over Zen 3 is mostly not because of architectural changes, but mainly due to the node jump from TSMC N7 to N4.

What people forget is, the exact same advantage applies to Intel too.

Meteor Lake is shifting from Intel 7 to Intel 4. And it's a complete "full" node jump. Intel 4 offers upto 20% performance increase or upto 40% reduction in power (but not both) over Intel 7. If Intel uses the PPW budget properly, Meteor Lake will end up with 10% performance increase over RPL at a given frequency & will be 20% more power efficient than RPL.

Meaning, when it comes to IPC, Meteor Lake cpu @ 5.4GHz will be as fast as RPL cpu @ 6.0GHz.

MTL doesn't even have to hit 6GHz to beat RPL. It just has to be at 5.5GHz or slightly above. Thats all!
 
Last edited:

SiliconFly

Golden Member
Mar 10, 2023
1,541
897
96
Lots of things that have never happened before are happening (or will happen) at Intel. Like using TSMC for anything.

Plus Intel almost directly ported Nehalem over to 32nm as Gulftown, so while I doubt the work was done simultaneously, there was probably some overlap.

If I remember right, they skipped both nehalem & westmere and jumped straight to sandy bridge for desktop & laptop cpus
 

DrMrLordX

Lifer
Apr 27, 2000
22,065
11,693
136
What I find very surprising is the fact that most people tend to forget the performance gains when an cpu core (logic) is moved from an old node to a more advanced node.

Not all node improvements increase clockspeed. Look at what happened with Intel went from 32nm to 22nm, or 14nm++(+?) to 10nm+. Clockspeed regression.
 
  • Like
Reactions: igor_kavinski

Hulk

Diamond Member
Oct 9, 1999
4,524
2,515
136
What I find very surprising is the fact that most people tend to forget the performance gains when an cpu core (logic) is moved from an old node to a more advanced node.

I don't understand? Are you referring to IPC increases due to the new architecture having more transistors, which are in turn allowed do to the increased transistor density of the new node?

Or due to frequency increase of the new node.

32nm Sandy Bridge to 22nm Ivy Bridge was a clock speed regression.

22nm Haswell to 14nm Broadwell was a clock speed regression.

14nm Comet Lake to 10nm Canon Lake, Ice Lake, and Tiger Lake was also a clock speed regression. It took quite a few "+'s" for Intel to get the clocks ramped up at 10nm. Actually they started in 2015 and now in 2023 they're still at it with 10nm.

It doesn't seem like node changes are the instantly result in less power, more frequency, and greater transistor density that used to be the case when shrinking a node. As we approach atomic sized features that pesky quantum physics is causing problems that take time to be worked around.


Meaning, when it comes to IPC, Meteor Lake cpu @ 5.4GHz will be as fast as RPL cpu @ 6.0GHz.

MTL doesn't even have to hit 6GHz to beat RPL. It just has to be at 5.5GHz or slightly above. Thats all!

Assuming Intel's first whack at Intel 4 could hit 5.5GHz you are also assuming a 9% IPC increase for ML compared to RL. IntelUser2000 a few posts above already wrote it looks like ML IPC gains will be really small, perhaps 1%.

It's all possible of course but it seems hard to believe Intel 4 will be hitting 5.5GHz in 2023.
 

SiliconFly

Golden Member
Mar 10, 2023
1,541
897
96
Not all node improvements increase clockspeed. Look at what happened with Intel went from 32nm to 22nm, or 14nm++(+?) to 10nm+. Clockspeed regression.

Clock speed regression is always expected in a new node. Thats why I mentioned:

MTL doesn't even have to hit 6GHz to beat RPL. It just has to be at 5.5GHz.

And if they manage to bring in any new uArch performance improvements, that also adds up.

Even if they bring in a meager 5% uArch performance improvement to MTL over RPL, Meteor Lake will beat RPL at 5.2GHz

Intel has something excellent in its hand right now. Just wondering they don't eff it up like before! :eek:
 

SiliconFly

Golden Member
Mar 10, 2023
1,541
897
96
I don't understand? Are you referring to IPC increases due to the new architecture having more transistors, which are in turn allowed do to the increased transistor density of the new node?

Or due to frequency increase of the new node.

32nm Sandy Bridge to 22nm Ivy Bridge was a clock speed regression.

22nm Haswell to 14nm Broadwell was a clock speed regression.

14nm Comet Lake to 10nm Canon Lake, Ice Lake, and Tiger Lake was also a clock speed regression. It took quite a few "+'s" for Intel to get the clocks ramped up at 10nm. Actually they started in 2015 and now in 2023 they're still at it with 10nm.

It doesn't seem like node changes are the instantly result in less power, more frequency, and greater transistor density that used to be the case when shrinking a node. As we approach atomic sized features that pesky quantum physics is causing problems that take time to be worked around.




Assuming Intel's first whack at Intel 4 could hit 5.5GHz you are also assuming a 9% IPC increase for ML compared to RL. IntelUser2000 a few posts above already wrote it looks like ML IPC gains will be really small, perhaps 1%.

It's all possible of course but it seems hard to believe Intel 4 will be hitting 5.5GHz in 2023.

1% IPC gains is just plain wrong.

And yes, clock speed regression is expected.

When a foundry mentions 20% PPW they say performance increase due to additional logic and caches in the die due to increase in logic density alone. And not just due to frequency increase itself.

Like I said, if Intel does it right, MTL @ 5.5GHz will beat RPL @ 6GHz

If you include a meager single digit 5% uArch improvement, MTL @ 5.2GHz will comfortably beat RPL @ 6GHz
 
Last edited:

Hitman928

Diamond Member
Apr 15, 2012
6,186
10,693
136
Clock speed regression is always expected in a new node. Thats why I mentioned:

MTL doesn't even have to hit 6GHz to beat RPL. It just has to be at 5.5GHz.

And if they manage to bring in any new uArch performance improvements, that also adds up.

Even if they bring in a meager 5% uArch performance improvement to MTL over RPL, Meteor Lake will beat RPL at 5.2GHz

Intel has something excellent in its hand right now. Just wondering they don't eff it up like before! :eek:

How will MTL at 5.5 GHz beat RPL at 6 GHz without any architecture performance improvements? That makes zero sense.
 

Hulk

Diamond Member
Oct 9, 1999
4,524
2,515
136
According to my sources The 14900K is a refresh Raptor Cove design on Intel Super 7+, Highly binned to extract the most out of the process

Wouldn't this mean that only the 13900 sku's would need the new process because the lower bins could have increased clocks off the "old" Intel 7 process.

14600K could move to 5.3GHz all core, 14700K could be 5.6GHz all core and 14900K perhaps 5.9GHz all core. If that's all the Refresh turns out to be I hope they don't slap a "14" on it.
 

Hulk

Diamond Member
Oct 9, 1999
4,524
2,515
136
If I remember right, they skipped both nehalem & westmere and jumped straight to sandy bridge for desktop & laptop cpus

Nehalem was desktop and mobile. Westmere was Nehalem without the Integrated Memory Controller (performance suffered as expected) and was also desktop and mobile.
 

Exist50

Platinum Member
Aug 18, 2016
2,452
3,102
136
Yup. But they only launched that as mobile only because 10nm was still not ready at that point . IIRC Tiger Lake launched as 4 core parts first and then 8 core parts, and never had exceptional volume until a while after launch too.
The 8c die's design was staggered after the 4c one. They probably could have used it for desktop, but at that point, wasn't worth the overhead of switching away from Rocket Lake. Remember that a 6c Rocket Lake mobile part was planned as well. That would have been a disaster.
 

Exist50

Platinum Member
Aug 18, 2016
2,452
3,102
136
I think 5.5GHz is actually a pretty reasonable estimate for where we'll see Meteor Lake land. Heck, even for mobile chips. But obviously, that doesn't make for a particularly compelling desktop chip, assuming negligible IPC gains. I'm also rather worried about gaming performance with the MTL/ARL architecture. Gaming is quite sensitive to memory latency, and there's not just the move to chiplets, but also their SoC fabric has apparently seen a lot of change, and been the source of much trouble. Lost the main architect midway through, apparently. Even if Arrow Lake provides nice core gains, they might be muted by SoC considerations.
 

SiliconFly

Golden Member
Mar 10, 2023
1,541
897
96
How will MTL at 5.5 GHz beat RPL at 6 GHz without any architecture performance improvements? That makes zero sense.

A increase in density usually gives an opportunity to increase L2/L3 cache without any significant changes to the architecture. This in turn increases IPC. Meaning, more performance at the same frequency.
 
Last edited:
  • Like
Reactions: Henry swagger

SiliconFly

Golden Member
Mar 10, 2023
1,541
897
96
I think 5.5GHz is actually a pretty reasonable estimate for where we'll see Meteor Lake land. Heck, even for mobile chips. But obviously, that doesn't make for a particularly compelling desktop chip, assuming negligible IPC gains. I'm also rather worried about gaming performance with the MTL/ARL architecture. Gaming is quite sensitive to memory latency, and there's not just the move to chiplets, but also their SoC fabric has apparently seen a lot of change, and been the source of much trouble. Lost the main architect midway through, apparently. Even if Arrow Lake provides nice core gains, they might be muted by SoC considerations.

MTL is Intel's first mainstream venture into chiplets (ignoring the now defunct Lakefield). And I'm sure it'll face a ton of issues. Thats why I mentioned a while ago that 2023 isn't Intel's year. AMD has a good chance to beat MTL this year. But, if Intel does things right, they may not fallback behind too much.
 

Geddagod

Golden Member
Dec 28, 2021
1,296
1,368
106
The 8c die's design was staggered after the 4c one. They probably could have used it for desktop, but at that point, wasn't worth the overhead of switching away from Rocket Lake. Remember that a 6c Rocket Lake mobile part was planned as well. That would have been a disaster.
Its a shame no one picked up the 11980HK on top of chinese motherboards and tried OCing it to the moon to compare perf at different wattages vs RKL (11900k).
Ik people have tested the 11900H online, but IIRC the 11980HK mobos has better VRMs and power delivery than the 11900H's mobos, so more OC potential. Cooling also was fine on the 11900H while OCing IIRC, so I don't think that would be an issue either.