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

Page 529 - 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
782
750
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.



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: 24,025
  • LNL.png
    LNL.png
    881.8 KB · Views: 25,517
Last edited:

jdubs03

Golden Member
Oct 1, 2013
1,257
889
136
diamond rapids is 2026. It will compete against Zen6. CWF I agree it could shake the current landscape but we’ll see.
Right right, probably early 2026. And that “Panther Cover-X” core should be quite competitive with Zen 6. And if it releases before, another bonus.
 
Last edited:

Doug S

Diamond Member
Feb 8, 2020
3,346
5,865
136
NO, when driving even a smartphone can use the car to receive calls, and be charged by car. The laptop does not need to be used in a car, but can charge from there.

SOMEBODY please give me a use case where you NEED a battery, and its longer than 8 hours (or so) and no power of any kind is available. In a reasonable use case. Playing games for a businessman is not a valid use case. Even students have external power there they are not in class. And they should not be playing games most places but be in class or studying.

Even if someone can find such a use case, it doesn't mean that smartphones or laptops should be designed with the needs of the few in mind. The people who need that kind of thing have options to buy smartphones with ridiculous batteries in the Android world. If they want an iPhone and even the Pro Max's lengthy battery life is too little for them, or they want an Android but don't like those models with the giant battery capacity they can get one of those cases with a built in battery (yeah it makes the phone bigger but so would putting in a gargantuan 10000 mah battery these cretins want)

If they don't want one of those cases then they can carry one of those little mini brick portable chargers. Then when they are stuck in these mythical places where they need to have their phone going constantly but have no access to a charger, they can charge it off their portable charger. And if they complain about "waaah! I don't want to carry another thing with me" then I'll play them a tune on the world's smallest violin.

In the laptop world you might need more of a "brick" portable charger but if @9949asd wants to cool in the cafe not being plugged in I guess he can hit it in his bag so unless someone notices the cord going into his bag they won't know he isn't cool :cool:
 

AcrosTinus

Senior member
Jun 23, 2024
213
220
76
I think Turin will have a considerable lead in efficiency. Performance wise I’m not so sure. There is a significant gap to cover.
Let us wait and see, both have their advantages. At work though I will be pushing for XEON6 adoption.
 
  • Like
Reactions: 511

Doug S

Diamond Member
Feb 8, 2020
3,346
5,865
136
Indeed, Intel clocks higher at the cost of density. And recent TSMC nodes can now achieve similar clocks with less density penalty.
Not to mention 18A was downgraded from what was previously expected, so it is now N3P PPA overall instead of N2.
Not to mention the cost and throughput are worse than equiv TSMC nodes, yields are also worse but serviceable.

Isn't TSMC's SRAM cell size smaller than Intel's, even with TSMC's little backtrack between N3B and N3E? Given that cache is no longer shrinking that's a more and more important factor as time goes on.
 
  • Like
Reactions: lightmanek

naukkis

Golden Member
Jun 5, 2002
1,004
849
136
Both process and core just aren't quite there yet to match TSMC + zen. GNR 3.2ghz all core is quite low

You really cannot directly compare such a wildly different hardware implementations. Intel mesh sure takes a lot of power to feed data - and that is compared to vcache Epyc where data is as locally stored as possible. Workloads that doesn't share data would favor Epyc hardware implementation - massively. But Intel mesh is giving 2x performance against that V-cache Epyc in some workloads - and probably Phonorix test won't even have workloads where mesh advantage is biggest.

Phoronix didn't show or test how different SNC configuratios change situation. Sub clustering disabled test would give glimpse of actual interconnect efficiency.
 

dttprofessor

Member
Jun 16, 2022
163
45
71
Granite rapids is for HPC & enterprise market, it's much profitable than 4th&5th exon which are no profit for intel.
Sierra forest is for classical cloud, it's like a big E5V4.
 
  • Like
Reactions: SiliconFly

cannedlake240

Senior member
Jul 4, 2024
247
138
76
You really cannot directly compare such a wildly different hardware implementations. Intel mesh sure takes a lot of power to feed data - and that is compared to vcache Epyc where data is as locally stored as possible. Workloads that doesn't share data would favor Epyc hardware implementation - massively. But Intel mesh is giving 2x performance against that V-cache Epyc in some workloads - and probably Phonorix test won't even have workloads where mesh advantage is biggest.

Phoronix didn't show or test how different SNC configuratios change situation. Sub clustering disabled test would give glimpse of actual interconnect efficiency.
Epyc 9655 96C/400w apparently has a boost clocks of 4.1Ghz, 28% higher than 96C GNR. Add the Zen 5 IPC advantage and they're back to having a massive lead. Some of that's probably due to yield with 600mm2 tiles and the large mesh
 

511

Diamond Member
Jul 12, 2024
3,204
3,148
106
With Clearwater forest they can do 7:1 consolidation of Ice lake servers easily

96 core GNR vs 96 Core Genoa
 

Josh128

Golden Member
Oct 14, 2022
1,078
1,631
106
Lion Cove and Zen 5 perform within 1-2% of each other, with Lion Cove getting a 1-2% edge in integer performance as well as efficiency going off of David Huang’s data. The 1T performance is going to be completely dependent on the laptop OEM and the chassis it is in. For all intents and purposes, the 1T performance is the same.


Agreed, it is apples to oranges. That’s why trying to compare nT performance makes no sense. Despite the defensiveness of multiple posters, nobody is saying Strix is disappointing.
Indeed, they are, I was just pointing out the hypocrisy. Somebody was definitely saying Strix was disappointing, see below. I know this is an Intel thread, but its always good to stay in the realm of reality. Krakan Point will be AMDs direct competition to Lunar. Obviously will have weaker iGPU, but even on 4nm will have comparable 1T, nT, and power efficiency. Apparently AMD thought they were going to target "premium" AI laptops with Strix rather than thin and light despite its ability to do so, and prioritized that over Krakan thinking it would be more profitable.

Siliconfly wrote:

[About Krakan Point] Why bother about a silly step-down product when Strix itself is getting hammered this gen?

AMD Strix is will remain a disappointment this gen.
 
Last edited:

511

Diamond Member
Jul 12, 2024
3,204
3,148
106
GNRs must not be much expensive than SPR based on RCP though RCP is totally wrong many times but if this is true why buy Sapphire They are DOA 🤣
Screenshot_20240927-184800.png
 

Hitman928

Diamond Member
Apr 15, 2012
6,649
12,274
136
Didnt know 43% is almost double.

BTW, measured at wall: 14" LNL; 16" Strix

Correct, that’s power at the wall with different screens/systems. For SoC power, MTL, LNL, and STX were all limited to 28W. While that doesn’t mean they all actually used that much, there shouldn’t be much difference in SoC power during gaming.
 

Joe NYC

Diamond Member
Jun 26, 2021
3,322
4,858
136
And you expect what from It?
It should be 4*Zen5 + 4*Zen5c and 8CU IGP according to leaks.
I would not expect any miracles, but in theory, ST performance of Kraken should match ST performance of Strix Point. MT performance should be in Lunar Lake range.

But one thing that can definitely be expected from Kraken is being a mainstream CPU, position to gain volume shipments.

Lunar Lake, Strix Point and Snapdragon X are all fighting on the same turf in $1,400 range price, Kraken based laptops are going to be at $1,000 and below, as an MSRP.

PS: I see Snapdragon already being discounted heavily. I noticed Samsung Snapdragon based laptops at $999, with $450 (!!!) discount from $1,499
 

Josh128

Golden Member
Oct 14, 2022
1,078
1,631
106
Didnt know 43% is almost double.

BTW, measured at wall: 14" LNL; 16" Strix
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.
 
  • Like
Reactions: lightmanek

cebri1

Senior member
Jun 13, 2019
373
405
136
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.
ah, the irony.