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

Page 646 - 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
700
615
106






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 + 4E012 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 (N3B)Lunar LakePanther Lake
PlatformMobile H/U OnlyDesktop & Mobile H&HXMobile U OnlyMobile H
Process NodeIntel 4TSMC N3BTSMC N3BIntel 18A
DateQ4 2023Desktop-Q4-2024
H&HX-Q1-2025
Q4 2024Q1 2026 ?
Full Die6P + 8P8P + 16E4P + 4E4P + 8E
LLC24 MB36 MB ?12 MB?
tCPU66.48
tGPU44.45
SoC96.77
IOE44.45
Total252.15



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.)



 

Attachments

  • PantherLake.png
    283.5 KB · Views: 24,009
  • LNL.png
    881.8 KB · Views: 25,496
Last edited:

mzocyteae

Junior Member
Dec 29, 2020
6
3
51
Based on the benches below it seems as though the L2 is where ARL is falling behind RPL, yes?

View attachment 110559

View attachment 110560
Those numbers don't make sense, and most likely a bugged benchmark.
If ARL L2 is that bad, its performance definitely should sunk much more.

According to RPL L2 numbers, this benchmark is measuring p-core (32B/c read @5.6G), excluding e-core.
While the ARL numbers are unclear yet, the LNL L2 numbers are known: p-core 32B/c; e-core ~27B/c.
With ARL p-core 5.4G/e-core 4.6G, this benchmark's numbers are way off (much lower than e-core) and looks broken.
 

GTracing

Member
Aug 6, 2021
138
321
106
Widening of SIMD unit is undeniably going to happen to 512Bits AVX10 clears the fragmentation issues

Oh yes the community revolted on this CnC and Felix jad a nice writeup
I think your wording is causing some confusion. There's a difference between supporting 512bit instructions and having a 512bit wide pipeline. For example, Strix Point supports AVX512, but they do so through double-pumping a 256bit SIMD pipeline. I think most people agree that Intel and AMD will continue to use double pumping on client CPUs.

The point of Felix Leclair's article was that AVX10/128 should be dropped. And as Kepler said, AVX10/128 is dead. It was dropped in the AVX10.2 spec.
 
Reactions: Elfear

Hitman928

Diamond Member
Apr 15, 2012
6,330
11,171
136
This doesn't make sense cause SPEC and R23 also showed that Arrow Lake made a generational increase over Raptor Lake.
Do we also disregard SPEC and R23 because these also failed to show regression in Arrow Lake?

IMO, these benchmarks should be taken as a guide and let real world productivity/gaming tests show the increases for your workflows.

Who did a Spec comparison for ARL? I haven’t seen it yet.
 

MS_AT

Senior member
Jul 15, 2024
311
691
96
I think your wording is causing some confusion. There's a difference between supporting 512bit instructions and having a 512bit wide pipeline. For example, Strix Point supports AVX512, but they do so through double-pumping a 256bit SIMD pipeline. I think most people agree that Intel and AMD will continue to use double pumping on client CPUs.

The point of Felix Leclair's article was that AVX10/128 should be dropped. And as Kepler said, AVX10/128 is dead. It was dropped in the AVX10.2 spec.
The leaks so far suggests that Intel will stick to AVX10/256 as efficient implementation of AVX10/512 with "double-pumping" would require full 512b shuffle units as shown by Zen4. That would not be doable if they want E-cores to keep using 128b execution units.

AMD hasn't yet announced anything about the AVX10 support, therefore we don't know if they will go Zen4/Zen5 mobile route [support 512b instructions but have 256b execution units], Zen 5 route [full 512b support] or Intel route [stick to 256b units].

Who did a Spec comparison for ARL? I haven’t seen it yet.
David Huang has the scores for 265k on his blog SPEC(int) section, but he hasn't produced an article about Arrow Lake afaik.https://blog.hjc.im/spec-cpu-2017
 

511

Senior member
Jul 12, 2024
746
659
96
I think your wording is causing some confusion. There's a difference between supporting 512bit instructions and having a 512bit wide pipeline. For example, Strix Point supports AVX512, but they do so through double-pumping a 256bit SIMD pipeline. I think most people agree that Intel and AMD will continue to use double pumping on client CPUs.
Looks like it i was doing multiple things so i screwed it up
 
Reactions: GTracing

MS_AT

Senior member
Jul 15, 2024
311
691
96
This, is, big big disappointment if true. Unless Intel ditch big-little hybrid to have full 512bit support but it wouldn't happen until Intel combine big and little cores.
Why disappointment? AVX10/256 doesn't mean you cannot use 128b, just that you have to support also 256b. This means if E-cores stay with 128b units, they will use the same trick Zen4 used for 512b compatibility, but targeting 256b compatibility. Nevertheless both cores will get other benefits like 32 architectural registers, masking and new instructions. This should be a noticeable boost, especially since spilling SIMD registers is expensive and it will also mean that Intel will reach ARM Neon parity in consumer market as Neon has had 32 architectural registers from the beginning.
 

GTracing

Member
Aug 6, 2021
138
321
106
Looks like it i was doing multiple things so i screwed it up
No worries. I might have been a little aggressive with my response. I just get frustrated when people use vague language and argue over things that they actually might agree on.

The leaks so far suggests that Intel will stick to AVX10/256 as efficient implementation of AVX10/512 with "double-pumping" would require full 512b shuffle units as shown by Zen4. That would not be doable if they want E-cores to keep using 128b execution units.

AMD hasn't yet announced anything about the AVX10 support, therefore we don't know if they will go Zen4/Zen5 mobile route [support 512b instructions but have 256b execution units], Zen 5 route [full 512b support] or Intel route [stick to 256b units].
Thanks for the correction. I guess I meant to say that they aren't necessarily moving to 512bit wide SIMD execution pipelines.

This, is, big big disappointment if true. Unless Intel ditch big-little hybrid to have full 512bit support but it wouldn't happen until Intel combine big and little cores.
Intel could also add AVX10/256 support to the little cores.

Here's the only source I could find on it. But the tweet is from the same guy who wrote the Chips and Cheese article. Sounds like it's actually AVX10.3 that drops 128, not AVX10.2, so I got that wrong.

 

511

Senior member
Jul 12, 2024
746
659
96
M4 Pro is 1600 r24 ( from slides)(45-50W based on M3)
M2 Pro is 1000
U9 185(1000 at 65W) iirc
Strix is 1250(65W)

 

Attachments

  • Screenshot_20241030-203638.png
    375.1 KB · Views: 14

Kosusko

Member
Nov 10, 2019
185
155
116

Intel's Arrow Lake chips aren't winning any awards for gaming performance but I think its new E-cores deserve a gold star

source: https://www.pcgamer.com/hardware/pr...t-i-think-its-new-e-cores-deserve-a-gold-star


P.S. Paradigm shift: 12c .LITTLE Atom Skymont E-cores beats 8C big Lion Cove P-cores !
source:
 
Last edited:
Reactions: igor_kavinski

511

Senior member
Jul 12, 2024
746
659
96
GPU performance in what? Doesn't looks like It's in a game.
Don't ask me i don't make apple slides they didn't mention anywhere what did they test.I am pretty sure that LNL and M4 GPU are within 10% of each other.
They also claim M4 Pro 20 Core gpu is 2X faster than M4 GPU(10 Core Version) additionally it has 2X bandwidth it would be half

It’s up to 2.2x faster than the CPU in M1 Max and up to 2.5x faster than the latest AI PC chip. M4 Max
The GPU features up to 20 cores for graphics performance that is 2x that of M4, and up to 2.4x faster than the latest AI PC chip M4 Pro
Apparently 2.4X of 258V and 2X for M4 apparently U9 is 5% faster so M4 iGPU is roughly 15% faster CPU being 150% faster
 
Reactions: TESKATLIPOKA

Hitman928

Diamond Member
Apr 15, 2012
6,330
11,171
136
The leaks so far suggests that Intel will stick to AVX10/256 as efficient implementation of AVX10/512 with "double-pumping" would require full 512b shuffle units as shown by Zen4. That would not be doable if they want E-cores to keep using 128b execution units.

AMD hasn't yet announced anything about the AVX10 support, therefore we don't know if they will go Zen4/Zen5 mobile route [support 512b instructions but have 256b execution units], Zen 5 route [full 512b support] or Intel route [stick to 256b units].


David Huang has the scores for 265k on his blog SPEC(int) section, but he hasn't produced an article about Arrow Lake afaik.https://blog.hjc.im/spec-cpu-2017

Thanks, I don’t see any real increase from RPL to ARL in that test. The 265k is slightly behind in raw score and ~2% ahead in IPC.

Edit: Updated IPC comment as I used the wrong freq. initially to compare.
 
Last edited:
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |