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

[Tigerick]

Wildcat Lake (WCL) Preliminary Specs

Intel Wildcat Lake (WCL) is upcoming mobile SoC replacing ADL-N. WCL consists of 2 tiles: compute tile and PCD tile. It is true single die consists of CPU, GPU and NPU that is fabbed by 18-A process. Last time I checked, PCD tile is fabbed by TSMC N6 process. They are connected through UCIe, not D2D; a first from Intel. Expecting launching in Q2/Computex 2026. In case people don't remember AlderLake-N, I have created a table below to compare the detail specs of ADL-N and WCL. Just for fun, I am throwing LNL and upcoming Mediatek D9500 SoC.

		Intel Alder Lake - N	Intel Wildcat Lake	Intel Lunar Lake		Mediatek D9500
Launch Date		Q1-2023	Q2-2026 ?	Q3-2024		Q3-2025
Model		Intel N300	?	Core Ultra 7 268V		Dimensity 9500 5G
Dies		2	2	2		1
Node		Intel 7 + ?	Intel 18-A + TSMC N6	TSMC N3B + N6		TSMC N3P
CPU		8 E-cores	2 P-core + 4 LP E-cores	4 P-core + 4 LP E-cores		C1 1+3+4
Threads		8	6	8		8
Max Clock		3.8 GHz	?	5 GHz
L3 Cache		6 MB	?	12 MB
TDP		7 W	Fanless ?	17 W		Fanless
Memory		64-bit LPDDR5-4800	64-bit LPDDR5-6800 ?	128-bit LPDDR5X-8533		64-bit LPDDR5X-10667
Size		16 GB	?	32 GB		24 GB ?
Bandwidth			~ 55 GB/s	136 GB/s		85.6 GB/s
GPU		UHD Graphics		Arc 140V		G1 Ultra
EU / Xe		32 EU	2 Xe	8 Xe		12
Max Clock		1.25 GHz		2 GHz
NPU		NA	18 TOPS	48 TOPS		100 TOPS ?

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.

 

[igor_kavinski]

Does Lion Cove even have such "disabled AVX-512 and HT circuitry"?

I hope so! Otherwise they are screwed!

At least when the sales get really bad, they can release some microcode update to unblock both.

It's kinda funny if you think about it. Maybe Intel's first CPU with close to P4 level madness and Conroe level elegance on the same package!

 

[511]

I hope so! Otherwise they are screwed!

At least when the sales get really bad, they can release some microcode update to unblock both.

It's kinda funny if you think about it. Maybe Intel's first CPU with close to P4 level madness and Conroe level elegance on the same package!

Made by TSMC to further Salt on their Wounds

 

[igor_kavinski]

And since Lion Cove has no server parts, it's likely they might have not included it.

What's taking up all that die space then? Ideas?

 

[igor_kavinski]

Made by TSMC to further Salt on their Wounds

More like red chilly

And the wound is where the sun don't shine

I think the MTL-S with 6Ps was a good idea in retrospect. ARL-S could've gotten an extra 8Es instead of two lousy Lions.

 

[igor_kavinski]

They are like sleeping volcanos.

Yeah, looks like you gonna waste at least $580 on the 285K

 

[SiliconFly]

.

 

[Tup3x]

I'm having trouble reconciling that slide with... reality. How did they manage to score 1370 on the XPS 9350 /w U7 256V?

Here's what a quick search turned out from a laptop review outlet:
View attachment 109954

It's indeed unplugged results. Still, even if they compared to plugged in results it looks pretty bad for Intel.

 

[Hulk]

The question then becomes what features are you looking for in an upgrade from a 1 year old top-end CPU (that doesn’t exhibit any instability)?

As I wrote it's honestly more of a just want the newest thing to fool around with.

Most of that gear is built into the Soundcraft UI24r stagebox.

 

[511]

Not much. Maybe @Hitman928 / @DavidC1 can share some interesting stuff.


Ha ha.. true. All the hype surrounding this gen cpus (both camps) should get back to normal once the reviewers get their hands on them. Making purchase decisions based on hype will make us end up with expensive duds.

MTL had a lot of hype. Many dodged the bullet safely after seeing the reviews. Just feel sad for those who took one for the team.

I took the bullet but i am happy cause i upgraded from 8250u 🙂 it depends from which platform did you upgrade from

 

[511]

It's indeed unplugged results. Still, even if they compared to plugged in results it looks pretty bad for Intel.

I have a bridge to sell you if you are buying Qualcomm marketing

 

[Magio]

Comparing against unplugged figures on an undisclosed lower power mode is absolutely ridiculous from Qualcomm. We have the power consumption figures so even if there are power modes in which LNL performance drops that much we still know that it absolutely holds its own against X Elite in ST PPW and that while it's weaker in MT it is capable of giving out its full performance on an unplugged power budget.

It's especially puzzling why they'd go to the trouble of such a biased comparison when showcasing their phone chip (on N3E) which actually does make LNL look quite underwhelming without such cheap tricks from QC.

 

[511]

phone is also running Linux Kernel vs Windows Kernel no one seems to point it out in these tests i thought Intel/AMD Marketing was the worst and than i saw Qualcomm

 

[poke01]

phone is also running Linux Kernel vs Windows Kernel no one seems to point it out in these tests i thought Intel/AMD Marketing was the worse and than i saw Qualcomm

Not just that it’s a laptop and a phone comparison is absurd to me.

 

[511]

Anyways when is the ARL Embargo ending?

 

[SiliconFly]

.

 

[Ghostsonplanets]

Anyways when is the ARL Embargo ending?

Should be either tomorrow or on the 24th.

 

[maddie]

If it is true that Arrow Lake has equal MT to 14900K, then it would seem to me that Arrow Lake is going to get beaten in MT by Zen 5. I must be missing something in this discussion.

Faith.

 

[AcrosTinus]

Such a beautiful CPU cannot be bad!

View attachment 109963

It is in fact not bad CPU and will run anything you throw at it...
We have to recenter our expectations, this is not early stage of computing anymore.
Without huge breakthrough or paradigm shifts to increase the transistor count everything is basically hitting a ceiling somewhere.

 

[Philste]

How did they manage to score 1370 on the XPS 9350 /w U7 256V?

Looking at other slides from the presentation, I think it has to do with performance while unplugged. The XPS laptop may be too aggressive with throttling while on battery.

Sadly Windows Power Plans are really bugged with Lunar Lake. My Zenbook S14 behaves the same way. Windows Power Plan Balanced will show these low results, CPU won't Boost and run at ~7W ST and ~13W MT through whole Geekbench 6. It's the same when plugged in.

But battery and plugged in is different power plans, so Qualcomm probably set it to performance and then unplugged it. You have to set it to performance seperately on battery.

And yes, it's overwriting ASUS Power modes, Asus full Power Mode with Windows Power Plan balance on charger still Delivers 1300-1400 ST score.

 

[511]

If Windows Disabled Turbo Boost than it is a Problem cause the base TDPs/Frequency are ridiculous my MTL can do. 6*2.3ghz + 8*1.8ghz at 29W in r24 while Intel Ark lists it at 45W for the same clocks

 

[OneEng2]

I do a lot of multitrack audio work. Playing a gig tomorrow actually that I will be recording and I run sound as well. Soundcraft UI24R stagebox mixer, controlled via wireless, 24 tracks, fx, etc... running the sytem in real time with no perceptible latency to the musicians, 8 aux mixes, feedback control...AND the ability to multitrack record all 24 tracks. It's nuts. 30 years ago it would have taken a good sized box truck full of racks of gear to do that.


I am going to take a good, hard look at the 9950X vs Arrow Lake if I upgrade this generation. I know, I know, there isn't much additional performance to be had from my 14900K. But the sad reality is there are more upgrades behind me than ahead of me if I upgrade at my old rate. This is a fun hobby for me and I splurge more on it these days.

What is the stock all-core clock for the 9950X BTW?

I have both the Ui24 and X32 Rack (I play in a band). I have a friend that has the CQ20B and one with an XR18. My point exactly. These little boxes are doing the processing of an entire 22 space rack full of gear from the early 90's and doing it in real time. All of this is possible through the use of hardware processing dedicated to digital sound algorithms.

My belief is that specific computing hardware can be many orders of magnitude (3 to 4 orders of magnitude) than general computing can be.

I believe that this will drive heterogeneous CPU design moving forward; however, Operating Systems will need to evolve to handle the much more complex scheduling needed.

It may have been present but is disabled considering the die size for ARL-S

Seems like a VERY expensive way to go. Why pay premium prices for N3B die space for a feature you have no intention of enabling? Additionally, Intel previously had thermal issues with running AVX512 due to the many more active transistors.

 

[moinmoin]

I believe that this will drive heterogeneous CPU design moving forward; however, Operating Systems will need to evolve to handle the much more complex scheduling needed.

We already have accelerators. GPUs are ones, sound cards pretty much were ones. Specifically lower latency sound mixing I honestly think that ship sailed ages ago already. We had excellent hardware mixing of sound already, but then Microsoft came and instead of standardizing hardware mixing made all the driver models incompatible with the new Windows OS release. Since then cheap on board sound and software mixing had been considered good enough for like 99.99% of all use cases.

 

[Hulk]

I believe that this will drive heterogeneous CPU design moving forward; however, Operating Systems will need to evolve to handle the much more complex scheduling needed.

Or maybe software developers will be able to specifically access certain cores in the CPU through the software for specific routines in the code.

This takes me back to the early '90's when I was using the CardD and Software Audio Workshop (SAW), written by Bob Lentini in Assembly. It was absolutely miraculous that you could multitrack record and mix with a pentium class computer. I was at Javits centers in NY way back when saw the demo of CardD and SAW. I bought them within a week, convining the band we HAD to have them. We also had a small recording studio. I used it instead of a DAT recorder to digitize final mixdown and for mastering. It was fabulous for the time and even holds up today. SAW, coded in Assembly was a few MB and ran from the exe file. It was stable, tiny, fast, and brilliant. I remember Bob had studied the Windows API for a bit and decided he couldn't do this through Windows, it was going to have to happen in Assembly and just made it happen.

Good software can be the way "around" insufficient hardware and it is the more elegant approach. We've already seen what good game drivers can do.

 

[Hitman928]

Not much. Maybe @Hitman928 / @DavidC1 can share some interesting stuff.

These pictures aren't really detailed enough for me to give any insights into individual modules. Most likely, a lot of it probably just comes down to Intel pushing bigger structures trying to push for higher frequencies, that's kind of been their M.O. for quite a while now.

 

[AMDK11]

What's taking up all that die space then? Ideas?

Including larger Front-End, 8x wider predictor, 8-Wide decode, L1-I 64KB, UOP Cache(L0) 5250, Queue 192, ROB 576, Branch Order Buffer 180, Scheduler FP 114 and PRF ~400, Scheduler INT 97 and PRF 290, Execution units on 10 (4x FP + 6x ALU) ports instead of 5 (3x FP/ALU + 2x ALU), Non-Scheduling-Queue Buffer, L0-D 48KB, L1-D 192KB and L2 3MB. + all resource control logic.

Also, splitting the unified FP/ALU scheduler with 97 entries for 5 execution ports into a separate scheduler for 6 ALU units with 97 entries and 4 FP units with 114 entries required large resources.