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.

 

[ondma]

And one more exciting bit of info that i came across recently:

AMD's Infinity Fabric is substrate-based (outdated tech). And Intel Foveros is interposer-based. Meaning, Foveros is light-years ahead of Infinity Fabric.

Interestingly, Foveros has up to 10X power-efficiency and very low latency compared to Infinity Fabric. Intel also has nextgen improved foveros too in the pipeline for ARL & LNL. Whereas, AMD has no such solution for Zen 5. I think AMD is in a lot of trouble. It appears MTL has managed to unleash the beasts Intel had in its unused arsenal.

The only "beast" Intel has right now is power consumption. RL "refresh" is a horrible idea. Eke out a few more percent performance at the expense of even more outrageous power consumption. Meteor Lake sounds like a replay of Tiger Lake. New, more efficient chip confined to mobile, but a only a refresh on desktop. Tiger Lake actually, except for the absolute performance crown, probably would have been a better product even on desktop, compared to whatever skylake refresh they sold at the time. I am afraid Meteor Lake is going to be the same situation. Another "refresh" also makes me very apprehensive that Arrow Lake is going to be delayed (even more) and/or mediocre in performance.

What I really would like to see, now that Intel hopefully is moving to a more efficient process with a bigger transistor budget is more P cores, not continual spamming of e cores. I really wonder if 8 P cores is going to compete well with 16 cores from AMD when games get even more demanding. (The 5090 is rumored to be a very big performance increase.) Right now, the e cores do very little or nothing for gaming. I think Intel will have to figure some way to leverage the e cores in gaming to remain competitive.

BTW, most on these forums know that I would like to see Intel be competitive or even beat AMD, so I am not an Intel basher by any means. I would love to see them remain competitive with AMD or even gain the lead. AldL/RL was a very good step for them, but I fear they are falling behind again.

 

[SiliconFly]

The only "beast" Intel has right now is power consumption. RL "refresh" is a horrible idea. Eke out a few more percent performance at the expense of even more outrageous power consumption. Meteor Lake sounds like a replay of Tiger Lake. New, more efficient chip confined to mobile, but a only a refresh on desktop. Tiger Lake actually, except for the absolute performance crown, probably would have been a better product even on desktop, compared to whatever skylake refresh they sold at the time. I am afraid Meteor Lake is going to be the same situation. Another "refresh" also makes me very apprehensive that Arrow Lake is going to be delayed (even more) and/or mediocre in performance.

What I really would like to see, now that Intel hopefully is moving to a more efficient process with a bigger transistor budget is more P cores, not continual spamming of e cores. I really wonder if 8 P cores is going to compete well with 16 cores from AMD when games get even more demanding. (The 5090 is rumored to be a very big performance increase.) Right now, the e cores do very little or nothing for gaming. I think Intel will have to figure some way to leverage the e cores in gaming to remain competitive.

BTW, most on these forums know that I would like to see Intel be competitive or even beat AMD, so I am not an Intel basher by any means. I would love to see them remain competitive with AMD or even gain the lead. AldL/RL was a very good step for them, but I fear they are falling behind again.

RPL refresh is a bad joke. A stop gap measure something we should forget as a bad dream. Thankfully, it's only for the desktop where power efficiency isn't that big a deal. And with the end of RPL refresh, Intel finally gets rid of the damn outdated power hungry node Intel 7. Good riddance. And forget RPL refresh.

And with Intel 4 out of the picture by early 2025, Intel finally gets rid of all the godforsaken outdated baggage they've been carrying all these years. All thanks to all the paper-pushing, being-counter, illiterate & incompetent CEOs who were running Intel into the ground (to put it mildly). Intel's true turnaround starts only next year. Sadly, this year they're taking a huge leap just to catch up!

 

[SiliconFly]

Just an MTL analysis. Rumors and leaks suggest MTL's ST performance uplift will be in a single digit which I totally agree. And we already know MTLs gonna be very power efficient. But by how much?

(1) The AT V/F curve suggests, Intel 4 provides 35% to 40% power efficiency which is in line with Intel's claims. Meaning, MTL can operate at significantly less power at iso-freq. I just did some math, and it clearly shows that a MTL cpu can easily run at less than 30W compared to an equivalent RPL cpu running at 45W. And thats just purely because of the node shrink alone.

(2) And rumors suggest MTL is getting DLVR. Thats another solid 15% to 20% boost in power efficiency.

(3) Rumors suggest that the LP E-cores in the SoC tile has the capability to completely turn off the CPU tile & GPU tile while idling which massively saves power.

(4) I've seen MTL ES idling at around 480 MHz (idle clock) which is impressive. Probably the SoC E-cores.

To summarize, if Intel can get all these tech to work in tandem with stable BIOS & proper Windows patches, MTL power-efficiency will be off the charts! Might easily provide all-day battery life for latops under standard productivity workloads. Meaning, it's not just going to be better than Zen 4. I think it's gonna be better than M2 macs as well in power efficiency. And it's coming soon...

 

[Anhiel]

Welp, nothing really new and all exactly within my previous predictions and calculation including how bad Panther Lake would be.

So the change with these new Cougar Cove & "Beast Lake" is what's changing things.
I suppose the 8+32 configuration is up again after it's clear Zen6 would get ~24-32 cores (depend on CCD have 8,12 or 16 cores).
If I assume ARL gets the ST increase as claimed and Zen5/Zen6 each get 20-25 then ARL and Zen6 will be head to head with Intel winning with clockspeed. On the low end Intel will win with their new L4 dies and power Via. I presume the L4 cache die will be standard on all SKUs. Zen6 has to be on TSMC's N2 to compete with backside power delivery, 3D cache or onboard DRAM. Let's hope that's becoming standard. Regardless all these will cost a lot for consumers.
Beast Lake sounds like a prove of concept design for Royal taken up or it's just a refresh combined with a node change or improved node.

Anyhow, the real important news is that Intel has partnered with the japanese RIKEN. This will change future designs beyond Nova Lake and its refreshe(s).
Judging from this article

AMD Talks Hybrid Ryzen CPU Concepts, Avoiding Intel's AVX-512 Problem

AMD wants to avoid Intel's mistakes.

www.tomshardware.com

AMD is already on the ball so 1-2 gens ahead on the future designs. Just look at the ISSCC presentation the last few years.
So Zen6's successor or new architecture will be interesting unless AMD decides to go with another Zen architecture round. That would turn to be their loss.

 

[eek2121]

This image shows exactly what i mentioned previously. Roughly between 16% to 24% improvement in frequency at iso-power with Intel 4 8VT cells compared to Intel 7. Averaging at 20% performance gains like Intel's Intel 4 claims.

60% is not shown anywhere in this pic.

DLVR & the new ULP E-cores sound very exciting. Waiting for more info abt these...

We already know the frequency uplift is 60%. Did you miss the screenshots?

 

[Exist50]

And one more exciting bit of info that i came across recently:

AMD's Infinity Fabric is substrate-based (outdated tech). And Intel Foveros is interposer-based. Meaning, Foveros is light-years ahead of Infinity Fabric.

Interestingly, Foveros has up to 10X power-efficiency and very low latency compared to Infinity Fabric. Intel also has nextgen improved foveros too in the pipeline for ARL & LNL. Whereas, AMD has no such solution for Zen 5. I think AMD is in a lot of trouble. It appears MTL has managed to unleash the beasts Intel had in its unused arsenal.

The advantages and disadvantages of different packaging techniques are far more complex than that. You really need to factor in cost as well. AMD makes monolithic chips for mobile (i.e. what MTL will compete against), and have demonstrated plenty of willingness to adopt exotic packaging. They just have very different considerations from Intel this gen. Seems unnecessary to reduce this to black vs white.

 

[A///]

never read so much bullocks in my life and I read the tabloids at the checkout.

 

[SiliconFly]

We already know the frequency uplift is 60%. Did you miss the screenshots?

No I didn't.

Performance uplift of 60% for ARL over MTL is pretty much an viable option. But not frequency uplift. Thats impossible.

MTL is expected to run at around 5 GHz easily. And a 60% frequency uplift will put ARL at 8 GHz which just can't happen.

 

[SiliconFly]

never read so much bullocks in my life and I read the tabloids at the checkout.

Anything worthwhile to contribute rather than snide comments. Or at least please be clear on who ur targeting. Your vague comments might hurt more people than you think.

 

[SiliconFly]

The advantages and disadvantages of different packaging techniques are far more complex than that. You really need to factor in cost as well. AMD makes monolithic chips for mobile (i.e. what MTL will compete against), and have demonstrated plenty of willingness to adopt exotic packaging. They just have very different considerations from Intel this gen. Seems unnecessary to reduce this to black vs white.

Actually, it's crucial. Both techniques are pretty much performant. And AMD's focus seems to on cost-savings. While Intel's focus seems to be on power-efficiency and heavy tile use. AMD has the cost advantage. But Intel now has technical superiority. It matters very much on the long run.

 

[A///]

Anything worthwhile to contribute rather than snide comments. Or at least please be clear on who ur targeting.

the mystery is what makes it all fun!

 

[SiliconFly]

the mystery is what makes it all fun!

Maybe for you. But not for others. If you want to point out bs, jut quote it. Makes life simple for everyone & it tends to avoid unnecessary misunderstandings.

 

[Geddagod]

In short, raichu simply says, it's fake.

I swear, every other leaker besides MLID thinks it is fake

I think AMD is in a lot of trouble. It appears MTL has managed to unleash the beasts Intel had in its unused arsenal.

AMD is staying monolithic for mobile CPUs anyway, which IMO, is a better strategy since the die size for mobile isn't that large.
Maybe if MTL has a HCC model that competes against Dragon Range, Intel's interconnect advantage might be useful, but until then...
As for server, I think the fact that Mesh consumes way more power due to how Intel connects it's cores and cache vs AMD negates the EMIB advantage.

 

[Henry swagger]

And one more exciting bit of info that i came across recently:

AMD's Infinity Fabric is substrate-based (outdated tech). And Intel Foveros is interposer-based. Meaning, Foveros is light-years ahead of Infinity Fabric.

Interestingly, Foveros has up to 10X power-efficiency and very low latency compared to Infinity Fabric. Intel also has nextgen improved foveros too in the pipeline for ARL & LNL. Whereas, AMD has no such solution for Zen 5. I think AMD is in a lot of trouble. It appears MTL has managed to unleash the beasts Intel had in its unused arsenal.

Foveros omni will be even bigger.. enhanced super glue😏

 

[A///]

I swear, every other leaker besides MLID thinks it is fake

mlid's a dumba**

 

[DrMrLordX]

mlid's a dumba**

The name is Dumas(s)

 

[uzzi38]

Actually, it's crucial. Both techniques are pretty much performant. And AMD's focus seems to on cost-savings. While Intel's focus seems to be on power-efficiency and heavy tile use. AMD has the cost advantage. But Intel now has technical superiority. It matters very much on the long run.

It would be crucial if AMD's mainstream mobile platforms switched over to a chiplet design.

They're not. And no, I do not consider Dragon Range and it's successors to be "mainstream", same goes for other larger designs we'll perhaps be seeing in the future.

 

[uzzi38]

No I didn't.

Performance uplift of 60% for ARL over MTL is pretty much an viable option. But not frequency uplift. Thats impossible.

MTL is expected to run at around 5 GHz easily. And a 60% frequency uplift will put ARL at 8 GHz which just can't happen.

In no universe is ARL going to be 60% faster than MTL.

Also I don't think MTL is expected to hit 5GHz when even MTL-H45 ES2 caps at 4.5GHz 1T (but I will be clear - I don't know MTL QS targets or anything, I'm guessing here).

https://twitter.com/x/status/1664375853021802496

I think Arrow Lake is going to be a great generation, but 60% is just too far.

 

[Exist50]

In no universe is ARL going to be 60% faster than MTL.

Also I don't think MTL is expected to hit 5GHz when even MTL-H45 ES2 caps at 4.5GHz 1T (but I will be clear - I don't know MTL QS targets or anything, I'm guessing here).

https://twitter.com/x/status/1664375853021802496

I think Arrow Lake is going to be a great generation, but 60% is just too far.

I wouldn't worry about MTL frequency so much. They'll probably hit somewhere north of 5GHz. Maybe not by much, and that still doesn't put MTL in a good position, but I think the concerns there are a tad overblown. We're basically seeing the process ramp up in real time. If they're on ES2, hopefully we should see QS samples start popping up by August.

I think Arrow Lake is going to be a great generation, but 60% is just too far.

Yeah, 60% ST is very unrealistic. Hopefully Lion Cove still has some decent IPC gains, but it's only really +1 gen over Golden Cove. And I can't see huge increases in frequency going from "Intel 7 Ultra", or whatever they're calling it, to N3. Maybe they'll land in the 6GHz range, but certainly not 7GHz. MT will probably a more interesting story from a couple of angles.

 

[mikk]

In no universe is ARL going to be 60% faster than MTL.

Also I don't think MTL is expected to hit 5GHz when even MTL-H45 ES2 caps at 4.5GHz 1T (but I will be clear - I don't know MTL QS targets or anything, I'm guessing here).

Just a reminder that ADL-S ES2 125W managed 4.6 Ghz 1T Turbo. 5 Ghz is expected for MTL.

Exclusive info about Intel’s Alder Lake-S – Will the Intel Core-1800 be a threat to AMD? | igor´sLAB

Disclaimer: The following article is machine translated from the original German, and has not been edited or checked for errors. Thank you for understanding!

www.igorslab.de

 

[eek2121]

No I didn't.

Performance uplift of 60% for ARL over MTL is pretty much an viable option. But not frequency uplift. Thats impossible.

MTL is expected to run at around 5 GHz easily. And a 60% frequency uplift will put ARL at 8 GHz which just can't happen.

I see where the disconnect is. I was talking about base clocks. I was not implying boost clocks would see a 60% uplift. I brought up base clocks because it tells us MTL-S is way more efficient than Raptor Lake. 1.9-3.4 ghz is the uplift.

As you go up in frequency, efficiency goes out the window. You would not see a 60% increase at the top end, it would be something more like 5-10%. However, that doesn’t matter. Intel is going to be conservative with clocks on these chips to keep power consumers down and keep them from being faster than upper tier parts.

 

[nicalandia]

Some people never learn...

 

[Tup3x]

Some people never learn...

View attachment 81361

That's just Intel version of AMDGPU_.

https://twitter.com/AMDGPU_

 

[TESKATLIPOKA]

That's just Intel version of AMDGPU_.

https://twitter.com/AMDGPU_

Maybe It's the same guy with 2 accounts.

 

[Exist50]

Maybe It's the same guy with 2 accounts.

"I'm playing both sides so that I always come out on top"