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.

 

[mikk]

i lost track of what post of yours i wanted to quote but i'd point mtl being like tigerlake on the scale but better.

Tigerlake-U was quite good. The biggest downside was the subpar multithread performance with max 4 cores. There are tons of devices still in the market.

Raichu being a Weibo resident and often just reposts stuff to Twitter has been known for ages now, this isn't some new revelation. I know for sure that I've said as much here before as well.

He does occasionally post slides and stuff that are legit, but there are also times where there are huge misses. What he's said about MTL is one of those cases.

Then please prove it. I would like to have links to his Lion Cove claims about the wide and increased ROB size. Maybe there is more in weibo which he didn't post. I'm also interested in the Redwood Cove voltage/frequency examination from here. Raichu is one of the most accurate on twitter, in fact he was one of the highest in the old Hardware Leak Tracker.

 

[S'renne]

Tigerlake-U was quite good. The biggest downside was the subpar multithread performance with max 4 cores. There are tons of devices still in the market.



Then please prove it. I would like to have links to his Lion Cove claims about the wide and increased ROB size. Maybe there is more in weibo which he didn't post. I'm also interested in the Redwood Cove voltage/frequency examination from here. Raichu is one of the most accurate on twitter, in fact he was one of the highest in the old Hardware Leak Tracker.

Is this happening because he was one of the highest in Hardware Leak Tracker? (no offense btw)

 

[uzzi38]

Tigerlake-U was quite good. The biggest downside was the subpar multithread performance with max 4 cores. There are tons of devices still in the market.



Then please prove it. I would like to have links to his Lion Cove claims about the wide and increased ROB size. Maybe there is more in weibo which he didn't post. I'm also interested in the Redwood Cove voltage/frequency examination from here. Raichu is one of the most accurate on twitter, in fact he was one of the highest in the old Hardware Leak Tracker.

You've already seen MTL-P/H (they're the same thing now) ES2 publicly. Admittedly the screenshot I know that came out of that didn't show clock speeds that were useful (was a task manager screenshot showing <1GHz only), but I can tell you now they're a good bit away from 5GHz on ES2.

QS may well hit 5GHz, but ES2 did not. Afaik QS isn't out yet though so I wouldn't hedge my bets either way.

 

[mikk]

You've already seen MTL-P/H (they're the same thing now) ES2 publicly. Admittedly the screenshot I know that came out of that didn't show clock speeds that were useful (was a task manager screenshot showing <1GHz only), but I can tell you now they're a good bit away from 5GHz on ES2.

QS may well hit 5GHz, but ES2 did not. Afaik QS isn't out yet though so I wouldn't hedge my bets either way.

I wasn't talking about MTL-H. Is that all you have? That's why you are saying he is getting his infos over the last few years from weibu, just because of the MTL-H ES2 clock speeds? That's a new low I would say. You need more than this, much more. To me it looks like you accusations are completely wrong, it's just that you don't like him and try to discredit him. That's my feeling I got.

 

[uzzi38]

I wasn't talking about MTL-H. Is that all you have? That's why you are saying he is getting his infos over the last few years from weibu, just because of the MTL-H ES2 clock speeds? That's a new low I would say. You need more than this, much more. To me it looks like you accusations are completely wrong, it's just that you don't like him and try to discredit him. That's my feeling I got.

?

I don't have any personal issues with Raichu at all, I don't know where you got that from. Aside from him gunking up my Twitter feed lol. I honestly don't really care about Raichu at all, the only reason I know about him being on Weibo is because a guy I know is on there and says that about Raichu. That guy also happens to work at an OEM himself, so when he calls someone out I know for sure it's real. He makes a big deal out of it every time, which is why I know for sure this isn't the first time it's happened. I just don't really care to keep track of one of the many guys posting stuff to Twitter.

I'm not really sure what's up with you honestly, you're treating what I said as a personal attack, when in reality all I'm saying is the nature of how he gets his information means it's not 100% accurate, and that there were and almost certainly will be in the future instances where he is wrong. I think those instances should be called out I will do so. This is one of those cases, because I've heard from two entirely different people the exact same thing: that MTL-H ES2 doesn't touch 5GHz. Both of which I'd trust over Raichu any day of the week because I know they have direct lines of information.

Look no offense Mikk, but rather than try to declare new lows perhaps you should recognise the real reason you don't like me saying something "negative" about Raichu (it's not actually negative in just explaining a way in which someone gets their info when it's still a private circle where one of the sources of said information is well protected from the general public, and that there very often is good information that comes out it, but just every now and then you get duds, and that 5GHz ES2 MTL-H is one of those duds.)

 

[Exist50]

?

I don't have any personal issues with Raichu at all, I don't know where you got that from. Aside from him gunking up my Twitter feed lol. I honestly don't really care about Raichu at all, the only reason I know about him being on Weibo is because a guy I know is on there and says that about Raichu. That guy also happens to work at an OEM himself, so when he calls someone out I know for sure it's real. He makes a big deal out of it every time, which is why I know for sure this isn't the first time it's happened. I just don't really care to keep track of one of the many guys posting stuff to Twitter.

I'm not really sure what's up with you honestly, you're treating what I said as a personal attack, when in reality all I'm saying is the nature of how he gets his information means it's not 100% accurate, and that there were and almost certainly will be in the future instances where he is wrong. I think those instances should be called out I will do so. This is one of those cases, because I've heard from two entirely different people the exact same thing: that MTL-H ES2 doesn't touch 5GHz. Both of which I'd trust over Raichu any day of the week because I know they have direct lines of information.

Look no offense Mikk, but rather than try to declare new lows perhaps you should recognise the real reason you don't like me saying something "negative" about Raichu (it's not actually negative in just explaining a way in which someone gets their info when it's still a private circle where one of the sources of said information is well protected from the general public, and that there very often is good information that comes out it, but just every now and then you get duds, and that 5GHz ES2 MTL-H is one of those duds.)

I think this whole discussion would be much easier if there was any actual link given to this content on Weibo before Raichu posts it. Without that evidence, is it unreasonable for someone to attribute him as the primary source? At best, it becomes a he-said-she-said.

 

[A///]

this conversation will be more interesting at wine o clock.

 

[Tigerick]

Seems like QS sample can only hit up to 4.8GHz for P core and 2.2GHz for tGPU, hopefully Ultra 9 would turbo up to 5GHz....

 

[msj10]

View attachment 82884

Seems like QS sample can only hit up to 4.8GHz for P core and 2.2GHz for tGPU, hopefully Ultra 9 would turbo up to 5GHz....

so RWC needs around 10% IPC just to match RPL. I guess the rumors about the IPC being only a low single improvement were wrong unless MTL has lower ST than RPL but I don't see that happening.

 

[Exist50]

View attachment 82884

Seems like QS sample can only hit up to 4.8GHz for P core and 2.2GHz for tGPU, hopefully Ultra 9 would turbo up to 5GHz....

No idea about the GPU, but they'll at least hit 5GHz. Reasonably confident of that. I think a top SKU in the 5.1-5.3 range is pretty likely.

I guess the rumors about the IPC being only a low single improvement were wrong unless MTL has lower ST than RPL but I don't see that happening.

About that...

 

[uzzi38]

I think this whole discussion would be much easier if there was any actual link given to this content on Weibo before Raichu posts it. Without that evidence, is it unreasonable for someone to attribute him as the primary source? At best, it becomes a he-said-she-said.

Ah yes, let me post a screenshot/link to a closed off private group that I already said I don't have access to personally. If you're so desperate to believe something multiple people have claimed is wrong (even publicly) then be my guest.

View attachment 82884

Seems like QS sample can only hit up to 4.8GHz for P core and 2.2GHz for tGPU, hopefully Ultra 9 would turbo up to 5GHz....

Back to reality now, yeah this is right currently.

 

[Exist50]

Ah yes, let me post a screenshot/link to a closed off private group that I already said I don't have access to personally. If you're so desperate to believe something multiple people have claimed is wrong (even publicly) then be my guest.

I'm not mikk, btw. But as I said, this is just a he-said-she-said. Why get angry at someone for asking for evidence for a particularly forceful claim being made? And you are kind of coming across as annoyed that people are taking Raichu seriously to begin with, despite an objectively good tract record. And regarding the MTL ES thing, there may not be a 5.1GHz ES2, but I don't believe that number was fabricated either, if that makes sense, so I don't see it as a real point of criticism, much less dismissal of everything he's posted.

Anyway, as I said, MTL will hit 5GHz, and likely a touch more. I don't think they're even shipping QS chips yet, so no idea what specific SKUs are out, but I'm reasonably confident of that claim for launch silicon. MTL has its problems, but the clocks on Intel 4 are tolerable enough. I do hope someone does a proper launch review of more than just the compute die though...

 

[uzzi38]

I'm not mikk, btw. But as I said, this is just a he-said-she-said. Why get angry at someone for asking for evidence for a particularly forceful claim being made? And you are kind of coming across as annoyed that people are taking Raichu seriously to begin with. There may not be a 5.1GHz ES2, but I don't believe that number was fabricated either, if that makes sense, so I don't see it as a real point of criticism.

Anyway, as I said, MTL will hit 5GHz, and likely a touch more. I don't think they're even shipping QS chips yet, so no idea what specific SKUs are out, but I'm reasonably confident of that claim for launch silicon. MTL has its problems, but the clocks on Intel 4 are tolerable enough. I do hope someone does a proper launch review of more than just the compute die.

I'm not annoyed at people taking Raichu seriously, I'm getting annoyed you're not listening to all of the people that have said publicly that this one specific rumour is not true. From my perspective, that rumour just wasn't true. Just because the numbers line up with expectations for what final silicon is aiming at, doesn't mean that at the time of the claim it was true.

If you think it's fine to pass on it, sure, feel free to do so, but I don't. Either it is what the ES2 does, or its an expectation of what final silicon will be able to do. You don't get to say it's one but treat it as the other. Letting little stuff like that fly is how people accept guys like MLID.

 

[uzzi38]

Oh and as an aside, I think at some point I said I thought GNR-AP would be decently competitive with Turin at the top end.

I'd like to recind that statement. I was very wrong. I'm sure it'll beat Genoa, but Turin is a bigger jump than I was expecting.

 

[Exist50]

I'm getting annoyed you're not listening to all of the people that have said publicly that this one specific rumour is not true

So where have I been saying anything about it? Frankly, I haven't been satisfied with any of the MTL ES claims/leaks and the general confusion the leakers seem to have between the different SKUs, milestones, etc., down to what's planned vs what's shipped. And I've been quite vocal about that.

I can see any number of ways for valid information to translate into misleading leaks, but for the purpose of setting expectations for final silicon, this isn't worth the crusade vs any number of far more baseless takes. And also doesn't address the Raichu vs Weibo leak question.

You see anyone who is projecting MTL silicon accurately? Because I sure haven't.

I'd like to recind that statement. I was very wrong. I'm sure it'll beat Genoa, but Turin is a bigger jump than I was expecting.

That, I'm curious about. I think the hype train has been going a little too hard for Zen 5 in general. I expect Turin to win, but not by anything huge. Still expect the two to be approximately competitive, all said and done.

 

[uzzi38]

So where have I been saying anything about it? Frankly, I haven't been satisfied with any of the MTL ES claims/leaks and the general confusion the leakers seem to have between the different SKUs, milestones, etc., down to what's planned vs what's shipped. And I've been quite vocal about that.

I can see any number of ways for valid information to translate into misleading leaks, but for the purpose of setting expectations for final silicon, this isn't worth the crusade vs any number of far more baseless takes.

The biggest issue has been that plans have been changing considerably for MTL/ARL for a while now. For the longest time nobody could guarantee what the SKU stack looked like. Things are probably locked in by this point, but we've seen entire dies get dropped in the last year or so (like the old 6+16 MTL and 8+32 ARL). We literally all just found out that neither 6+8 nor 8+16 for ARL is being fabbed in-house for the main core die, so nothing is really left in-house. So when it comes to specific SKUs the big issue is that it was all in flux for the longest time. I don't think anyone wanted to say anything about it. But I'm just guessing what other people are thinking at the end of the day.

As for what's planned vs what's shipped, that much at at the very least I think you can figure out with good filtering.

That, I'm curious about. I think the hype train has been going a little too hard for Zen 5 in general. I expect Turin to win, but not by anything huge. Still expect the two to be approximately competitive, all said and done.

I think I've made myself clear already. I'm purposefully moving away from that stance.

 

[Exist50]

The biggest issue has been that plans have been changing considerably for MTL/ARL for a while now. For the longest time nobody could guarantee what the SKU stack looked like.

To some degree, yes, but MTL-S was always on-again off-again. The MTL M/P roadmap never really changed. Little bearing on clocks either.

We literally all just found out that neither 6+8 nor 8+16 for ARL is being fabbed in-house for the main core die

There should be two 6+8 dies. N3B and 20A. I still don't think that 20A one is canceled. Not going to claim it as fact, but I certainly would not trust rumors.

I think I've made myself clear already. I'm purposefully moving away from that stance.

Yes, I was just explaining my own understanding of the matchup.

 

[adroc_thurston]

I think the hype train has been going a little too hard for Zen 5 in general

Wdym hype train; the samples for non-Dense Zen5 are everywhere.

I expect Turin to win, but not by anything huge. Still expect the two to be approximately competitive, all said and done.

HAHAHAHHAHAHAHA ohohohohohoohhohoho jesus.
GNR is approximately competitive with maybe Genoa if it's lucky.
Turin is just a different realm of existence.

 

[yuri69]

Turin is just a different realm of existence.

If this is the case then the usual suspects would leak something, right?

 

[adroc_thurston]

If this is the case then the usual suspects would leak something, right?

I assume soon™; Turin perf numbers are pretty widely known already.

 

[mikk]

View attachment 82884

Seems like QS sample can only hit up to 4.8GHz for P core and 2.2GHz for tGPU, hopefully Ultra 9 would turbo up to 5GHz....

On Sisoft there are tGPU entries with 2.25 Ghz:

Details for Result ID Intel(R) Graphics (1152S 128C SM3.0 2.25GHz, 4MB L2, 6.3GB) (OpenCL)

ranker.sisoftware.co.uk

It's exactly a TFLOPS doubling over Xe LP 1.5 Ghz (4.6 vs 2.3 tflops). By the way earlier this year Raichu told the original target was 2250-2300 Mhz. Seems like they can hit this target.

 

[Exist50]

Wdym hype train; the samples for non-Dense Zen5 are everywhere.

HAHAHAHHAHAHAHA ohohohohohoohhohoho jesus.
GNR is approximately competitive with maybe Genoa if it's lucky.
Turin is just a different realm of existence.

If you have numbers for both, than post them. Otherwise, this is unhelpful at best. As for GNR, are you seriously claiming that ~128c wouldn't even beat Genoa?

 

[eek2121]

I'm not annoyed at people taking Raichu seriously, I'm getting annoyed you're not listening to all of the people that have said publicly that this one specific rumour is not true. From my perspective, that rumour just wasn't true. Just because the numbers line up with expectations for what final silicon is aiming at, doesn't mean that at the time of the claim it was true.

If you think it's fine to pass on it, sure, feel free to do so, but I don't. Either it is what the ES2 does, or its an expectation of what final silicon will be able to do. You don't get to say it's one but treat it as the other. Letting little stuff like that fly is how people accept guys like MLID.

My info didn’t come from Raichu, and to be clear, I never said ES2 hit 5+ ghz. I said the highest speed reached by these chips was around 5.4ghz. That was for an early unfused sample. What I DID say was the final clocks would be 4.8-4.9 ghz, and well what do you know, the QS appears to be clocking in at 4.8 ghz.

View attachment 82884

Seems like QS sample can only hit up to 4.8GHz for P core and 2.2GHz for tGPU, hopefully Ultra 9 would turbo up to 5GHz....

Unless things have changed, there will be no Core 9 Ultra this gen. The i7 sample will clock around 4.8-4.9ghz and minor IPC/architectural improvements will carry the chip the rest of the way.

 

[TESKATLIPOKA]

Fastest 45W i7 13900HK has 5.4GHz turbo, If Meteor Lake will clock only at 4.9-5GHz then It needs 8-10% IPC gain just to be on par with It, anything less and you have regression in ST performance.
I think they will try to at least keep ST performance the same.
I kinda don't care about It unless It affects games. What I am more interested in is the MT, IGP and power consumption.

 

[mikk]

Unless things have changed, there will be no Core 9 Ultra this gen. The i7 sample will clock around 4.8-4.9ghz and minor IPC/architectural improvements will carry the chip the rest of the way.

There will be Ultra 9 variants.

你也改名？2分钟看懂全新酷睿产品命名规则_哔哩哔哩_bilibili

你也改名？2分钟看懂全新酷睿产品命名规则, 视频播放量 87830、弹幕量 159、点赞数 2085、投硬币枚数 154、收藏人数 569、转发人数 537, 视频作者 金猪升级包, 作者简介 既不独立，也不客观，更不公正，相关视频：英特尔14代酷睿重大变革：最高频率超6.2GHz，i9/i7/i5更名为Ultra9/Ultra7/Ultra5【宅同学】，全网最全电脑CPU性能排行榜！！！，英特尔Ultra系列和i系列会有什么差异？，英特尔新一代CPU又改名了？玩游戏到底选Ultra还是14代i9?，Intel Ultra 7 265K对比i7-14700KF 评测...

www.bilibili.com