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.

 

[511]

It wasn't a political decision, Intel validation teams just plain suck.
And SMT is extensive in terms of validation man-hours.

Their validation teams had soo many layoffs xDd

 

[adroc_thurston]

Their validation teams had soo many layoffs xDd

Yeah but like the SPR debacle was Intel validation teams being poopoo peepee.
Awful.

 

[511]

Yeah but like the SPR debacle was Intel validation teams being poopoo peepee.
Awful.

Intel's validation culture was badfor the last 15-20 years they would just tape out a new silicon with their fab advantage.

 

[igor_kavinski]

they would just tape out a new silicon with their fab advantage.

To keep their fab people occupied with something to do because they had no external customers to stay busy.

 

[DrMrLordX]

Yeah but like the SPR debacle was Intel validation teams being poopoo peepee.
Awful.

Didn't Intel fire a bunch of people from their validation team, leading to (among other things) the Sapphire Rapids debacle?

 

[511]

To keep their fab people occupied with something to do because they had no external customers to stay busy.

NEw tape out is different things lol it barely fills just few wafers.

Didn't Intel fire a bunch of people from their validation team, leading to (among other things) the Sapphire Rapids debacle?

They fired the whole team not bunch of people from the team.

 

[igor_kavinski]

NEw tape out is different things lol it barely fills just few wafers.

I'm not talking about ramping up production capacity but about gaining more experience with the process. More silicon to tape out means you get to try some new refinement alongside the silicon bugfixes to improve the yield or optimize the v/f curve for example.

 

[511]

I'm not talking about ramping up production capacity but about gaining more experience with the process. More silicon to tape out means you get to try some new refinement alongside the silicon bugfixes to improve the yield or optimize the v/f curve for example.

No ? It was the same freaking silicon you need volume for improving Yield and Tape out is not that

 

[LightningZ71]

and Intel is capable of shipping laptop parts yearly.

That's... Well there were certainly parts shipped. If we're not talking down powered desktop parts, ice lake was an overheating mess. Tiger Lake was what Ice Lake should have been and really wasn't great in it's 8 core variant. Alder Lake P and below were actually decent, but then raptor lake mobile... Was just Alder Lake again. Meteor Lake regressed in multiple areas, but it sure idled well. Are we counting Lunar Lake? That's not much volume. Arrow Lake is just... Meteor Lake again, or it's a desktop chiplet with a different iGPU or IoD.

So, yes. They are shipping... Something each year, barely in the case of meteor.

Don't get me wrong. AMD certainly is no better. Prior to Renoir they weren't really serious. Rembrandt barely existed. Phoenix/Hawk Point was their first truly competitive part. Strix/Kraken needs to carry them for years, maybe with a refresh.

 

[511]

Tiger Lake was what Ice Lake should have been and really wasn't great in it's 8 core variant. Alder Lake P and below were actually decent, but then raptor lake mobile... Was just Alder Lake again. Meteor Lake regressed in multiple areas, but it sure idled well. Are we counting Lunar Lake? That's not much volume. Arrow Lake is just... Meteor Lake again, or it's a desktop chiplet with a different iGPU or IoD.

Meteor Lake has better battery under load as well vs ADL. LNL is shipping in Volume. Arrow Lake is clear improvement over Meteor Lake in everything.

 

[regen1]

Meteor Lake has better battery under load as well vs ADL. LNL is shipping in Volume. Arrow Lake is clear improvement over Meteor Lake in everything.

Yeah
MTL got upgrades for battery life, IGPU, memory support, media engine, I/O.
RWC actually has tiny IPC uplift but the resultant ST performance vs some previous Gen SKUs is negated by lower clocks, increased L3 latency.
ARL-H is a good upgrade over MTL-H in ST, MT, memory support, IGPU perf(+Full-rate RT, XMX), battery life, I/O (integrated WiFi 7).
Also HX line got updated with ARL.

 

[Khato]

Didn't Intel fire a bunch of people from their validation team, leading to (among other things) the Sapphire Rapids debacle?

Correct. More specifically they fired the lower level pre-silicon validation teams on the idiotic notion of some executive that all pre-silicon validation could be done by the full chip team. As anyone with even the slightest clue could easily have guessed, the result was not good.

I'll agree that historicallyl Intel's CPU validation teams haven't done as well at keeping pace with industry trends. But as is par for the course with Intel, the major fumbles tend to be the result of executive stupidity, not engineer incompetence.

 

[511]

I'll agree that historicallyl Intel's CPU validation teams haven't done as well at keeping pace with industry trends. But as is par for the course with Intel, the major fumbles tend to be the result of executive stupidity, not engineer incompetence

It is these clueless executives that are the core problem they just make stupid decisions.

 

[igor_kavinski]

It is these clueless executives that are the core problem they just make stupid decisions.

Their supreme leader was Pat

 

[511]

Their supreme leader was Pat

You should get your fact checked cause he was against the infinite stepping and hot lots

 

[igor_kavinski]

You should get your fact checked cause he was against the infinite stepping and hot lots

And that's how Intel lost BILLIONS!

Doesn't matter what he was for or against. What matters in the end is actions he took and decisions he made. And almost everyone agrees that he's the most unexpected brilliant idiot CEO Intel has had. He was supposed to be the chosen one.

We have an Intel engineer telling MLID that cancelling Royal Core was probably a big mistake. Who made that decision?

 

[regen1]

Their supreme leader was Pat

Nope. Pat was good.

And that's how Intel lost BILLIONS!

Intel lost those mostly cause those before him were more into buybacks than do R&D, invest in tools, etc. Had to spend a lot on tools and R&D, a lot of which should have been been done in BK or Swan era.
Doesn't mean he got everything right, but he understood Intel and the path ahead.

 

[regen1]

We have an Intel engineer telling MLID that cancelling Royal Core was probably a big mistake. Who made that decision?

Many believe cancelling Royal Core was good. Intel still might have lot of IP related to it.
MLID doesn't even know the timeline Royal Core, he thinks it was supposed to come with Nova Lake.
Intel's major revenue = Notebooks and Server chips, much bigger share than DIY.
What if Royal Core with poor area-efficiency was not that suitable for either notebooks and servers ?
Unified core would probably be lot better than Royal Core for those market aspects.

 

[regen1]

New PantherLake steppings

https://twitter.com/x/status/1943301390421422464

https://review.coreboot.org/c/coreboot/+/88130/3/src/include/cpu/intel/cpu_ids.h

 

[coercitiv]

We have an Intel engineer telling MLID that cancelling Royal Core was probably a big mistake. Who made that decision?

Probably MLID.

 

[511]

We have an Intel engineer telling MLID that cancelling Royal Core was probably a big mistake. Who made that decision?

Bruh using MLID as a source means the Intel engineer was a vaporware.

 

[LightningZ71]

Yeah
MTL got upgrades for battery life, IGPU, memory support, media engine, I/O.
RWC actually has tiny IPC uplift but the resultant ST performance vs some previous Gen SKUs is negated by lower clocks, increased L3 latency.
ARL-H is a good upgrade over MTL-H in ST, MT, memory support, IGPU perf(+Full-rate RT, XMX), battery life, I/O (integrated WiFi 7).
Also HX line got updated with ARL.

I was specific for lower power mobile. Arrow Lake U IS just Meteor Lake with the CCD on Intel 3 now. It's slightly faster, but the uncore and supporting chiplet haven't changed at all.

Arrow Lake H is the only decent one of the lot. HX is the desktop setup with the LP cores enabled and mobile firmware.

For most mobile users on the business world, MTL improved battery life without making much of a performance improvement. Since most of them sit docked all day, it's not any different for them than ADL was. The improvement in media engine and iGPU performance is lost on them.

 

[igor_kavinski]

Nope. Pat was good.

Intel lost those mostly cause those before him were more into buybacks than do R&D, invest in tools, etc. Had to spend a lot on tools and R&D, a lot of which should have been been done in BK or Swan era.
Doesn't mean he got everything right, but he understood Intel and the path ahead.

This is getting really old and tired. So if the current CEO fails, we can simply say that Pat left Intel in such a sad state that LBT couldn't do anything better even if he wanted to.

And if the current CEO succeeds, people will say, oh, it was all Pat's strategy. LBT simply executed it blindly without changing a thing!

Yeah right. Probably why Pat was fired because the only thing he couldn't do was execute. You know, something good CEOs are supposed to do?

And I personally think (and outright accuse) Pat sympathizers of benefiting from the massive stock price damage Pat did. They feel grateful to him for letting them buy enough shares on the cheap for comfortable retirement.

 

[regen1]

I was specific for lower power mobile. Arrow Lake U IS just Meteor Lake with the CCD on Intel 3 now. It's slightly faster, but the uncore and supporting chiplet haven't changed at all.

Arrow Lake H is the only decent one of the lot. HX is the desktop setup with the LP cores enabled and mobile firmware.

ARL-H is the main volume part of Arrowlake on laptops and it has enough battery endurance to compete with ARL-U plus has lot more OEM designs and volume.

For most mobile users on the business world, MTL improved battery life without making much of a performance improvement. Since most of them sit docked all day, it's not any different for them than ADL was. The improvement in media engine and iGPU performance is lost on them.

We can pick whatever way but the improvements are to be done and welcome by other user segments. By that line of thinking for many casual users an 8th-10th gen processor laptop with an SSD would suffice.

 

[regen1]

This is getting really old and tired. So if the current CEO fails, we can simply say that Pat left Intel in such a sad state that LBT couldn't do anything better even if he wanted to.

And if the current CEO succeeds, people will say, oh, it was all Pat's strategy. LBT simply executed it blindly without changing a thing!

Yeah right. Probably why Pat was fired because the only thing he couldn't do was execute. You know, something good CEOs are supposed to do?

And I personally think (and outright accuse) Pat sympathizers of benefiting from the massive stock price damage Pat did. They feel grateful to him for letting them buy enough shares on the cheap for comfortable retirement.

Sure then BK, Frank Yeary, Swan must be the Intel greats for you !! ?
Looks like you have emotional damage with Pat, brother. Please see beyond some stupid conspiracy and stock nonsense.
May be try see how much damage was done by buyback and less investment in R&D among numerous other stuff and missing many boats.
Anyway I don't want to waste further time on these when you bring MLID's nonsense about Royal Core, who doesn't even know its details and timeline.