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

Tigerick · Aug 22, 2022

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

Model	Code-Name	Date	TDP	Node	Tiles	Main Tile	CPU	LP E-Core	LLC	GPU	Xe-cores
Core Ultra 100U	Meteor Lake	Q4 2023	15 - 57 W	Intel 4 + N5 + N6	4	tCPU	2P + 8E	2	12 MB	Intel Graphics	4
?	Lunar Lake	Q4 2024	17 - 30 W	N3B + N6	2	CPU + GPU & IMC	4P + 4E	0	8 MB	Arc	8
?	Panther Lake	Q1 2026 ?	?	Intel 18A + N3E	3	CPU + MC	4P + 8E	4	?	Arc	12

Comparison of die size of Each Tile of Meteor Lake, Arrow Lake, Lunar Lake and Panther Lake

	Meteor Lake	Arrow Lake (20A)	Arrow Lake (N3B)	Arrow Lake Refresh (N3B)	Lunar Lake	Panther Lake
Platform	Mobile H/U Only	Desktop Only	Desktop & Mobile H&HX	Desktop Only	Mobile U Only	Mobile H
Process Node	Intel 4	Intel 20A	TSMC N3B	TSMC N3B	TSMC N3B	Intel 18A
Date	Q4 2023	Q1 2025 ?	Desktop-Q4-2024 H&HX-Q1-2025	Q4 2025 ?	Q4 2024	Q1 2026 ?
Full Die	6P + 8P	6P + 8E ?	8P + 16E	8P + 32E	4P + 4E	4P + 8E
LLC	24 MB	24 MB ?	36 MB ?	?	8 MB	?
tCPU	66.48
tGPU	44.45
SoC	96.77
IOE	44.45
Total	252.15

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.)

Oneto · Feb 15, 2024

This thread is so entertaining. It is. I have no idea what half of what is being said means but that doesn't matter I guess.

FlameTail · Feb 15, 2024

igor_kavinski said:
Damn shame that the Crater Lake laptop can't touch the Ryzen laptop even with a larger battery.

View attachment 93906
The first three tests the MTL SoC was supposed to excel at and even in the last test, the Ryzen laptop gives it a run for its money.

But don't worry, Intel. Your fans won't let you sweat. They will keep you nice and cool with their disposable dollars!

Now where is Apple M3😏

Markfw · Feb 15, 2024

igor_kavinski said:
Yeah my COO is just like that. Thinks of these people as servants for hire and ignores their suggestions. I keep wondering how long before we post a loss. Then it's gonna be him pointing fingers at everyone but himself.

Slightly OT, but the company I retired from, I told a senior VP about something that need to change for better customer service and his reply was "Its's beyond my pay grade to effect such a change". There was only 2 steps above him. That is sad. Thats exactly what happened at Intel, is what I think you are all agreeing to.

adroc_thurston · Feb 15, 2024

SiliconFly said:
MTL is foveros

In name only.

igor_kavinski said:
Damn shame that the Crater Lake laptop can't touch the Ryzen laptop even with a larger battery.

View attachment 93906
The first three tests the MTL SoC was supposed to excel at and even in the last test, the Ryzen laptop gives it a run for its money.

But don't worry, Intel. Your fans won't let you sweat. They will keep you nice and cool with their disposable dollars!

ooohh that's embarassing

Hitman928 · Feb 15, 2024

FlameTail said:
Now where is Apple M3😏

Top is H264 and middle is WiFi. Note that I don’t think the Mac uses an OLED screen but someone can correct me if I’m wrong on that.

Saylick · Feb 15, 2024

igor_kavinski said:
Yeah my COO is just like that. Thinks of these people as servants for hire and ignores their suggestions. I keep wondering how long before we post a loss. Then it's gonna be him pointing fingers at everyone but himself.

The worst part is that C-suite execs get a golden parachute if they get sacked.

I'm glad my company is employee-owned and the entire upper management, except for CFO and legal, are all engineers.

igor_kavinski · Feb 15, 2024

Markfw said:
Slightly OT, but the company I retired from, I told a senior VP about something that need to change for better customer service and his reply was "Its's beyond my pay grade to effect such a change". There was only 2 steps above him. That is sad. Thats exactly what happened at Intel, is what I think you are all agreeing to.

Yeah I think these people are so comfy with their skyhigh wages that they don't want to disturb the hornet's nest and would rather inflate their bank account in peace rather than risk trying to be loyal to the entity that pays them and then be seen as THE PROBLEM instead of the one IDENTIFYING the problem.

Funny thing happened last year to us. My entire department (operations) was looked over during the performance evaluations and no one got even a cent of an increment. Everyone around me was furious. It was decided with the department head that we would all go in unison to the top management and report our grievance because some other departments with good actors (as in good at drama) were able to secure some nice increments. So we were not sure why we were left out.

One day before the "march" to the COO's office was supposed to happen, our dept head goes to the COO and tells him the situation. Next day, the COO identifies some mistake by one of our members and rocks the entire office with his booming voice, chiding us, even those who had nothing to do with the mistake. And thus, it was no longer possible to "demand" an explanation since he would just bring this up as an example of how incompetent we were.

Our weasel of a dept head is enjoying his fat paycheck by the way

SiliconFly · Feb 15, 2024

adroc_thurston said:
In name only.

Oh yeah! Thats gotta to be it! Considering MTL's foveros packaging is far more advanced than LakeField.

adroc_thurston · Feb 15, 2024

SiliconFly said:
Oh yeah! Thats gotta to be it!

Yeah.

SiliconFly said:
Considering MTL's foveros packaging is far more advanced than LakeField.

Tighter bump pitch (36um) but the base is wholly passive slab (welcome from 2012).

H433x0n · Feb 15, 2024

Little bit of hopium fuel for ARL-S gaming performance not being a total trainwreck.

Seems to be roughly at par with HawkPoint here (except for MW3 & CyberPunk). It's also a lower clocking SKU, so maybe the 165H or 185H would close the gap a little bit.

Hitman928 · Feb 15, 2024

H433x0n said:
Little bit of hopium fuel for ARL-S gaming performance not being a total trainwreck.

View attachment 93912

View attachment 93913

Seems to be roughly at par with HawkPoint here (except for MW3 & CyberPunk). It's also a lower clocking SKU, so maybe the 165H or 185H would close the gap a little bit.

The MTL laptop dGPU is running at 23% higher power.

Edit: I’m not predicting anything about ARL-S with this comment, just noting that the MTL GPU being able to use more power blurs the ability to draw gaming performance conclusions from these results.

SiliconFly · Feb 15, 2024

adroc_thurston said:
...but the base is wholly passive slab (welcome from 2012).

And Zen5 uses routing wires in substrate! (welcome from 1980s)

(Yet again) To sum it up, ARL uses very advanced foveros packaging & NOC interconnect compared to Zen5's substrate & infinity fabric which ever way you see it.

And like I said, the final outcome of ARL vs Zen5 now depends solely on how LNC performs in real world. Not chiplets or packaging or infinity fabric, etc.

H433x0n · Feb 15, 2024

Hitman928 said:
The MTL laptop dGPU is running at 23% higher power.

It has a 23% higher power budget, not sure if that matters when running cpu limited at 1080p.

adroc_thurston · Feb 15, 2024

SiliconFly said:
And Zen5 uses routing wires in substrate!

They're very magical carriers overall.

SiliconFly said:
To sum it up, ARL uses very advanced foveros packaging

It's a passive 2.5D slab. nothing special or interesting.

SiliconFly said:
NOC interconnect

It's IOSF. From 2012.

Hitman928 · Feb 15, 2024

H433x0n said:
It has a 23% higher power budget, not sure if that matters when running cpu limited at 1080p.

It uses all 80 W running Witcher 3 at 1080p Ultra. How much it uses at lower settings isn't shown but at such low power budgets, it wouldn't surprise me if it still uses 80 W, or near 80 W, even at lower settings. Some games, like Avatar, will for sure use all 80 W, even at 1080p low.

Tigerick · Feb 15, 2024

Tigerick said:
OK, here comes some leaks not from me but credential leaker about ARL & PTL:

ARL-H (6P+8E) comes with iGPU (Build by newer process, N4P) 8 Xe1 LPG+ cores clocking at 2.3GHz

ARL-HX (8P+16E, H and HX's tCPU are different dies) comes with 4 Xe cores

SoC has been updated to support WiFi7, still on N6

PTL-H's tCPU integrates with 4P+8E+4LPe without HT, SoC and MC

Here comes interesting part, PTL's tGPU comes with 12 Xe3 LPG core clocking at 2.5GHz. Yeah it is GT3 with 7.7 TF, slightly higher than Strix Point. And they are built on Intel 3+ process. Finally, IFS build tGPU.

ADM so far is dead.

Compared to MTL-H's iGPU, below is the major changes in ARL's iGPU:

Doubled L2 cache.
Doubled RT throughput.
XMX cores added.

igor_kavinski · Feb 15, 2024

It's funny that Intel is reliably firing on all cylinders in...GRAPHICS!

From joke to fairly desirable product, it's been quite a journey for them and they ain't stoppin'!

adroc_thurston · Feb 15, 2024

igor_kavinski said:
It's funny that Intel is reliably firing on all cylinders in...GRAPHICS

No?
Their IP still sucks.

igor_kavinski said:
From joke to fairly desirable product, it's been quite a journey for them and they ain't stoppin'!

They're still not competitive on raw PPA metrics.

igor_kavinski · Feb 15, 2024

adroc_thurston said:
Their IP still sucks.

Is that envy? Remember, they are No.2 in upscaling (XeSS) and also in raytracing.

Plus, they put out a paper on enabling real time performant raytracing on iGPUs which could be a real gamechanger if there is virtually no performance hit on enabling RT.

Abwx · Feb 15, 2024

Hitman928 said:
The MTL laptop dGPU is running at 23% higher power.

Edit: I’m not predicting anything about ARL-S with this comment, just noting that the MTL GPU being able to use more power blurs the ability to draw gaming performance conclusions from these results.

Despite these 23% higher dGPU power gaming perfs are comparable, even the 3DMark score is the same within margin of error, so what is exactly this higher power used for is a question that has no logic answer at first glance, in principle it should had yielded 8-10% higher perf on average including in 3DMark.

gdansk · Feb 15, 2024

igor_kavinski said:
Is that envy? Remember, they are No.2 in upscaling (XeSS) and also in raytracing.

Don't get ahead of yourself. They're not yet #2 in raytracing. 406mm² N6 DG2-512 vs 204mm² N6 N33 and they're only 20% ahead in RT (at 50% higher power too). Gotta wait for BM before talking up anything about Intel graphics outside of their upscaling. Oh and I guess MTL is doing good catching up.

H433x0n · Feb 15, 2024

gdansk said:
Don't get ahead of yourself. They're not yet #2 in raytracing. 406mm² N6 DG2-512 vs 204mm² N6 N33 and they're only 20% ahead in RT (at 50% higher power too). Gotta wait for BM before talking up anything about Intel graphics outside of their upscaling. Oh and I guess MTL is doing good catching up.

No, they are. We're not talking about absolute perf/watt with RT. The RT performance of an architecture is the relative performance hit from raw raster. As illustrated in the charts below.

gdansk · Feb 15, 2024

H433x0n said:
No, they are. We're not talking about absolute perf/watt with RT. The RT performance of an architecture is the relative performance hit from raw raster. As illustrated in the charts below.

No, even if they made a reticle limit DG2-1024 at 450W they'd still be behind the 7900 XTX in raytracing performance. So-called architectural advantage does not exist if you can't build bigger. It doesn't exist so they're not number 2. Same thing with AMD saying one of the advantages of their multi-chip architecture is that they could build a supermassive 400mm + 8 MCD RDNA3 chip to match the 4090. OK cool, but they didn't so it's whiteboard cope & dreams and they stay #2.

To summarize: if your raster performance is 0.5x RDNA3 per area on iso process it doesn't matter that you have -23% RT performance and they have -45% you're still 0.7x RT performance per area on iso process. Their raster is just really, really bad.

adroc_thurston · Feb 15, 2024

igor_kavinski said:
they are No.2 in upscaling (XeSS)

Doesn't matter?

igor_kavinski said:
also in raytracing.

Doesn't matter?
But a 406mm^2 die being the perf of 204mm^2 at higher power sure does.

igor_kavinski said:
if there is virtually no performance hit on enabling RT.

that's never happening, RTRT is inherently pretty tough.

itsmydamnation · Feb 15, 2024

I feel like this thread needs the AMD MOAR CORES meme updated with but muh package.

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

Senior member

Attachments

Member

Platinum Member

Moderator Emeritus, Elite Member

Platinum Member

Diamond Member

Diamond Member

Lifer

Golden Member

Platinum Member

Senior member

Diamond Member

Golden Member

Senior member

Platinum Member

Diamond Member

Senior member

Lifer

Platinum Member

Lifer

Lifer

Platinum Member

Senior member

Attachments

Platinum Member

Platinum Member

Platinum Member