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

[Tigerick]

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	12 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)	Lunar Lake	Panther Lake
Platform	Mobile H/U Only	Desktop Only	Desktop & Mobile H&HX	Mobile U Only	Mobile H
Process Node	Intel 4	Intel 20A	TSMC N3B	TSMC N3B	Intel 18A
Date	Q4 2023	Q1 2025 ?	Desktop-Q4-2024 H&HX-Q1-2025	Q4 2024	Q1 2026 ?
Full Die	6P + 8P	6P + 8E ?	8P + 16E	4P + 4E	4P + 8E
LLC	24 MB	24 MB ?	36 MB ?	12 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.)

 

[Josh128]

P All Core 5.4

I can believe that...tell us more!

 

[Josh128]

Thanks. People still don't seem to realize that HT has drawbacks. One of which is the extra transistor flipping and data movement uses more power and creates more heat. Meaning HT limits clock rates (not a lot, but some). And the efficiency improvements come in-part by no longer having to do the logic checks to keep threads apart on the same core for security reasons (you don't want one thread to be able to access even the cache from another thread).

Obviously higher performance requires higher power consumption. HT more fully utilizes a cores resources. No performance is "free".

 

[dullard]

Obviously higher performance requires higher power consumption. HT more fully utilizes a cores resources. No performance is "free".

Apparently that isn't obvious to all people.

 

[9949asd]

OMG cudimm xmp 9600！ running at gear2 confirmed!

 

[Henry swagger]

OMG cudimm xmp 9600！ running at gear2 confirmed!View attachment 107469

The cores will be well feed woth data.. gaming crown will be taken with ease

 

[9949asd]

The cores will be well feed woth data.. gaming crown will be taken with ease

Z890 APEX will have high chance run 10000+.

 

[Josh128]

OMG cudimm xmp 9600！ running at gear2 confirmed!View attachment 107469

1.5V?? Yikes! XMP-9600 on its own is impressive though.

 

[Hulk]

P All Core 5.4

I was thinking the same. It may be that 5.6 will only be attainable under very low load/low temp conditions all-core? Like under 70C or something?

 

[511]

ultra 5?

https://twitter.com/x/status/1834630852917240195

 

[cebri1]

ultra 5?

https://twitter.com/x/status/1834630852917240195

ugh. Exactly the same results as the 14600K.

edit:Actually 8-10% faster in multi, the 14600K hovers around 10K. Skymont covering for 6 threads less and more.

 

[9949asd]

ugh. Exactly the same results as the 14600K.

edit:Actually 8-10% faster in multi, the 14600K hovers around 10K. Skymont covering for 6 threads less and more.

The leaker said bios still has some problems, the frequently is 0.1-0.2g lower, and the power should be a bit lower.

 

[cebri1]

The leaker said bios still has some problems, the frequently is 0.1-0.2g lower, and the power should be a bit lower.

Even then, the result is still quite low compared to expectations.

4% ST above 14600K with only a minimal clock regression doesn't align with IPC gains.

 

[511]

Guess the reason they are delayed but they can't be delayed cause they will be on time of Q4

 

[Hail The Brain Slug]

OMG cudimm xmp 9600！ running at gear2 confirmed!View attachment 107469

I'm cudimming at the very thought!!!

 

[9949asd]

Even then, the result is still quite low compared to expectations.

4% ST above 14600K with only a minimal clock regression doesn't align with IPC gains.

ARL will not show real performance on cpuz now just like zen. On r23 r24 spec will show the correct uplift.

 

[Wolverine2349]

ARL will not should real performance on cpuz now just like zen. On r23 r24 spec will show the correct uplift.

Yeah may not. But Zen 5 is a flop in uplift with almost no gain in gaming and other consumer workloads that do not use AVX512. Will Arrow Lake actually have a meaningful gain over RPL in gaming and other general consumer workloads even if only 10-15%. I really want to see that plus much lower power for cooler running CPUs and oh stability stability stability which Raptor Lake lacks.

Raptor Lake may even lack stability and still degrade too easily even with corrected microcode and default Intel power profile updates. It was a rushed design and flawed afterall.

 

[9949asd]

Yeah may not. But Zen 5 is a flop in uplift with almost no gain in gaming and other consumer workloads that do not use AVX512. Will Arrow Lake actually have a meaningful gain over RPL in gaming and other general consumer workloads even if only 10-15%. I really want to see that plus much lower power for cooler running CPUs and oh stability stability stability which Raptor Lake lacks.

Raptor Lake may even lack stability and still degrade too easily even with corrected microcode and default Intel power profile updates. It was a rushed design and flawed afterall.

Zen5 architecture is more focus on servers workload, that’s normal. Servers cpu are way more profitable🤪 who cares about MSDT 😂

 

[Wolverine2349]

Zen5 architecture is more focus on servers workload, that’s normal. Servers cpu are way more profitable🤪 who cares about MSDT 😂

Yes it is, but AMD using the exact same stepping for everything and just cramming a bunch of the same 8 core CCDs into one package for server EPYC parts. And lying to consumers about performance uplift when there was almost 0.

AMD is just too cheap to do anything different. They could have separate dies for consumer and server but nope.

At least Intel kind of always has had separate dies for consumer and server parts cause they ae not so cheap like AMD. So they can optimize al their mobile and desktop Arrow Lake/Lunar Lake parts for consumer workloads while focusing on a totally different granite Rapids/Clearwater forest and Sierra Forrest Xeon dies for server/enterprise workloads and not constrained by the same problem of using one due for all like AMD.

AMD has gotten so greedy when they have gained space in server market they are not innovating like they did when they had to comeback and attract desktop first which lead to the wonderous progress through Zen 3 and lesser extent Zen 4 on desktop space.

 

[Josh128]

Yes it is, but AMD using the exact same stepping for everything and just cramming a bunch of the same 8 core CCDs into one package for server EPYC parts. And lying to consumers about performance uplift when there was almost 0.

AMD is just too cheap to do anything different. They could have separate dies for consumer and server but nope.

At least Intel kind of always has had separate dies for consumer and server parts cause they ae not so cheap like AMD. So they can optimize al their mobile and desktop Arrow Lake/Lunar Lake parts for consumer workloads while focusing on a totally different granite Rapids/Clearwater forest and Sierra Forrest Xeon dies for server/enterprise workloads and not constrained by the same problem of using one due for all like AMD.

AMD has gotten so greedy when they have gained space in server market they are not innovating like they did when they had to comeback and attract desktop first which lead to the wonderous progress through Zen 3 and lesser extent Zen 4 on desktop space.

I dont think its that they're cheap, I think its that desktop is too low margin to create a new design for. Look whats getting the 3nm / 16 core CCX this gen: EPYC. Look whats getting 40 CUs of RDNA3.5 and what looks to be new uncore/+ultra low power cores? Strix Halo/ Laptop.

Desktop is like the redheaded stepchild of the bunch. In my opinion, AMD had to know the asking prices were too high for desktop Zen 5 but priced them that way anyway as they are already so much less profitable than EPYC. Its pretty sad, but it'll pay off in the long run.

 

[Markfw]

Yes it is, but AMD using the exact same stepping for everything and just cramming a bunch of the same 8 core CCDs into one package for server EPYC parts. And lying to consumers about performance uplift when there was almost 0.

AMD is just too cheap to do anything different. They could have separate dies for consumer and server but nope.

At least Intel kind of always has had separate dies for consumer and server parts cause they ae not so cheap like AMD. So they can optimize al their mobile and desktop Arrow Lake/Lunar Lake parts for consumer workloads while focusing on a totally different granite Rapids/Clearwater forest and Sierra Forrest Xeon dies for server/enterprise workloads and not constrained by the same problem of using one due for all like AMD.

AMD has gotten so greedy when they have gained space in server market they are not innovating like they did when they had to comeback and attract desktop first which lead to the wonderous progress through Zen 3 and lesser extent Zen 4 on desktop space.

Not to derail this thread, but you brought this up. Zen 4 and 5 are actually giant leaps in performance. EXCEPT for gaming. Those who live for gaming obviously will not see the value. But first, as mentioned, servers are immensely more profitable than desktop. To the point of possibly funding the rest of the cpu stack. And look at Intels financial situation. They are having trouble since they are losing market share in servers. And desktop, instead of innovating they are simply trying to overclock what they have to the point of instability and degrading.

Do you want a company that is financially strong with good stability ? Or just a good gaming processor ? If I was an investor, I pick number one.

 

[Wolverine2349]

Not to derail this thread, but you brought this up. Zen 4 and 5 are actually giant leaps in performance. EXCEPT for gaming. Those who live for gaming obviously will not see the value. But first, as mentioned, servers are immensely more profitable than desktop. To the point of possibly funding the rest of the cpu stack. And look at Intels financial situation. They are having trouble since they are losing market share in servers. And desktop, instead of innovating they are simply trying to overclock what they have to the point of instability and degrading.

Do you want a company that is financially strong with good stability ? Or just a good gaming processor ? If I was an investor, I pick number one.

Why not both. Intel well before their financial troubles proved both could be done. They had separate dies for lots.

AMD has just been using one die and nothing else.

 

[Wolverine2349]

I dont think its that they're cheap, I think its that desktop is too low margin to create a new design for. Look whats getting the 3nm / 16 core CCX this gen: EPYC. Look whats getting 40 CUs of RDNA3.5 and what looks to be new uncore/+ultra low power cores? Strix Halo/ Laptop.

Desktop is like the redheaded stepchild of the bunch. In my opinion, AMD had to know the asking prices were too high for desktop Zen 5 but priced them that way anyway as they are already so much less profitable than EPYC. Its pretty sad, but it'll pay off in the long run.

Well Intel has separate dies for both and they did well before their current financial troubles despite desktop being too low of a margin to care.

I think AMD is being cheap as such.

 

[gdansk]

Why not both. Intel well before their financial troubles proved both could be done. They had separate dies for lots.

AMD has just been using one die and nothing else.

Doesn't one wonder if having dozens of "chips" in-flight simultaneously perhaps contributed toward their financial problems? Maybe not.

 

[Wolverine2349]

Doesn't one wonder if having dozens of "chips" in-flight simultaneously perhaps contributed toward their financial problems? Maybe not.

They have been like that for more than a decade. Only recently have financial problems came up and largely due to Raptor Lake degradation/stability issues.

 

[gdansk]

They have been like that for more than a decade. Only recently have financial problems came up and largely due to Raptor Lake degradation/stability issues.

Since about Broadwell it hasn't been working right. That's almost a decade now.

I don't think making a gaming specific core is a good financial decision for anyone. X3D with a standard core is as close as you'll get and it's the side effect of a oddly-specific server part.