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.)

Philste · Nov 1, 2023

Henry swagger said:
skymont aiming for alder lake ipc and arm X1 core efficiency.. 1w per core ? 🤔💻✍️

eek2121 said:
I can see the IPC, though if they really are shooting for 1w/core it won’t be performing like Golden Cover or even Sunny Cove.

The efficiency claim is completely made up by Prakhar Verma. Dude is hyping Intel stuff on Twitter without understanding half of the information.

2 weeks ago, when Raichu first wrote about Skymont, kopite replied with "Aim at ADL", now Raichu backs it up with the Cortex X1 claim. Cortex X1 and Golden Cove are quite similar in Terms of IPC (Geekbench), Cortex obviously clocks much lower.

So yeah, since both Raichu and Kopite are kinda reliable, I would say we should assume Skymont is in the Cortex X1/Golden Cove IPC Region. The efficiency stuff is BS. With that in mind i would assume that Lion Cove indeed is a decent IPC uplift and Arrow Lakes mediocre Performance predictions are frequency related. Skymont and Lion Cove less than 10% apart in IPC would be really weird to see.

Henry swagger · Nov 1, 2023

Philste said:
The efficiency claim is completely made up by Prakhar Verma. Dude is hyping Intel stuff on Twitter without understanding half of the information.

2 weeks ago, when Raichu first wrote about Skymont, kopite replied with "Aim at ADL", now Raichu backs it up with the Cortex X1 claim. Cortex X1 and Golden Cove are quite similar in Terms of IPC (Geekbench), Cortex obviously clocks much lower.

So yeah, since both Raichu and Kopite are kinda reliable, I would say we should assume Skymont is in the Cortex X1/Golden Cove IPC Region. The efficiency stuff is BS. With that in mind i would assume that Lion Cove indeed is a decent IPC uplift and Arrow Lakes mediocre Performance predictions are frequency related. Skymont and Lion Cove less than 10% apart in IPC would be really weird to see.

Thats why they are cutting smt because skymont ipc is big enough to compensate

Henry swagger · Nov 1, 2023

eek2121 said:
I can see the IPC, though if they really are shooting for 1w/core it won’t be performing like Golden Cover or even Sunny Cove.

That is why I chuckle a bit when people say Intel can’t possibly compete. Add 5-10% IPC to the mont cores, a node shrink, and even more mont cores and suddenly the chip is 40-60% faster. 5-10% adds up when applied to 16 (now 18) cores. Shoot, even 1% adds up.

They can do it if they copy apple core desgn.. apple has golden cove ipc at low watts because of a wide decode and 600+ robs

igor_kavinski · Nov 1, 2023

Henry swagger said:
They can do it if they copy apple core desgn

You sound like you've been designing CPUs since you were a toddler. I want to hear more!

dullard · Nov 1, 2023

coercitiv said:
Oh yes, you showed me in red bold text just in case I was having problems reading charts. The reality is you got triggered and jumped to defend Intel even if I wasn't really attacking them. They provide less info on expected future performance/efficiency, that by itself has no clear meaning until we get to see the product. It could be a good thing. It could even be a temporary lapse, they may share more about ARL for example. But hey, you do you, find someone who offends your sense of technological justice and help set the record straight.

So, you want to see the product 6 weeks before the launch of the product? Seems like you need realistic expectations.

You were caught in a flat out lie about Intel being silent. Ne need to call be triggered just because I point that out. Feel free to edit your post to be correct, something along the lines that Intel isn't sharing the information you want to see.

mikk · Nov 1, 2023

With Tremont Intel introduced the 2x3 out of order decode which also Gracemont is using.

3x3 for Skymont is a beefy upgrade if true. Tremont gained over 30% IPC with this dual decoder (combined with all the other changes).

Goldmont Plus= 3-wide
Tremont= 2x3-wide
Gracemont= 2x3-wide
Skymont= 3x3-wide?

Saylick · Nov 1, 2023

Philste said:
The efficiency claim is completely made up by Prakhar Verma. Dude is hyping Intel stuff on Twitter without understanding half of the information.

Holy moly. That dude's profile is insane. Over 90k posts. I know we don't throw around the word fanboy lightly here but that guy is on another level. Almost every reply just hyping up Intel, spamming others with the same repeated pro-Intel claims, and/or blatantly making things up about MTL/ARL without any sources.

SiliconFly · Nov 1, 2023

Geddagod said:
Look at Zen 3 vs GLC. GLC is ~20% performant, ~ as efficient, but like 75% larger...

Equalize the process, Intel cores are still disproportionately larger...

GLC is around 75% larger than Zen 3. It's really sad. And I'm really really glad that they're ditching their stupid fat P cores starting next year in favor of LNC which I'm hoping will be a lot leaner, more performant and efficient and more in line with Zen cores.

Also, raichu seems to be very excited about the skymont E cores in ARL. He's hinting at huge E core perf gains. That should make ARL very competitive in MT benchmarks.

SiliconFly · Nov 1, 2023

Henry swagger said:
Thats why they are cutting smt because skymont ipc is big enough to compensate

I don't think they're cutting SMT. Since LNC is brand new, I think it still hasn't "gained" SMT. That would be my guess.

DrMrLordX · Nov 1, 2023

eek2121 said:
Add 5-10% IPC to the mont cores, a node shrink, and even more mont cores and suddenly the chip is 40-60% faster. 5-10% adds up when applied to 16 (now 18) cores. Shoot, even 1% adds up.

If those numbers were true, Arrow Lake-S would be vastly more-performant than Raptor Lake-S even with a common 8+16 config.

SiliconFly · Nov 1, 2023

DrMrLordX said:
If those numbers were true, Arrow Lake-S would be vastly more-performant than Raptor Lake-S even with a common 8+16 config.

Maybe in MT. But if ST sucks, then Intel still has a huge problem. Especially considering the fact that the competition now have vastly better ST cores (like apple, amd & even qualcomm).

AMDK11 · Nov 1, 2023

SiliconFly said:
Maybe in MT. But if ST sucks, then Intel still has a huge problem. Especially considering the fact that the competition now have vastly better ST cores (like apple, amd & even qualcomm).

Not if the IPC is much higher but the clock speed is lower.We'll see what it will look like in the final product.

Geddagod · Nov 1, 2023

Henry swagger said:
https://twitter.com/x/status/1719634325795443080
skymont aiming for alder lake ipc and arm X1 core efficiency.. 1w per core ? 🤔💻✍️

3, 3 wide decoders sounds pretty weird tbh. And one of the decoders is disabled for a total of 8 wide?
Maybe they want SKMT to target extremely low power so they could power gate to just 2 or 3 wide decoders if they want to, idk

Henry swagger · Nov 1, 2023

Geddagod said:
3, 3 wide decoders sounds pretty weird tbh. And one of the decoders is disabled for a total of 8 wide?
Maybe they want SKMT to target extremely low power so they could power gate to just 2 or 3 wide decoders if they want to, idk

Mayb 1 complex decode and 8 simple

JoeRambo · Nov 3, 2023

mikk said:
3x3 for Skymont is a beefy upgrade if true. Tremont gained over 30% IPC with this dual decoder (combined with all the other changes).

Goldmont Plus= 3-wide
Tremont= 2x3-wide
Gracemont= 2x3-wide
Skymont= 3x3-wide?

My first reaction was to doubt it, but after giving it some thought it is logical for Intel to do it. The reason is:
(source as always Chips and Cheese)

There is nowhere for Atom cores to go after 3+3, as they don't have uOP cache to help things, they are stuck with instructions decoded from 32 bytes. At first i thought that 8 wide from earlier "leaks" might mean obvious "expansion" to 4+4, but that does not solve much.

So they are preparing atom core for closing the gap with big cores -> introducing the 3rd cluster, having 3 fetch queues and chewing 36-48 bytes per clock from L1I

igor_kavinski · Nov 6, 2023

Intel Meteor Lake CPU delays leave OEMs 'royally pissed'

Intel's various delays with its new Meteor Lake CPU launch have left OEMs 'royally pissed' at the company, according to the latest leaks.

www.tweaktown.com

Despite this being MLID, I'm inclined to believe him for once. Intel is late and probably kept OEMs waiting and promising them new CPUs and told them at the last minute that they couldn't provide enough in time.

Geddagod · Nov 6, 2023

igor_kavinski said:
Intel Meteor Lake CPU delays leave OEMs 'royally pissed'

Intel's various delays with its new Meteor Lake CPU launch have left OEMs 'royally pissed' at the company, according to the latest leaks.

www.tweaktown.com

Despite this being MLID, I'm inclined to believe him for once. Intel is late and probably kept OEMs waiting and promising them new CPUs and told them at the last minute that they couldn't provide enough in time.

Doubt it. Intel couldn't keep OEMs waiting till the "last minute", literally them lacking shipments would have told them that they didn't have chips to sell lol. They would have known it for months before the actual launch of new products. Stuff like this is communicated before hand, especially since lack of which could prob easily leave them open to lawsuits. Besides, it's not like they don't have any CPUs at all to sell (like cough AMD cough), RPL should keep selling throughout the holiday season.

mikk · Nov 6, 2023

Strange that MLID never complains when AMD paper launches their mobile lineup. Some OEMs were not amused about the lack of Phoenix CPUs half a year after the launch.

trivik12 · Nov 6, 2023

igor_kavinski said:
Intel Meteor Lake CPU delays leave OEMs 'royally pissed'

Intel's various delays with its new Meteor Lake CPU launch have left OEMs 'royally pissed' at the company, according to the latest leaks.

www.tweaktown.com

Despite this being MLID, I'm inclined to believe him for once. Intel is late and probably kept OEMs waiting and promising them new CPUs and told them at the last minute that they couldn't provide enough in time.

MLID is garbage. Intel just in their earnings call have said they have shipped MTL to OEMs. Multiple OEMs have MTL laptops designed and ready to release. Let us see next month how many are launched on 12/14. Unlike a normal CPU launch, this seems way more high profile looking at the event. I am sure we will have major OEMs there launching them. Rest we will hear in CES.

Based on what I have seen in ADL/RPL, we generally see some laptops even in store in couple of months post release. Unlike say AMD with phoenix where after 3 quarters I have yet to see one in person. Good news is at least online availability of phoenix is good these days including great deals.

Ultimately customer is the winner when all these laptops are available and competing against not just between these 2 companies but with Apple and even Qualcomm next year. Its exciting times indeed.

jpiniero · Nov 7, 2023

Wouldn't be surprised if it's more that OEM employees don't like the date. If you can't make mid November, just delay the thing to CES.

SiliconFly · Nov 7, 2023

Well, what MLID forgot to mention was... once MTL launches next month, it's not going to have any real competition for a long time.

Remember, MLID himself mentioned that Strix has been pushed back to Q3 2023, which makes it a ARL competitor (not MTL).

Thats probably the reason Intel is being bullish & trying to price MTL higher than it should.

coercitiv · Nov 7, 2023

SiliconFly said:
Well, what MLID forgot to mention was... once MTL launches next month, it's not going to have any real competition for a long time.

If we suddenly care what MLID says... let's quote him.

With this leak we can conclude that Intel's Meteor Lake should, because of the delays, have a solid 1 to 3 months in the Sun as a premium product. But it does also seem like the cheaper to manufacture Hawk Point is going to be a bit stronger than we thought it might be, that AMD has a mobility lineup ready for the 2024 back-to-school season, while Intel likely will have large gaps in mobility until 2026.

Personally I would be far more interested in the absolute attributes of MTL rather than relative positioning versus AMD's product stack. Let's see the massive improvement in battery life, the strong MT improvement due to higher base clocks, let's see the GPU come to life with Intel's mature driver stack, let's see it come together as a well rounded tiled design to justify the price premium. After that we can start looking around for any challenger.

DrMrLordX · Nov 7, 2023

igor_kavinski said:
Despite this being MLID, I'm inclined to believe him for once. Intel is late and probably kept OEMs waiting and promising them new CPUs and told them at the last minute that they couldn't provide enough in time.

Is this news at this point? Wasn't it a given that actual availability of Meteor Lake in Q4 2023 would be nearly non-existent and very expensive?

DrMrLordX · Nov 7, 2023

trivik12 said:
Unlike say AMD with phoenix where after 3 quarters I have yet to see one in person. Good news is at least online availability of phoenix is good these days including great deals.

It's been like that pretty much through the entire Zen era. AMD ships some mobile SoCs but you never really know where they're going. Availability at retail is bad for months on end.

Mopetar · Nov 7, 2023

DrMrLordX said:
It's been like that pretty much through the entire Zen era. AMD ships some mobile SoCs but you never really know where they're going. Availability at retail is bad for months on end.

You can find something, but not really the particular model you're looking for, or if you can it's in some absurd product that doesn't make a lot of sense. There might be one particular product that actually makes sense, but they only made 500 units and it has a 3 month waitlist before your order gets canceled because they aren't making more.

Intel generally had better availability and even though you can find a fair share of bizarre products with an Intel Inside (TM) there's usually enough of a selection such that there's something you might actually want.

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

Senior member

Attachments

Member

Senior member

Senior member

Lifer

Elite Member

Diamond Member

Diamond Member

Golden Member

Golden Member

Lifer

Golden Member

Senior member

Golden Member

Senior member

Golden Member

Lifer

Golden Member

Diamond Member

Senior member

Lifer

Golden Member

Diamond Member

Lifer

Lifer

Diamond Member