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

DKR · Jan 8, 2025

gdansk said:
Which reviewer config? CB 2024 1T shows 140+ in most reviews. Based on this chart I'm supposed to expect 210?

https://twitter.com/x/status/1876454826017743069

dullard · Jan 8, 2025

Joe NYC said:
Maybe in some spiritual form, but Gelsinger reiterated that Lunar Lake is a one off, with no direct successor.

He said memory on the package was a one-off, not that the concept of Lunar Lake type chip (low power, energy efficient, 4P+4E, aimed for tasks that are more GPU and NPU heavy than multi-tasking CPU heavy) was a one-off. Here is his one-off quote for context:

"That will not be the case with Panther Lake, Nova Lake, and its successors as well. We’ll build it in a more traditional way with the memory off package in the CPU, GPU, NPU, and I/O [input/output] capabilities in the package. But volume memory will be off-package in the road map going forward."

And the analysis of the whole Gelsinger discussion by those in the room with him: "However, Gelsinger went on to say that the complex integration of on-package memory affects Intel’s profit margin too much, and it appears that the company isn’t going to repeat this design in Panther Lake and Nova Lake CPUs. Although both are considered direct successors to Lunar Lake, they won’t feature on-package memory." https://finance.yahoo.com/news/intel-ceo-says-lunar-lake-125926535.html

As in, there are direct Lunar Lake successors, such as Panther Lake. But, the memory will not be the package.

del42sa · Jan 8, 2025

https://www.tomshardware.com/pc-com...celestial-xe-gpu-cores-power-intel-sneak-peek

Intel's CES 2025 exhibition features a stack of computers from the near future; six Panther Lake and Celestial-based systems. The next-generation CPU architecture and next-next-gen Celestial Xe GPU cores arrived at the show with little fanfare, receiving little mention in Intel's CES press release or keynote address.

Joe NYC · Jan 8, 2025

dullard said:
He said memory on the package was a one-off, not that the concept of Lunar Lake type chip (low power, energy efficient, 4P+4E, aimed for tasks that are more GPU and NPU heavy than multi-tasking CPU heavy) was a one-off. Here is his one-off quote for context:

"That will not be the case with Panther Lake, Nova Lake, and its successors as well. We’ll build it in a more traditional way with the memory off package in the CPU, GPU, NPU, and I/O [input/output] capabilities in the package. But volume memory will be off-package in the road map going forward."

And the analysis of the whole Gelsinger discussion by those in the room with him: "However, Gelsinger went on to say that the complex integration of on-package memory affects Intel’s profit margin too much, and it appears that the company isn’t going to repeat this design in Panther Lake and Nova Lake CPUs. Although both are considered direct successors to Lunar Lake, they won’t feature on-package memory." https://finance.yahoo.com/news/intel-ceo-says-lunar-lake-125926535.html

As in, there are direct Lunar Lake successors, such as Panther Lake. But, the memory will not be the package.

That's right. But Memory on Package was the unique part of Lunar Lake, and other properties, such as arrangement of tiles is less unique.

I think Lunar Lake is relatively successful in market place, and that's really what Intel wants Panther Lake to be a successor to.

Philste · Jan 8, 2025

Which of the Panther Lake Designs would report 12 Cores/Threads? I thought P/H is 4+8+4 and U is 4+0+4?

DavidC1 · Jan 8, 2025

DKR said:
https://twitter.com/x/status/1876454826017743069

Ugh. What a disaster.

It seems their dishonesty is getting worse and worse. Are they not aware most of the review sites aren't using the pre-launch BIOS? They have to be THAT selective about it?

So they said "let's compare against the non-functional BIOS" that made it perform like a Skylake core CPU?

dullard · Jan 8, 2025

Joe NYC said:
That's right. But Memory on Package was the unique part of Lunar Lake, and other properties, such as arrangement of tiles is less unique.

I think Lunar Lake is relatively successful in market place, and that's really what Intel wants Panther Lake to be a successor to.

The memory on package was unique, interesting academically, and shows Intel's packaging capabilities. But, since it did not translate much at all to performance, that is about it. Lunar Lake has so, so much more than the memory chip location.

Lunar Lake's focus is on efficiency first, not core spamming. Guess what, so is Panther Lake, especially Panther Lake-U.
Lunar Lake prioritizes the E cores over the P cores. Not only were the number of P cores capped at 4 (the top Meteor Lake and Arrow Lake mobile chips have 6 P cores) but the E cores getting the workload first. Same with Panther Lake--which might have up to 12 E+LPE cores with P cores capped at 4.
Lunar Lake featured their highest end GPU (Battlemage was on Lunar Lake first before being a graphics card). Panther will do the same with being the introduction to their Celestial GPU.
Lunar Lake is NPU heavy compared to their other mobile chips. Panther will be too.
Lunar Lake is for ultralight with long battery life. Same with Panther (at least Panther-U).
Etc.

To say that Panther is not a successor simply because one thing was removed is like saying Windows 11 and 12 aren't successors to Windows 7, 8, and 10 because Internet Explorer was removed.

mikk · Jan 8, 2025

Philste said:
Which of the Panther Lake Designs would report 12 Cores/Threads? I thought P/H is 4+8+4 and U is 4+0+4?

Maybe not fully enabled or task manager can't see the LPE cores.

gdansk · Jan 8, 2025

dullard said:
The memory on package was unique, interesting academically, and shows Intel's packaging capabilities. But, since it did not translate much at all to performance, that is about it. Lunar Lake has so, so much more than the memory chip location.

Lunar Lake's focus is on efficiency first, not core spamming. Guess what, so is Panther Lake, especially Panther Lake-U.

Lunar Lake prioritizes the E cores over the P cores. Not only were the number of P cores capped at 4 (the top Meteor Lake and Arrow Lake mobile chips have 6 P cores) but the E cores getting the workload first. Same with Panther Lake--which might have up to 12 E+LPE cores with P cores capped at 4.

Lunar Lake featured their highest end GPU (Battlemage was on Lunar Lake first before being a graphics card). Panther will do the same with being the introduction to their Celestial GPU.

Lunar Lake is NPU heavy compared to their other mobile chips. Panther will be too.

Lunar Lake is for ultralight with long battery life. Same with Panther (at least Panther-U).

Etc.

To say that Panther is not a successor simply because one thing was removed is like saying Windows 11 and 12 aren't successors to Windows 7, 8, and 10 because Internet Explorer was removed.

I feel like all this successor talk may be a waste of time. By these criteria isn't also the Apple M4 a successor to Lunar Lake? It showed up at another company before it and with completely different DNA but it meets all the criteria and even has MoP too!

dullard · Jan 8, 2025

gdansk said:
I feel like all this successor talk may be a waste of time. By these criteria isn't also the Apple M4 a successor to Lunar Lake? It showed up at another company before it and with completely different DNA but it meets all the criteria and even has MoP too!

One company does not make successors to another company's products. The Ford F-150 will never be a successor to a Dodge Ram regardless of features.

To me, it comes down to months and months and months of people arguing here that Intel's successes in Lunar Lake products will never be repeated. I highly doubt that moving a memory chip to a different location will end Intel's ability to sell CPUs in similar products with similar features.

gdansk · Jan 8, 2025

dullard said:
One company does not make successors to another company's products. The Ford F-150 will never be a successor to a Dodge Ram regardless of features.

On that I have to disagree. Intel's 486 has successors other than the Pentium that weren't made by Intel and are more similar to 486 than the Pentium was. That they were inferior to Intel's chosen successor is maybe another matter.

dullard said:
I highly doubt that moving a memory chip to a different location will end Intel's ability to sell CPUs in similar products with similar features.

That's true they can hit a similar mark in different ways.

dullard · Jan 8, 2025

gdansk said:
On that I have to disagree. Intel's 486 has successors other than the Pentium that weren't made by Intel and are more similar to 486 than the Pentium was. That they were inferior to Intel's chosen successor is maybe another matter.

That is the definition of a competitor, not a successor. Lets just think how that commercial would go: "Introducing the Reimagined 2026 Ford Dodge Ram 1500 this time Made by Ford!" Nope, never heard such a commercial in my life. Because that isn't how we define successors.

gdansk · Jan 8, 2025

dullard said:
That is the definition of a competitor, not a successor. Lets just think how that commercial would go: Introducing the new 2026 Ford Dodge Ram 1500 this time Made by Ford! Nope, never heard such a commercial. Because that isn't how we do successors.

Yeah, that absurdity happened with the 486. Because they can't trademark numbers.

I threw time out the window to be as inane as possible but the part that actually happened shouldn't be that odd.

dullard · Jan 8, 2025

gdansk said:
Yeah, that absurdity happened with the 486. Because they can't trademark numbers.

I threw time out the window to be as inane as possible but the part that actually happened shouldn't be that odd.

Those still are not successors. They are competing me-too products.

gdansk · Jan 8, 2025

dullard said:
Those still are not successors. They are competing me-too products.

Being an improved 486 does make them competing me-too products. But they carried 486 to new levels, lower power and even static variants. An improved derivative
has a good claim to being called a successor.

Much like Lunar Lake is an aping of M1 and PTL-U is not but will end up being better than LNL and worse than M4 anyway. Some times inspiration gives away half their intent. And there is no intent to compete but ship good enough at lowest possible BoM.

dullard · Jan 8, 2025

gdansk said:
Being an improved 486 does make them competing me-too products.

Oh, I see you are inventing your own definitions to words now. https://dictionary.cambridge.org/us/dictionary/english/me-too

A company's me-too product is one that is designed to be similar to a very popular product made by another company.

If you are a me-too product with advantages, there is a case for that too. https://radiusinsights.com/blog/should-you-develop-a-me-too-product/

Perhaps the way to look at it is not so much as a “me too,” but a “me too” with advantages.” If you can hit on those advantages, then you can justify moving forward.

No reason to keep this line of discussion going on in this thread. Go ahead and post your rebuttal. We'll just laugh at your incorrect usage of the English language.

gdansk · Jan 8, 2025

dullard said:
No reason to keep this line of discussion going on in this thread. Go ahead and post your rebuttal. We'll just laugh at your incorrect usage of the English language.

Here I'll get it back on topic for you. LNL has no derivative. It has a successor. But it's as much of a successor as the AM5x86 is to the Intel 486. It's another product hoping to hit the same market. But this time by the same company.

ajsdkflsdjfio · Jan 8, 2025

gdansk said:
Here I'll get it back on topic for you. LNL has no derivative. It has a successor. But it's as much of a successor as the AM5x86 is to the Intel 486. It's another product hoping to hit the same market. But this time by the same company.

FWIW, I'd say Panther lake is a lot closer to Lunar than anything else Intel has released so far in the mobile segment (raptor, arrow, meteor). The rumored PL1 and PL2 power (https://www.windowscentral.com/hardware/cpu-gpu-components/what-is-intel-panther-lake) also seems to line up with Lunar lake somewhat, although it likely won't have a 17W variant minus the 4 + 0 + 4 + 4XE3 variant.
Source: https://www.techpowerup.com/324540/...-cores-12-xe3-gpu-cores-and-five-tile-package

"The processor incorporates five distinct tiles, with three playing active roles in its functionality. The central compute operations are handled by one "Die 4" tile with CPU and NPU, while "Die 1" is dedicated to platform control (PCD). Graphics processing is managed by "Die 5", leveraging Intel's Xe3 technology."

What defined lunar lake wasn't the on package memory saving .5-1.5 W, or even the monolithic design of the product compared to meteor lake, but the excellent skymont-LPE cores and the decent onboard graphics. Lion cove was still relatively inefficient even compared to other x86 products like strix point, but the LPE-cores which were on the same tile as the rest of the cores (being a monolithic chip) and were powerful enough to actually be used to offload light workloads in many situations allowing for excellent battery life in light compute scenarios.

Panther-lake has the similarly excellent LPE cores on the same compute tile (I think, although a 4-core LPE darkmont cluster on the uncore die might be interesting since this time around it might actually be useful enough to turn off the Compute tile and run only the LPE tile), unlike Meteor Lake which had them on the SOC tile and in general had extremely weak and gutted LPE cores that weren't useful for anything in real use. The tiled architecture also isn't comparable to meteor-lake or arrow-lake since it only has 3 tiles, crucially integrating various functions of the "SOC" tile back into the main compute tile which in my view will likely eliminate much of the performance penalties meteor-lake and arrow-lake had in regards to their tiled architecture.

You only have 2 other tiles in addition to the compute tile, a graphics tile which is traditionally separated from the CPU anyways, and a die for "uncore" functions which I read as things such as I/O controllers, media complex, PCIE which do not need to be as closely connected to the rest of the CPU functions, I mean up until recent years there was a another die on the motherboard for some of these functions like PCIE/IO. There might be an argument for reduced power efficiency from inter-tile communication, but I wouldn't call it world-shifting, either way the "Core" of the CPU really seems to be in one tile this time around.

gdansk · Jan 8, 2025

ajsdkflsdjfio said:
...
What defined lunar lake wasn't the on package memory saving .5-1.5 W...

I agree.
PTL-U should be better than LNL in most ways technically. But also financially it is much more aligned with playing to Intel and its OEM customers strengths than LNL. 4+0+4/12 at similar power should not be hard unless something went very wrong. A few minor trade offs almost every partner is willing to make if the chip is cheaper and they get to mix memory as needed.

LNL was Intel's instinctual response. PTL-U is rethinking it and moving to a product design that fits with their business model and product stack. The supply chain doesn't favor Intel in producing a marginally better fit for a MBA that will never exist. But damn it is neat that they built that once anyway.

DavidC1 · Jan 8, 2025

ajsdkflsdjfio said:
The rumored PL1 and PL2 power (https://www.windowscentral.com/hardware/cpu-gpu-components/what-is-intel-panther-lake) also seems to line up with Lunar lake somewhat, although it likely won't have a 17W variant minus the 4 + 0 + 4 + 4XE3 variant.

That article says nothing about PL2. There was an early leak and PL2 is *much* higher on PTL-U than on Lunarlake. 57W if I remember correctly for -U.

Saying Pantherlake-U is close to Lunarlake could be also said about Meteorlake as well.

Battery life wise it'll be a regression compared to Lunarlake even if it might be better than Meteorlake. Just the optimized memory PHY with on package memory saves anywhere from 0.5-1W, which is a big deal on idle assuming the laptop doesn't use the power hungry 1440p+ screens.

Three negatives:
-Lack of on package memory which allows for PHY power reduction(not always but LNL Intel opted to save power)
-Existence of a separate LPE core, meaning the ones on the compute core isn't for saving power, but purely for performance. If you do not handle the hand-off properly to LPE you'd have a scenario where you are worse off in power due to two tiles being active. And because off die is still slower in many cases even if the LPE is as capable as the one in LNL, less cases for powering down.
-PTL is meant for multiple segments, thus SLC cache makes little sense, because it'll require the -U to be on a different die. LNL's SLC was dog slow but great for power savings.

AMD does just as well as Meteorlake without needing such complex setups. I expect maybe 10-15% over Meteorlake.

@dullard On package memory isn't unique. Every other vendor other than x86 did it and for more than a decade. Intel did it too with Medfield and successors. Actually mobile vendors do even better by stacking it on top and having a super compact package. My Bay Trail Tablet's internals showed that in a 15mm x 15mm area it had the SoC and memory all in one.

ajsdkflsdjfio · Jan 8, 2025

DavidC1 said:
That article says nothing about PL2. There was an early leak and PL2 is *much* higher on PTL-U than on Lunarlake. 57W if I remember correctly for -U.

You're right about the PL1/PL2 thing, I thought panther lake lineup would follow lunarlake trend of closer PL1 and PL2 numbers, but instead they're more close to Meteor Lake. IG that makes sense given panther-lake is meant to compete in a wide gamut of laptop performance tiers/form factors so they want the extra option of eeking out some additional performance at higher power.

Three negatives:
-Lack of on package memory which allows for PHY power reduction(not always but LNL Intel opted to save power)
-Existence of a separate LPE core, meaning the ones on the compute core isn't for saving power, but purely for performance. If you do not handle the hand-off properly to LPE you'd have a scenario where you are worse off in power due to two tiles being active. And because off die is still slower in many cases even if the LPE is as capable as the one in LNL, less cases for powering down.
-PTL is meant for multiple segments, thus SLC cache makes little sense, because it'll require the -U to be on a different die. LNL's SLC was dog slow but great for power savings.

For one, I'm not so sure that the LPE cores are on a separate die this time around since the "uncore" die is a lot smaller than the SOC tile of meteor lake, and is also small relative to the rest of the panther tiles, which makes me think the NPU and LPE cores are moved into the compute tile since both Lunar and Meteor had their NPU and LPE cores grouped together. Either way, if it is on the "uncore" tile it may not have the same issues as meteor-lake with handing off workloads to the LPE cores or atleast not nearly as much. Also why does being "off-die" necessitate that the LPE cores in panther lake are somehow worse than Lunar-Lake. I understand there might be some form of performance penalty, but the amount is unclear/varies based on implementation and the off-die e-cores of meteor-lake weren't ineffective simply because they were off-die.

Also what are you referring to with SLC cache, I might be uninformed about what SLC means in regards to CPU cache, but I thought that term is about cache layers for DRAM not SRAM in cpus. Are you referring to the cache hierarchy in lunar lake LPE cores? If so, I don't see how that's necessarily different than the cache setup on Panther lake LPE cores, after all, they are distinguishing between LPE and E-cores for a reason, leading me to believe LPE cores on Panther also have an altered cache setup for lower power operation.

I think the main differences between Meteor lake and Lunar lake power efficiency are the weak LPE-cores and the poor tiled architecture configuration specifically concerning the SOC tile, Panther lake solves both of these main issues. I agree, it probably won't have the same ultra-low power video streaming, or whatever it maybe, battery performance. But I'd argue it'd be closer to lunar-lake than meteor-lake, while offering performance outside of purely ultrabook levels. Overall as a product i'd say it's much more reminiscent of Lunar Lake than meteor lake in terms of power efficiency, architecture, and also with the excellent GPU. I think of it more like a beefed up lunar-lake rather than a successor to Meteor, but IG it doesn't really matter.

MoistOintment · Jan 8, 2025

PTL-U will be similar to LNL in that the SoC design will be overhauled to more closely resemble LNL, dropping the MTL/ARL design. But this is also true of PTL-H.

PLT-U differs from LNL in that it's not targeting the premium thin-and-light segment. It's focusing on the segment that U has always focused on: Lower cost.

PTL-U will be lower cost with no MoP (which was not implemented for performance but for idle power consumption). It'll be lower cost with less use of PMICs. It'll be lower cost with a smaller iGPU relative to LNL. It's NPU won't be any noticeably different than PTL-H. It's going to have worse idle efficiency than LNL and will probably have a gaming regression as well. It's not a 1:1 successor. It's more of a successor to "ARL-U", and any similarities it shares with LNL is true for the entire PTL family.

Philste · Jan 9, 2025

DavidC1 said:
There was an early leak and PL2 is *much* higher on PTL-U than on Lunarlake. 57W if I remember correctly for -U.

57W PL2 (or MTP how they call it now) was MTL-U. Even PTL-P with 4+8+4 and 12Xe has 64W MTP, so for now the data looks impressive.

DavidC1 · Jan 9, 2025

MoistOintment said:
It's going to have worse idle efficiency than LNL and will probably have a gaming regression as well. It's not a 1:1 successor. It's more of a successor to "ARL-U", and any similarities it shares with LNL is true for the entire PTL family.

It'll be worse in light load efficiency not just pure idle efficiency.

Laptop chips go between active and sleep states multiple times a second. Microsoft's alignment with Haswell's S0iX had 16ms period where the CPU can sleep. Back in the Pentium III days the CPUs could power down in between media playback frames. When you are talking hardware + software latency being that low, and your power management isn't fast enough then you'll either sacrifice performance or battery life.

@ajsdkflsdjfio Why would they have all three cores in the same tile? The whole point of Meteorlake's LPE was so you could turn off the compute tile entirely. When you are dealing with two dies though, you have to shut one off and hand the data over to the other all within that period I'm talking about. If the transition is too slow, you'll have more scenarios where two dies are active at once, increasing power.

They should have stuck with the two tile setup like Lunarlake, which is basically monolithic plus the PCH on the package with faster interconnects. There's no need for more than two tiles - the die size is small enough anyway. The tiles for sake of tiles should stop.

Ok, fine you saved money by not needing expensive extra PMICs and On package memory. Pretty sure having a dedicated mask just for the -U is worth it.

Either way, if it is on the "uncore" tile it may not have the same issues as meteor-lake with handing off workloads to the LPE cores or atleast not nearly as much. Also why does being "off-die" necessitate that the LPE cores in panther lake are somehow worse than Lunar-Lake. I understand there might be some form of performance penalty, but the amount is unclear/varies based on implementation and the off-die e-cores of meteor-lake weren't ineffective simply because they were off-die.

Why do you assume PTL is better than MTL in this regard other than speculative "improved execution"? Off-die makes the bad, worse, because you are introducing latency, which is a critical factor for good user experience while saving power. PTL changes nothing in this regard.

Oh, and Skymont doesn't take up much space so you can't tell from die size whether it's on it or not.

And the only reason GPU is on TSMC is because the company is split and disjointed and can't agree on anything.

But I'd argue it'd be closer to lunar-lake than meteor-lake, while offering performance outside of purely ultrabook levels.

There's no way a part with that many regressions in the power management department will come close to one that is to-the-wall optimized for power efficiency such as Lunarlake.

Philste said:
57W PL2 (or MTP how they call it now) was MTL-U. Even PTL-P with 4+8+4 and 12Xe has 64W MTP, so for now the data looks impressive.

PTL-U is 54W. It's basically same as 57W.

ajsdkflsdjfio · Jan 9, 2025

DavidC1 said:
There's no way a part with that many regressions in the power management department will come close to one that is to-the-wall optimized for power efficiency such as Lunarlake.

To-the-wall optimized doesn't necessarily mean above and beyond the competition, it just means that they took every step to improve power efficiency no matter how big/small the improvements were.

The only major regressions I see are the lack of on-package memory, less use of PMICs, and possibly power loss with inter-die communication. But the biggest ingredient of Lunar lake's efficiency over meteor-lake, the improved architecture and layout of the cores, are still applicable to panther lake AFAIK. I can definitely see like a 20-30% efficiency regression especially in very specific light workloads, like maybe in a scrolling webpage for 24 hours test it'd fall behind, but overall I'd say that PTL-U will be close in efficiency to lunar lake.

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

Senior member

Attachments

Junior Member

Elite Member

Member

Platinum Member

Senior member

Golden Member

Elite Member

Diamond Member

Diamond Member

Elite Member

Diamond Member

Elite Member

Diamond Member

Elite Member

Diamond Member

Elite Member

Diamond Member

Member

Diamond Member

Golden Member

Member

Member

Senior member

Golden Member

Member