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	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.)

 

[Thunder 57]

Sorry about that. But when I keep seeing consistent posts stating AMD and 7840U are superior at everything this and that, I had to make a point to put a stop to that tunnel vision. I do know MTL's performance is not expected to be better than RPL and MTL's power efficiency under full load will be no different than RPL.

But MTL isn't targeted at that. Its expected to have very good power efficiency for average users in average use cases mentioned previously. Squarely focusing on weaknesses like CB R15 is a waste of time imho as it doesn't really matter. Thats was the point i was trying to make. And apologies for being a bit too blunt. 🫢

Nothing to be sorry about. Plenty of homers around and they are usually easy to spot. I for one look forward to MTL's launch. It'll be the first real "new" thing for Intel since Alder lake two years ago. RPL was a bit different but I don't think anyone really thought it would be that big of a deal.

They are doing a lot at once though. New node, tiles, improved uarchs, SoC CPU's. Pretty bold to change so much at once. That's why I don't think they will hit it out of the park. I think they'll get some wins (efficiency, battery life) but stagnet elsewhere (max performance, clocks). If they can produce them in large quantities I think they will sell well. I can't imagine the ODM's are happy about the release date though!

 

[H433x0n]

Sorry about that. But when I keep seeing consistent posts stating AMD and 7840U are superior at everything this and that, I had to make a point to put a stop to that tunnel vision. I do know MTL's performance is not expected to be better than RPL and MTL's power efficiency under full load will be no different than RPL.

It's the internet.. it's to be expected. I personally don't understand it since I think AMD's current products are a total snooze fest. The Zen 4 core is a work of art but everything surrounding it is meh and their products are painfully boring (w/ exception of MI300). Ryzen, Threadripper, Epyc.. It's all the same thing - a substrate with a varying amount of CCD's placed on it. I think their most interesting product is probably Phoenix.

But MTL isn't targeted at that. Its expected to have very good power efficiency for average users in average use cases mentioned previously. Squarely focusing on weaknesses like CB R15 is a waste of time imho as it doesn't really matter. Thats was the point i was trying to make. And apologies for being a bit too blunt. 🫢

It will still outperform RPL in workloads such as Cinebench at 45W. How it performs past 60W is almost entirely irrelevant IMO.

 

[eek2121]

Just an observation. I see many people use the assumption that in Intel CPUs, 1 P core equals 4 E cores. It's not the case. I used a ADL die shot to measure the area. That is, after deducting the L3 cache area for both, a E core complex is roughly 1.43x bigger than a P core. That is, 1 P core is only around 2.8 E cores.

Including the L3 cache will throw the calculations off as L3 cache sizes can vary which isn't an accurate way to measure. But if included, in this case, the E core complex with L3 is roughly 1.3x bigger than a P core with L3. That is, with L3 (for both), 1 P core is around 3 E cores. But I think we shouldn't include L3 as it's not the right way imho.

It has nothing to do with die area and everything to do with ring stops, cache, etc.

 

[FlameTail]

Have you guys seen this masterpiece?

 

[ondma]

Have you guys seen this masterpiece?

Some people have too much time on their hands.

 

[ondma]

Why are you so passionate about this particular topic? Intel has a less efficient CPU at the moment.. so what? It’s well known at this point.

I posed a simple hypothetical to another poster and you launched into a multi page tirade that was off topic from my original post

Why are you so passionate about this particular topic? Intel has a less efficient CPU at the moment.. so what? It’s well known at this point.

I posed a simple hypothetical to another poster and you launched into a multi page tirade that was off topic from my original post.

I dont see why anyone even bothers to respond to this poster. He is a long time poster who clearly has an agenda and is not interested in unbiased discussion. Just put him on ignore.

 

[SiliconFly]

Have you guys seen this masterpiece?

Awesome video. But darth maul (lisa sue) is an evil villain who dies in the end in original star wars remember?

 

[SiliconFly]

I dont see why anyone even bothers to respond to this poster. He is a long time poster who clearly has an agenda and is not interested in unbiased discussion. Just put him on ignore.

The actual amd biased poster is known for his typical modus operandi for a long time. Posts very focused disparaging content about Intel to elicit a response thru which he can learn something about basic Intel tech. He's got no understanding of basic silicon engineering or how tech operates or has no understanding of process tech in general.

Just uses this forum to further his understanding by posting brainless and baseless inflammatory content usually. Most of us here know it already, but in the end, its just interesting and hilarious at times. Why miss out on our very own court jester!?!

 

[Hans Gruber]

Nothing to see here until Arrow Lake.

 

[Hulk]

Awesome video. But darth maul (lisa sue) is an evil villain who dies in the end in original star wars remember?

Actually Darth Maul does not die at the end of episode I. He lives quite a long time, has many battles, eventually comes up against the Emperor, and finally meets his fate (again and finally) at the hands of Obi Wan.

On another note. I told my wife I will finish the turkey and she is not to throw out a single piece. You see she and my two daughters don't eat it after the first night. But I will get through all of it!

 

[DrMrLordX]

Actually Darth Maul does not die at the end of episode I. He lives quite a long time, has many battles, eventually comes up against the Emperor, and finally meets his fate (again and finally) at the hands of Obi Wan.

Yeah but by that point he's changed his name to Darth Mauled. hyuk

 

[Tup3x]

Nothing to see here until Arrow Lake.

I wouldn't put it like that. Performance wise probably so but I'm interested in efficiency gains and how well their glue solution (so to speak) works. This going to be very important step for Intel.

 

[FlameTail]

Arrow Lake vs Zen 5

Who will win?

 

[tamz_msc]

Arrow Lake vs Zen 5

Who will win?

The ultimate winner will be all those posters who do not veer off-topic.

 

[SiliconFly]

Arrow Lake vs Zen 5

Who will win?

Just guessing (no basis whatsoever). Zen 5 may hold the ST performance crown. And ARL may slightly edge out Zen 5 in efficiency and also MT performance (due to new E cores & increased core count).

 

[igor_kavinski]

Arrow Lake vs Zen 5

Who will win?

Zen 5 gonna be released months earlier. How can ARL win if it doesn't even show up for the race?

 

[DrMrLordX]

Arrow Lake vs Zen 5

Who will win?

Wario.

 

[TESKATLIPOKA]

Oh wait, so you want me to show you something that doesn't exist?

Didn't you confirm that 6P+8E is comparable to 10 Golden Cove cores only at high TDPs?
So I wanted data showing that It's different at lower TDPs, because I wasn't sure about that.
Are you saying you confirmed(agreed with) something without knowing If It's really true?

Extremely stupid smh (talking about that logic, not u lol)
Because the design itself slots in 4 E-cores into one P-core. This isn't just die size, or I would easily be claiming some weird ratio of Zen 4C cores = 1 Zen 4 core (like 2 Zen 4 cores = 3 Zen 4C cores) for mobile skus. The way the 4 E-core clusters are designed is to easily slot into 1 P-core cluster.

This talk about a 4 E-core cluster slotting into a single P-core is still about die size.
Intel used a combination of P and E-cores to maximize performance without pointlessly increasing die size, which they did because 6P+8E CPU is comparable to an imaginary 10 Golden Cove CPU in performance even though It has the size of an imaginary 8 Golden Cove CPU.

When you are comparing CPUs, what are you comparing or looking at first?
Performance or die size?
So I find It ridiculous that you call my logic extremely stupid and yours the correct one.
In my humble opinion, my logic makes much more sense than yours.

When the boost of the E-cores are far out of their ideal range in order to maximize their perf/area, sure. That's highlighted in the chip and cheese article.

I already showed in my previous reply that this is not true.
Here It is once more:
6P at 1GHz + 8E at 1.9GHz perform and consume the same as 10P at 1.51GHz in libx264.
In exact numbers It would be 10.1FPS and 18.8-19W in libx264.

 

[ondma]

Zen 5 gonna be released months earlier. How can ARL win if it doesn't even show up for the race?

Well, ARL should compete against the last half of the Zen 5 cycle. But yea, it looks like Intel is falling behind. ARL will most likely not match Zen 5, and the later part of ARL will have to compete against Zen 6, so I dont see Intel having the performance lead until ?????. Hopefully they will at least be much more competitive on efficiency, if anybody cares on the desktop.

 

[lightisgood]

i5-13500H VS. Ryzen 7 7840H comparison test: Core AI Paint is 3 times better than its opponent and has stronger battery life

i5-13500H VS. 锐龙7 7840H对比测试：酷睿AI画图3倍于对手、续航更强

一、前言：i5-13500H与锐龙7 7840H的越级对比不同于桌面Zen4处理器所采用的CPU-Die与I/O Die分离的Chiplets设计方案，为了降低功耗，笔记本端的Zen4构架锐龙处理器采用的是单Die设计，同时将

news.mydrivers.com

To reconfirm, RPL notebook has enough performance for competing with Zen4.
So, MTL could break Zen5.
Generally, the process node jump (RPL => MTL) is more efficient than uarch jump (Zen4 => Zen5) in mobile market, and then ARL is expected to achieve the double jump.

 

[Abwx]

i5-13500H VS. Ryzen 7 7840H comparison test: Core AI Paint is 3 times better than its opponent and has stronger battery life

i5-13500H VS. 锐龙7 7840H对比测试：酷睿AI画图3倍于对手、续航更强

一、前言：i5-13500H与锐龙7 7840H的越级对比不同于桌面Zen4处理器所采用的CPU-Die与I/O Die分离的Chiplets设计方案，为了降低功耗，笔记本端的Zen4构架锐龙处理器采用的是单Die设计，同时将

news.mydrivers.com

To reconfirm, RPL notebook has enough performance for competing with Zen4.
So, MTL could break Zen5.
Generally, the process node jump (RPL => MTL) is more efficient than uarch jump (Zen4 => Zen5) in mobile market, and then ARL is expected to achieve the double jump.

80W and 2164 pts in CB R15 for the 13500H.

65W and 2544 pts in CBR5 for the 7840H.

To be at same level than the 13500H the 7840H need only 45W, that s a massive 75% perf/watt advantage at same perfs, if that is considered as competetive enough, well, so good...

A 6 + 8 would do better by reducing the perf/watt disadvantage to a still big 1.65x, and that s also the perf/watt improvement required by MTL to catch up with PHX1.

 

[eek2121]

i5-13500H VS. Ryzen 7 7840H comparison test: Core AI Paint is 3 times better than its opponent and has stronger battery life

i5-13500H VS. 锐龙7 7840H对比测试：酷睿AI画图3倍于对手、续航更强

一、前言：i5-13500H与锐龙7 7840H的越级对比不同于桌面Zen4处理器所采用的CPU-Die与I/O Die分离的Chiplets设计方案，为了降低功耗，笔记本端的Zen4构架锐龙处理器采用的是单Die设计，同时将

news.mydrivers.com

To reconfirm, RPL notebook has enough performance for competing with Zen4.
So, MTL could break Zen5.
Generally, the process node jump (RPL => MTL) is more efficient than uarch jump (Zen4 => Zen5) in mobile market, and then ARL is expected to achieve the double jump.

Meteor Lake is going to be well behind Zen 5, but will do great at competing with current gen, perhaps even beating Zen 4 at perf/watt and absolute performance. (Except, perhaps, the desktop wannabe chip, 7945hx).

Zen 5 mobile is going to be an interesting beast, but I suspect Arrow Lake will have no issue keeping up.

I've been a bit out of the loop lately (job hunting), but I hope we get some new info regarding Meteor Lake soon. I would love to see some detailed numbers concerning power. I am considering the chip for my next laptop upgrade (depending on when I decide to pull the trigger)

 

[dullard]

I've been a bit out of the loop lately (job hunting), but I hope we get some new info regarding Meteor Lake soon. I would love to see some detailed numbers concerning power. I am considering the chip for my next laptop upgrade (depending on when I decide to pull the trigger)

2.5 weeks until the official launch. That said, laptop reviews aren't as immediate as desktop reviews. A proper laptop review takes far longer* and the laptops usually trickle in as it takes time for manufacturers to get everything ready to their satisfaction. Does a chip look good because the chip is good or because the laptop itself is good? You won't really know until a proper comparison with the same laptop model is out for all the chips you want to compare. So, while we'll have a lot of data on Dec 14, we won't really have a complete picture until probably January.

* A reviewer not only should test performance and power usage at initial bootup, but also at varying conditions. What about when the laptop is hot and thermally throttling (hot from the CPU use or just hot because the room is hot), what about when under battery power and Windows is not set to full performance mode, etc.

 

[Hulk]

2.5 weeks until the official launch. That said, laptop reviews aren't as immediate as desktop reviews. A proper laptop review takes far longer* and the laptops usually trickle in as it takes time for manufacturers to get everything ready to their satisfaction. Does a chip look good because the chip is good or because the laptop itself is good? You won't really know until a proper comparison with the same laptop model is out for all the chips you want to compare. So, while we'll have a lot of data on Dec 14, we won't really have a complete picture until probably January.

* A reviewer not only should test performance and power usage at initial bootup, but also at varying conditions. What about when the laptop is hot and thermally throttling (hot from the CPU use or just hot because the room is hot), what about when under battery power and Windows is not set to full performance mode, etc.

These points regarding laptop reviews are very important. The key will be trying to compare processors in the same laptop so that the cooling (thermal design) and power (battery) are equivalent.

 

[SiliconFly]

i5-13500H VS. Ryzen 7 7840H comparison test: Core AI Paint is 3 times better than its opponent and has stronger battery life

i5-13500H VS. 锐龙7 7840H对比测试：酷睿AI画图3倍于对手、续航更强

一、前言：i5-13500H与锐龙7 7840H的越级对比不同于桌面Zen4处理器所采用的CPU-Die与I/O Die分离的Chiplets设计方案，为了降低功耗，笔记本端的Zen4构架锐龙处理器采用的是单Die设计，同时将

news.mydrivers.com

To reconfirm, RPL notebook has enough performance for competing with Zen4.
So, MTL could break Zen5.
Generally, the process node jump (RPL => MTL) is more efficient than uarch jump (Zen4 => Zen5) in mobile market, and then ARL is expected to achieve the double jump.

Efficiency wise, ARL might beat Zen 5 in mobile. But Zen 5 might beat ARL in raw performance in mobile. Just thinking out loud.