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

[Tigerick]

Wildcat Lake (WCL) Specs

Intel Wildcat Lake (WCL) is upcoming mobile SoC replacing Raptor Lake-U. WCL consists of 2 tiles: compute tile and PCD tile. It is true single die consists of CPU, GPU and NPU that is fabbed by 18-A process. Last time I checked, PCD tile is fabbed by TSMC N6 process. They are connected through UCIe, not D2D; a first from Intel. Expecting launching in Q1 2026.

		Intel Raptor Lake U	Intel Wildcat Lake 15W?	Intel Lunar Lake		Intel Panther Lake 4+4+4
Launch Date		Q1-2024	Q2-2026	Q3-2024		Q1-2026
Model		Intel 150U	Intel Core 7	Core Ultra 7 268V		Core Ultra 7 365
Dies		2	2	2		3
Node		Intel 7 + ?	Intel 18-A + TSMC N6	TSMC N3B + N6		Intel 18-A + Intel 3 + TSMC N6
CPU		2 P-core + 8 E-cores	2 P-core + 4 LP E-cores	4 P-core + 4 LP E-cores		4 P-core + 4 LP E-cores
Threads		12	6	8		8
Max Clock		5.4 GHz	?	5 GHz		4.8 GHz
L3 Cache		12 MB		12 MB		12 MB
TDP		15 - 55 W	15 W ?	17 - 37 W		25 - 55 W
Memory		128-bit LPDDR5-5200	64-bit LPDDR5	128-bit LPDDR5x-8533		128-bit LPDDR5x-7467
Size		96 GB		32 GB		128 GB
Bandwidth				136 GB/s
GPU		Intel Graphics	Intel Graphics	Arc 140V		Intel Graphics
RT		No	No	YES		YES
EU / Xe		96 EU	2 Xe	8 Xe		4 Xe
Max Clock		1.3 GHz	?	2 GHz		2.5 GHz
NPU		GNA 3.0	18 TOPS	48 TOPS		49 TOPS

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.

 

[Khato]

Yes, but the Intel ARC 390 is not 60W full system but 70W CPU sustained, actualy
70W sustained PL1, 85W PL2 and 212W PL4, Computerbase also did a test, so there s a long shot from 60W full system to say 65W sustained CPU power.

You want me to set my computer to a 1000W TDP and then claim its efficiency is absolutely horrible because it's not running any faster at 1000W than it is at 250W?

Again, the 60W full system power was measured by PCWorld, around 2:45 in their video coverage -

 

[Abwx]

You want me to set my computer to a 1000W TDP and then claim its efficiency is absolutely horrible because it's not running any faster at 1000W than it is at 250W?

Again, the 60W full system power was measured by PCWorld, around 2:45 in their video coverage -

You cant always use power at the main because there s sometime instances where the CPU drain power from both the AC adaptator and the battery, Computerbase did a review with the CPU at 70-65W, check the numbers, if they are the same as PCW it means that the latter is not accurate, beside Computerbase test is done at stock and they have the power settings displayed in HVInfo.

 

[poke01]

Yes, but the Intel ARC 390 is not 60W full system but 70W CPU sustained, actualy
70W sustained PL1, 85W PL2 and 212W PL4, Computerbase also did a test, so there s a long shot from 60W full system to say 65W sustained CPU power.

Edit : That s about 2x the TDP of a LNL, so if the GPU is 2x bigger there s basicaly no improvement in perf/watt from N3 to 18A.

Core Ultra 300; Erste Intel Arc B390 Benchmarks

Intel Core Ultra 300 „Panther Lake“ umfasst auch Varianten mit der iGPU Intel Arc B390. Erste Benchmarks bestätigen deren hohe Leistung.

www.computerbase.de

The GPU is on N3E.

 

[poke01]

You cant always use power at the main because there s sometime instances where the CPU drain power from both the AC adaptator and the battery, Computerbase did a review with the CPU at 70-65W, check the numbers, if they are the same as PCW it means that the latter is not accurate, beside Computerbase test is done at stock and they have the power settings displayed in HVInfo.

Please. Measuring the at the wall is always perfect and is the correct way to measure power consumption

 

[adroc_thurston]

Measuring the at the wall is always perfect and is the correct way to measure power consumption

Kind of.
That quickly becomes the benchmark for PDN losses or panel efficiency at high APL.

 

[Abwx]

I did specify what the numbers were, and they were picked as the closest option to the full-system power number that was provided by the PCWorld preview. Now it's entirely plausible that PTL will show higher or lower power numbers when subjected to the same process in full notebookcheck review, we're just ballparking at the moment.

As for the comparisons against almost two year old chips... well, that's what happens when the competition phones it in just like Intel used to. AMD's focus has moved on to AI. We'll still get table scraps, but their announcements at CES make it clear how much they care about consumer devices.

PCW say that they could tell us the TDP but then the guy say nothing about it, just that it s the max perf profile, beside he show the power at the main a few seconds on a random scene, not average during all the test, a sample of a few seconds for power is not representative, it can be a non demanding scene, but the fact that he say that it s max perf profile mean that the CPU TDP is likely the same than in Computerbase s test.

 

[poke01]

If the TDP is set to 60 watts for Panther Lake as a whole meaning the GPU can access 2x the power when gaming then it does invalidate the efficiency gain over lunar lake.

We will see

 

[Hulk]

Would have been some pretty good foresight if Intel could have had some Panther Lake with 32GB RAM on the die like Lunar Lake coming off the presses. Yes, I know some people NEED 64GB, but in general the unwashed masses (like me) only need 32GB and rarely if ever crack open the laptop case.

 

[Abwx]

If the TDP is set to 60 watts for Panther Lake as a whole meaning the GPU can access 2x the power when gaming then it does invalidate the efficiency gain over lunar lake.

We will see

That s the power for the whole chip whatever how it s distributed, beside in games the CPU power is much less than what is drained by the GPU, the CPU drain something like 10-15W at most for the stated 70W PL1.

 

[poke01]

That s the power for the whole chip whatever how it s distributed, beside in games the CPU power is much less than what is drained by the GPU, the CPU drain something like 10-15W at most for the stated 70W PL1.

Exactly so you’re saying we need wait tests on how it performs on lower TDPs over lunar lake and Strix point

 

[Abwx]

Exactly so you’re saying we need wait tests on how it performs on lower TDPs over lunar lake and Strix point

Since the GPU is much bigger there wont be much lower perf with half the TDP, it should still be quite ahead of LNL and 890M by the virtue of the SP count.

 

[mikk]

I wonder if any of the proper reviews for PTL graphics will include Strix Halo running at comparable power. Anyone aware of existing reviews for Strix Halo graphics which aren't running at 100W+?

The comparison points in notebookcheck's brief CES PTL preview, including external monitor Cyberpunk 2077 power numbers from the full reviews - https://www.notebookcheck.net/Intel...in-our-first-gaming-benchmarks.1200743.0.html
- Radeon 8060S: 92.1 fps, 195W
- RTX 5050: 75.2 fps, 129W
- Arc B390: 44.8 fps, 60W (according to PCworld preview, full system)
- Arc 140V: 28.3 fps, 45W
- Radeon 890M: 26.5 fps, 58W

Wccftech tested Cyberpunk with Strix Halo at 54W and Panther Lake at ~50W. Maybe Strix Halo at 45W is as fast as 388H at 45W if this game reflects more games.

In Cyberpunk 2077, the Intel Arc B390 iGPU delivered 99.45 FPS on average at 1080p using the Medium Preset and Balanced XeSS preset. The Strix Halo iGPU scores 111.92 FPS in the same benchmark at a 54 Watt TDP

We Tested Intel's Panther Lake "Arc B390" iGPU In Several Games, & It Truly Is A Game-Changer

At CES, Intel gave us the opportunity to test their newest Panther "Arc B390" iGPU, and the gaming performance truly impressed us.

wccftech.com

 

[LightningZ71]

While testing at the mains is often a reasonable method, he is correct that many systems can draw power from both the mains AND the battery as needed. If you are doing that sort of test, you really need to tighten up your methodology. Among other things, it needs to be a LONG duration canned test, like at least an hour, preferably more. Start each system at 100% battery charge and verify the actual battery capacity at 100% listed as many only charge to only 85% of rated cell capacity and call it 100%. Then, you capture the draw over the test duration at the wall, and then add in the battery charge used during the test to find total watt/hrs consumed.

We did this at work years ago to validate manufacturer power efficiency claims on some commercial applications and found that, not shocking to anyone, manufacturers have fudged past numbers quite a bit.

For a handheld, I would wait for actual reviews of individual devices that detail how long you can play game X from a full battery and maintain 60fps at reasonable quality. So much comes down to actual device implementation.

 

[Khato]

Wccftech tested Cyberpunk with Strix Halo at 54W and Panther Lake at ~50W. Maybe Strix Halo at 45W is as fast as 388H at 45W if this game reflects more games.

Thanks! I'm so used to wccftech being a source for nothing more than random rumors and questionable news that I didn't even think of checking them for first-party results from the CES demo, haha.

Also appreciate the clear statement that despite the absurdly high TDP settings, actual PTL power consumption remained around 50W max. Matches what I know about PTL. (Again, you can set the TDP as high as you want, that doesn't mean the chip is going to use that much power.)

 

[511]

Yes, but the Intel ARC 390 is not 60W full system but 70W CPU sustained, actualy
70W sustained PL1, 85W PL2 and 212W PL4, Computerbase also did a test, so there s a long shot from 60W full system to say 65W sustained CPU power.

Edit : That s about 2x the TDP of a LNL, so if the GPU is 2x bigger there s basicaly no improvement in perf/watt from N3 to 18A.

Core Ultra 300; Erste Intel Arc B390 Benchmarks

Intel Core Ultra 300 „Panther Lake“ umfasst auch Varianten mit der iGPU Intel Arc B390. Erste Benchmarks bestätigen deren hohe Leistung.

www.computerbase.de

This is TDP not actual power consumption and people have taken reading from devices

 

[poke01]

turn off the autodubbing its aweful lol

 

[poke01]

Black myth wukong 1080 medium upscaling off

 

[511]

While testing at the mains is often a reasonable method, he is correct that many systems can draw power from both the mains AND the battery as needed. If you are doing that sort of test, you really need to tighten up your methodology. Among other things, it needs to be a LONG duration canned test, like at least an hour, preferably more. Start each system at 100% battery charge and verify the actual battery capacity at 100% listed as many only charge to only 85% of rated cell capacity and call it 100%. Then, you capture the draw over the test duration at the wall, and then add in the battery charge used during the test to find total watt/hrs consumed.

We did this at work years ago to validate manufacturer power efficiency claims on some commercial applications and found that, not shocking to anyone, manufacturers have fudged past numbers quite a bit.

For a handheld, I would wait for actual reviews of individual devices that detail how long you can play game X from a full battery and maintain 60fps at reasonable quality. So much comes down to actual device implementation.

You can't use this cheat on battery and PC world tested power draw on battery it was around ~60W System

 

[ToTTenTranz]

https://twitter.com/x/status/2009997592634245494

some DF testing

Hopefully, this means it's over for overpriced Strix Point and Phoenix in handhelds. Especially with AMD refusing to distribute FSR4 for RDNA3, the difference in performance for ISO IQ is enormous.

It's either Panther Lake or cheap Strix Halo. Finally, some decent performance evolution since 2023.

 

[regen1]

Yes, but the Intel ARC 390 is not 60W full system but 70W CPU sustained, actualy
70W sustained PL1, 85W PL2 and 212W PL4, Computerbase also did a test, so there s a long shot from 60W full system to say 65W sustained CPU power.

Again, the 60W full system power was measured by PCWorld, around 2:45 in their video coverage -

Computerbase Article Author's comment

Watt-meter @1:50 onwards, stays under 60W.
Quote from ETA Prime's video:

The next game I was able to test here was Cyberpunk 2077. And before we get into it, I do want to show you what we've got. So, this is a watt meter coming from the wall. And unfortunately, I wasn't able to use like hardware info to get a correct reading, but it does look like the TDP on this setup was at 65 watts. So, we're not quite drawing 65 watts even running Cyberpunk like this.

This is TDP not actual power consumption and people have taken reading from devices

Yeah, this is just the configuration settings, not necessarily the power consumption. Even the HWinfo page from CB test shows just that, there's no readings/sensor page shown there.

Anyway we will get numbers from tests/reviews not that far away.

 

[511]

Anyway we will get numbers from tests/reviews not that far away.

it's like 17 days from now

 

[regen1]

Panther Lake im Akkubetrieb: Auch an der Batterie soll Core Ultra 300 richtig schnell sein

ComputerBase hatte noch einmal die Gelegenheit, mit führenden Panther-Lake-Ingenieuren zu sprechen. Thema diesmal: Leistung ohne Stromkabel.

www.computerbase.de

Qualcomm and Intel battery performance claims/issue:

The story in short: Qualcomm had claimed about eight months ago that Intel is losing up to 45 percent of power in battery operation, which is not the case with its solutions. This number was not devised, it came about by a faulty firmware in a Dell notebook coupled with curious power profile. However, since the solution was exactly in the trade and could be bought by everyone, the accusation was valid and Qualcomm’s marketing decorated it accordingly.

The topic also made high waves for Intel, so that the question of ComputerBase went directly in this direction at the start of CES 2026 in Las Vegas: How should this be prevented and what will it look like in the future? Intel immediately made it clear that it was not so natural, the “normal loss of power” at Lunar Lake was on average “only” 14 percent when the power cable was removed.

The problem with this, however, was that the OEMs were able to interpret certain things very loosely and broadly, so that it could differ more here and there – that in turn was Intel’s miscalculation. Coupled with incorrect settings by the OEM, a worst-case scenario could be created as the one at Dell.

Panther Lake battery-performance claims:

Panther Lake is getting much faster in battery operation

For Panther Lake, Intel is now taking OEMs more into account. The power profiles are now intended to ensure that the power in battery operation continues to reach very close to that of the socket.

In a first short test on site, the performance was also at almost the same level in games, regardless of whether the notebook was connected to the power or on the road in battery mode.

 

[Hulk]

Computerbase Article Author's comment
View attachment 136512

Watt-meter @1:50 onwards, stays under 60W.
Quote from ETA Prime's video:




Yeah, this is just the configuration settings, not necessarily the power consumption. Even the HWinfo page from CB test shows just that, there's no readings/sensor page shown there.

Anyway we will get numbers from tests/reviews not that far away.

"I wasn't able to run any CPU benchmarks right now."

Hmm.

 

[511]

"I wasn't able to run any CPU benchmarks right now."

Hmm.

there is near 0 peak CPU Performance improvement vs ARL all you get is same performance at lower power

 

[Magio]

Really hope NVL is a return to serious CPU gains because it's been a rough couple of years with peak performance basically plateau-ing since Raptor Lake with the only gains since coming in the form of efficiency.

On another note, leaving aside CPU and GPU metrics one area where PTL's launch is a huge success IMO is in the sheer quantity of *quality* and distinct designs from all major OEMs. Intel's partners truly brought their A game to complement Panther Lake and it's just the start.