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.)

maddie · Jan 28, 2024

Thunder 57 said:
Because you are wrong, about everything? "MTL shines bright". Tell me you honestly have no bias?

That phrase brought a laugh.

When does a meteor shine bright before plunging in the lake?

jpiniero · Jan 28, 2024

FWIW Pat claims Intel is going to ship 40 million "AI PCs" in 2024. Do note that probably includes some amount of Arrow and Lunar.

branch_suggestion · Jan 28, 2024

tamz_msc said:
Entrust a capable OEM like Dell or Lenovo to make a capable unit by not rushing to be the first - unlike Asus, and Acer - and MTL shines bright:

Lenovo IdeaPad Pro 5i review: Meteor Lake dazzles on performance and endurance

Meteor Lake shoots for the stars

www.tomshardware.com

The thing costs too much to make, the perf and efficiency don't really matter, it is perf per area that Intel is losing their GM on.
EMIB/Foveros is overkill for such a part, LNL being compute die+PCH like every prior Intel mobile part is for a reason, it is better, cheaper, more efficient and just plain simple.
If MTL was the same scale as M3 Max, then it's approach would make some sense. It is a dud part, doing nothing to change the market landscape and the packaging paradigm is being replaced instantly.

tamz_msc · Jan 28, 2024

branch_suggestion said:
The thing costs too much to make, the perf and efficiency don't really matter, it is perf per area that Intel is losing their GM on.
EMIB/Foveros is overkill for such a part, LNL being compute die+PCH like every prior Intel mobile part is for a reason, it is better, cheaper, more efficient and just plain simple.
If MTL was the same scale as M3 Max, then it's approach would make some sense. It is a dud part, doing nothing to change the market landscape and the packaging paradigm is being replaced instantly.

That wasn't the point at all. The point was that using early samples that reviewers got did not give a complete picture of how well MTL achieved the goals laid out for it by Intel.

Hitman928 · Jan 28, 2024

tamz_msc said:
That wasn't the point at all. The point was that using early samples that reviewers got did not give a complete picture of how well MTL achieved the goals laid out for it by Intel.

How does this review change that picture?

Thunder 57 · Jan 28, 2024

tamz_msc said:
That wasn't the point at all. The point was that using early samples that reviewers got did not give a complete picture of how well MTL achieved the goals laid out for it by Intel.

So when Intel preivewed Core 2 and they were winning do they do? Ah yes, Meteor Lake shines bright.

Tigerick · Jan 28, 2024

Intel To Utilize TSMC's 2nm Process In Next-Gen "Nova Lake" CPUs, Apple a Primary Client As Well

TSMC's 2nm process has attracted significant client interest, with Apple and Intel lining up to secure the initial production batch.

wccftech.com

As expected, Nova Lake-S 16P+32E will be manufactured by TSMC's N2 process, likely debut at Q4 2026 to utilize DDR6 platform. tCPU should integrate with IOD like Panther Lake, if Nova Lake's tGPU is using same tGPU as Panther Lake, then we should be expecting 50-60TOPS NPU as well...

Let's see:-

Intel Panther Lake 4P+8E by Intel 18A
Intel Nova Lake-S 8P+16E by Intel Next ???
Intel Nova Lake-S 16P+32E by TSMC N2

There should have one more version of Nova Lake with half amount of P&E cores, Intel will try to make it inhouse, likely to be new process announced later.

At this stage, if anybody thinks 18A is comparable to N2, hoho

poke01 · Jan 28, 2024

From a technical perspective it’s amazing that the MacBook Air 15 beats the Lenovo despite packing a much smaller battery (66Wh vs 84Wh) by 3 hrs in that battery test. It really does showcase how far Apple is ahead in performance per watt and idle efficiency.

Hitman928 said:
another one with a very large battery.

DavidC1 · Jan 28, 2024

Tigerick said:
At this stage, if anybody thinks 18A is comparable to N2, hoho

Panther Lake is early 2026, thus Nova Lake is a 2027 product.

Why do you automatically assume it's not for some other tile? They did say the tile configuration is changing again with Panther Lake.

The media is also the one that wrongly peddled the "Meteorlake GPU is now N5 instead of N3 like previously rumored!!".

tamz_msc · Jan 28, 2024

UDN news has also been associated with disinformation in the past. So I would wait for some other sources to report on a product that is three years out from now.

mikk · Jan 28, 2024

DavidC1 said:
Panther Lake is early 2026, thus Nova Lake is a 2027 product.

Why do you automatically assume it's not for some other tile? They did say the tile configuration is changing again with Panther Lake.

The media is also the one that wrongly peddled the "Meteorlake GPU is now N5 instead of N3 like previously rumored!!".

Intel says 2025. Panther Lake is supposed to be a Lunar Lake successor with 4+8 cores whereas Nova Lake should be a replacement for Arrow Lake in the higher performance mobile and desktop segment. I don't see how Panther Lake affects the Nova Lake launch timeframe. Lunar Lake and Arrow Lake are launching close together because there is no correlation, they serve for different segments.

uzzi38 · Jan 29, 2024

tamz_msc said:
Entrust a capable OEM like Dell or Lenovo to make a capable unit by not rushing to be the first - unlike Asus, and Acer - and MTL shines bright:

Lenovo IdeaPad Pro 5i review: Meteor Lake dazzles on performance and endurance

Meteor Lake shoots for the stars

www.tomshardware.com

And you can always entrust Tom's Hardware to do a crappy job of reviewing a laptop, too!

No mention of power consumption with any of the tested devices, just performance and temperatures (the thing that is totally irrelevant unless your CPU is sitting at >90c under load or we're talking chassis temperatures).

The colour accuracy chart is clearly wrong, unless you think a device can cover 74.2% of the sRGB colour space and 104.8% of the DCI-P3 colour space.

No mention of battery sizes, which means the fact that the Swift Edge 16 has a 54WHr battery and the Macbook Air has a 66.5WHr while the Ideapad Pro 5i has an 84WHr battery is completely lost (which puts a significant dampener on the battery life figures).

No measurements of audio from the fans at all, just a vague mention of it being "noticeable but not objectionable".

But yeah, if you think this is a review that makes Meteor Lake "shine bright" then by all means, go ahead!

tamz_msc · Jan 29, 2024

uzzi38 said:
No mention of battery sizes, which means the fact that the Swift Edge 16 has a 54WHr battery and the Macbook Air has a 66.5WHr while the Ideapad Pro 5i has an 84WHr battery is completely lost (which puts a significant dampener on the battery life figures).

Okay, let's normalize the results:

Meteor Lake-H (Lenovo): 697 minutes with 84 Wh battery capacity = 8.3 minutes/Wh
Phoenix-U (Acer): 438 minutes with 54 Wh battery capacity = 8.1 minutes/Wh

In favor of Lenovo - LCD panel
Not in favor of Acer - OLED panel

So we need some other way to account for the difference in the displays, but in the end, it doesn't matter because, fortunately, notebookcheck measured "idle minimum power" for the Acer Swift Edge:

Idle minimum power (display set to minimum brightness, all wireless off, and power plan in Windows set to Energy Saving) = 7.1 W

So what should be the battery runtime be in this state for the Acer?

54 Wh/7.1 W = ~ 7.6 hours = 456 minutes

Which is just a tad more than the battery runtime that Toms Hardware got, while doing "something" instead of "nothing". So at least their result is logically consistent with the result that NBC got, even though they were testing the laptop doing different things.

Oh, let's go back to MTL. As per these two tests, the Lenovo with MTL ran for ~4 hours more than the Phoenix Acer doing SOMETHING instead of NOTHING.

So much for AMD's "energy efficiency".

Reiterating - Intel can deliver 4 HOURS more battery runtime doing SOMETHING, while AMD does

S'renne · Jan 29, 2024

tamz_msc said:
Okay, let's normalize the results:

Meteor Lake-H (Lenovo): 697 minutes with 84 Wh battery capacity = 8.3 minutes/Wh
Phoenix-U (Acer): 438 minutes with 54 Wh battery capacity = 8.1 minutes/Wh

In favor of Lenovo - LCD panel
Not in favor of Acer - OLED panel

So we need some other way to account for the difference in the displays, but in the end, it doesn't matter because, fortunately, notebookcheck measured "idle minimum power" for the Acer Swift Edge:

Idle minimum power (display set to minimum brightness, all wireless off, and power plan in Windows set to Energy Saving) = 7.1 W

So what should be the battery runtime be in this state for the Acer?

54 Wh/7.1 W = ~ 7.6 hours = 456 minutes

Which is just a tad more than the battery runtime that Toms Hardware got, while doing "something" instead of "nothing". So at least their result is logically consistent with the result that NBC got, even though they were testing the laptop doing different things.

Oh, let's go back to MTL. As per these two tests, the Lenovo with MTL ran for ~4 hours more than the Phoenix Acer doing SOMETHING instead of NOTHING.

So much for AMD's "energy efficiency".

Reiterating - Intel can deliver 4 HOURS more battery runtime doing SOMETHING, while AMD does

View attachment 92685

Amazing that you completely ignored the different battery capacities between the two laptops where meteor lake has 84 Wh while Phoenix has 54 as if battery maximum capacity doesn't affect total battery life....lmfao

tamz_msc · Jan 29, 2024

S'renne said:
Amazing that you completely ignored the different battery capacities between the two laptops where meteor lake has 84 Wh while Phoenix has 54 as if battery maximum capacity doesn't affect total battery life....lmfao

Normalization - do you know what that means?

Hitman928 · Jan 29, 2024

tamz_msc said:
Okay, let's normalize the results:

Meteor Lake-H (Lenovo): 697 minutes with 84 Wh battery capacity = 8.3 minutes/Wh
Phoenix-U (Acer): 438 minutes with 54 Wh battery capacity = 8.1 minutes/Wh

In favor of Lenovo - LCD panel
Not in favor of Acer - OLED panel

So we need some other way to account for the difference in the displays, but in the end, it doesn't matter because, fortunately, notebookcheck measured "idle minimum power" for the Acer Swift Edge:

Idle minimum power (display set to minimum brightness, all wireless off, and power plan in Windows set to Energy Saving) = 7.1 W

So what should be the battery runtime be in this state for the Acer?

54 Wh/7.1 W = ~ 7.6 hours = 456 minutes

Which is just a tad more than the battery runtime that Toms Hardware got, while doing "something" instead of "nothing". So at least their result is logically consistent with the result that NBC got, even though they were testing the laptop doing different things.

Oh, let's go back to MTL. As per these two tests, the Lenovo with MTL ran for ~4 hours more than the Phoenix Acer doing SOMETHING instead of NOTHING.

So much for AMD's "energy efficiency".

Reiterating - Intel can deliver 4 HOURS more battery runtime doing SOMETHING, while AMD does

View attachment 92685

If you read your own link for the Acer review in Notebookcheck. . .

Power consumption when idling is higher than on most other laptops with integrated GPUs due to the high 120 Hz refresh rate and OLED panel of our Acer configuration which are known to be more power demanding than the usual 60 Hz IPS. Displaying an all-white image at 120 Hz would consume as much as 19 W compared to less than half on the Framework Laptop.

The Lenovo has a higher refresh than the Framework laptop but still an IPS screen. The higher refresh won't bring the power consumption to near the Acer laptop screen consumption. The Acer laptop has a higher resolution, OLED screen which takes significantly more power than the IPS panel in the Lenovo model and is the overwhelming consumer of power at idle or low load conditions. So the fact that (when normalized for battery size) the Acer even comes close to the Lenovo in battery life is actually not a good look for the Lenovo unit.

In fact, the Framework laptop they mention with a 7840u and a 61 Whr battery gets over 12 hours of battery life in their web browsing test thanks largely to its more efficient screen than the Acer. If you gave it the same size battery as the Lenovo MTL laptop, it would get over 16.5 hours of battery life and the Framework doesn't even have the best battery life out of the Zen4 laptops. So no, the Lenovo's battery life test is not impressive given the type of display and size of battery it has.

tamz_msc · Jan 29, 2024

Hitman928 said:
If you read your own link for the Acer review in Notebookcheck. . .

The Lenovo has a higher refresh than the Framework laptop but still an IPS screen. The higher refresh won't bring the power consumption to near the Acer laptop screen consumption. The Acer laptop has a higher resolution, OLED screen which takes significantly more power than the IPS panel in the Lenovo model and is the overwhelming consumer of power at idle or low load conditions. So the fact that (when normalized for battery size) the Acer even comes close to the Lenovo in battery life is actually not a good look for the Lenovo unit.

In fact, the Framework laptop they mention with a 7840u and a 61 Whr battery gets over 12 hours of battery life in their web browsing test thanks largely to its more efficient screen than the Acer. If you gave it the same size battery as the Lenovo MTL laptop, it would get over 16.5 hours of battery life and the Framework doesn't even have the best battery life out of the Zen4 laptops. So no, the Lenovo's battery life test is not impressive given the type of display and size of battery it has.

The quoted paragraph that you cite from NBC is not talking what you think is idle minimum.

The idle minimum power consumption, that carries an asterisk next to it in the table under Energy Management section of the review, is a different metric altogether.

I have provided the link for the separate webpage that details what that is according to NBC.

At least read it before trying to argue.

Hitman928 · Jan 29, 2024

tamz_msc said:
The quoted paragraph that you cite from NBC is not talking what you think is idle minimum.

The idle minimum power consumption, that carries an asterisk next to it in the table under Energy Management section of the review, is a different metric altogether.

I have provided the link for the separate webpage that details what that is according to NBC.

At least read it before trying to argue.

I don't understand the point you are trying to make. Can you be more clear for how what I posted is not accurate?

Edit:

For clarity, since I did read it. Here is the idle minimum test procedure:

Idle: power consumption while the notebook is idle.
Minimum: all additional modules are off (Wi-Fi, Bluetooth, etc.), minimum brightness, and Windows power plan is set to "Energy Saving".

tamz_msc · Jan 29, 2024

Hitman928 said:
I don't understand the point you are trying to make. Can you be more clear for how what I posted is not accurate?

Edit:

For clarity, since I did read it. Here is the idle minimum test procedure:

And what is the idle minimum as measured by NBC for the Acer in their review?

Hitman928 · Jan 29, 2024

tamz_msc said:
And what is the idle minimum as measured by NBC for the Acer in their review?

7.1W which is 77.5% more than the Framework laptop with the same CPU but IPS screen. Again, what is your point?

tamz_msc · Jan 29, 2024

Hitman928 said:
7.1W which is 77.5% more than the Framework laptop with the same CPU but IPS screen. Again, what is your point?

The point is that OLED panels typically consume 2-3 W at zero brightness. So discounting the display, the Acer consumes 5 W at idle minimum.

Even with literally no display, and doing nothing, AMD implementation in the Acer delivers less battery runtime than the Lenovo, because 54/5 = 10.9 hours.

igor_kavinski · Jan 29, 2024

Henry swagger said:
mild said 6.5ghz can be reached on 8+16 not 8+32 yet..

Did he give a reason for that? Do the E-core clusters start melting when a single P-core is running at 6.5 GHz?

Or is it that 6.5 GHz matches the resonant frequency of the E-core cluster silicon and they start vibrating so hard that they continue vibrating seconds after the P-core activity ceases, causing latency so high that it becomes impractical?

Hitman928 · Jan 29, 2024

tamz_msc said:
The point is that OLED panels typically consume 2-3 W at zero brightness. So discounting the display, the Acer consumes 5 W at idle minimum.

Even with literally no display, and doing nothing, AMD implementation in the Acer delivers less battery runtime than the Lenovo, because 54/5 = 10.9 hours.

First, where do you get your OLED number from and what size screen is it in reference to? Second, the minimum idle test sets the panel to minimum brightness, not zero brightness (this would have to mean zero color signal to the OLED pixels or with the backlight turned off for an LCD panel which makes no sense), so even if your numbers were correct, they don't even apply. What you're describing is more of a connected standby situation which will consume less than 1 W because the screens are actually off in that situation though the CPU goes into a connected sleep state so not exactly the same. I'm honestly shocked you think Notebookcheck's minimum idle number is with the screen on, "zero brightness" and that the system in long idle is taking up to 5 W without the display.

Edit: I'll say this, let's say your observations with the Acer are correct and the whole systems without the display uses 5 watts in idle, then all that proves is that Acer screwed up their power plan settings because other laptops with regular DDR5 (unlike LPDDR5 in the Acer) and with their screens on can idle lower than 5 W (not OLED screens). Which again leads to the Lenovo laptop's battery life not being impressive when normalizing for the size of the battery. Heck, if we're just going to ignore the differences in screens, why not compare this HP with a 7840u that would get over 20 hours of battery life in their wi-fi or video playback tests if it had the same size battery as the Lenovo. That's way more impressive than what Tom's review of the Lenovo showed.

Edit 2: Another issue in comparing the minimum idle numbers that ties to the comments on power plans is that Notebookcheck measures the idle power when plugged in. While the reason they do this is obvious, you have to also realize that the AC idle power and battery idle power may not be the same depending on how the manufacturer sets up the power plan because it is not uncommon for them to configure it to have higher idle power (for better responsiveness) on AC power because the increased power consumption doesn't matter when plugged in. Additionally, because they are measuring at the plug, the efficiency of the power supplies will come into play and will be at their lowest efficiency at low power consumption. So the measured power readings at idle will be significantly higher than what the system itself is pulling from the external power supply with the rest being lost in the conversion. On top of that, each power supply will vary in efficiency though they should all be fairly close to each other.

tldr; the measured AC idle power is not what the system will draw when on DC power and can't be used as a stand-in for battery life at idle.

uzzi38 · Jan 29, 2024

tamz_msc said:
Okay, let's normalize the results:

Meteor Lake-H (Lenovo): 697 minutes with 84 Wh battery capacity = 8.3 minutes/Wh
Phoenix-U (Acer): 438 minutes with 54 Wh battery capacity = 8.1 minutes/Wh

Alright, I'm onboard with you here, this is true.

tamz_msc said:
In favor of Lenovo - LCD panel
Not in favor of Acer - OLED panel

This is also true.

tamz_msc said:
So we need some other way to account for the difference in the displays, but in the end, it doesn't matter because, fortunately, notebookcheck measured "idle minimum power" for the Acer Swift Edge:

Idle minimum power (display set to minimum brightness, all wireless off, and power plan in Windows set to Energy Saving) = 7.1 W

Let me stop you there for a moment. You do realise how bad 7.1W idle power is, right?

Here's a totally different device in the HP Elitebook 845 G10 with a 2560x1600 LCD panel sitting at an idle minimum of 4.3W.

OLEDs are egregiously bad in their power consumption, I really wouldn't just gloss over it like that. By the way, this is a 7840U device that gets 12 hours of battery life with a 51WHr battery. Just saying.

tamz_msc said:
So what should be the battery runtime be in this state for the Acer?

54 Wh/7.1 W = ~ 7.6 hours = 456 minutes

You can't really use those numbers to estimate overall battery life, I don't think they're accurate considering that's the minimum idle power, average idle power is higher, and the latter is how you should actually be "estimating" battery life.

Those numbers are the average idle power whilst monitoring the device, which is going to require you to spin up the chip to do in the first place. That's an extra workload not being discussed you're throwing into the mix.

tamz_msc said:
Which is just a tad more than the battery runtime that Toms Hardware got, while doing "something" instead of "nothing". So at least their result is logically consistent with the result that NBC got, even though they were testing the laptop doing different things.

Oh, let's go back to MTL. As per these two tests, the Lenovo with MTL ran for ~4 hours more than the Phoenix Acer doing SOMETHING instead of NOTHING.

So much for AMD's "energy efficiency".

Reiterating - Intel can deliver 4 HOURS more battery runtime doing SOMETHING, while AMD does

View attachment 92685

That's a lot of ranting to very conveniently ignore that Tom's Hardware got similar battery life (with the difference being in just a few minutes) with the same "workload" as the MTL device here. Like that's some real selective memory taking place here.

John Carmack · Jan 29, 2024

tamz_msc said:
Okay, let's normalize the results:

Meteor Lake-H (Lenovo): 697 minutes with 84 Wh battery capacity = 8.3 minutes/Wh
Phoenix-U (Acer): 438 minutes with 54 Wh battery capacity = 8.1 minutes/Wh

In favor of Lenovo - LCD panel
Not in favor of Acer - OLED panel

So we need some other way to account for the difference in the displays, but in the end, it doesn't matter because, fortunately, notebookcheck measured "idle minimum power" for the Acer Swift Edge:

Idle minimum power (display set to minimum brightness, all wireless off, and power plan in Windows set to Energy Saving) = 7.1 W

So what should be the battery runtime be in this state for the Acer?

54 Wh/7.1 W = ~ 7.6 hours = 456 minutes

Which is just a tad more than the battery runtime that Toms Hardware got, while doing "something" instead of "nothing". So at least their result is logically consistent with the result that NBC got, even though they were testing the laptop doing different things.

Oh, let's go back to MTL. As per these two tests, the Lenovo with MTL ran for ~4 hours more than the Phoenix Acer doing SOMETHING instead of NOTHING.

So much for AMD's "energy efficiency".

Reiterating - Intel can deliver 4 HOURS more battery runtime doing SOMETHING, while AMD does

View attachment 92685

Disinformation merchant.

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

Senior member

Attachments

Diamond Member

Lifer

Senior member

Diamond Member

Diamond Member

Platinum Member

Senior member

Senior member

Senior member

Diamond Member

Diamond Member

Platinum Member

Diamond Member

Member

Diamond Member

Diamond Member

Diamond Member

Diamond Member

Diamond Member

Diamond Member

Diamond Member

Lifer

Diamond Member

Platinum Member

Member