Intel Meteor Lake P-cores Show IPC Regression Over Raptor Lake?

[Solaris17]

So within its sphere, it's no laggard.

Lets not forget the addition to ARC instead or Iris XE. Compared to 13th gen and below (all the way to HD).

Intel Integrated Graphics Performance From Gen9 To Meteor Lake Arc Graphics - Phoronix

www.phoronix.com

 

[RandallFlagg]

Lets not forget the addition to ARC instead or Iris XE. Compared to 13th gen and below (all the way to HD).

Intel Integrated Graphics Performance From Gen9 To Meteor Lake Arc Graphics - Phoronix

www.phoronix.com

Yep, not really a gaming platform but much better than before, useable now. I personally wouldn't spend my own money on one of these, as I'd prefer a more powerful dGPU and at that point the iGPU is pointless. If I were in school, I'd probably go for the ML though, just so I didn't have to lug a gaming laptop and power brick around.

However, I'd really like a work laptop with one of these in it. I think that corporate laptop thing is really its niche, seems perfect for that and for students. A bit too high end in price IMO for say, grandma to pay her bills with though.

 

[Solaris17]

However, I'd really like a work laptop with one of these in it. I think that corporate laptop thing is really its niche, seems perfect for that and for students. A bit too high end in price IMO for say, grandma to pay her bills with though.

I use mine for this exact purpose. small like an ultrabook but much better compute density this gen than previously. Its nice to be able to work on things, render or compile and watch a training or even just youtube on another screen without it lagging. I enjoy doing dev work on mine a LOT its a breath of fresh air specifically to build and not be out of battery in 30min.

 

[chrcoluk]

Looking at that table, waiting for Apple to enter PC CPU market.

 

[lexluthermiester]

We shall see what further testing reveals, but looks like at intel they're still actively digging their grave...

Seriously with this fanboy crap? The thought never occurred to you that Intel made some architectural changes to make the series viable as an ultra-mobile and USFF CPU platform? I'm betting you didn't..

EDIT: Oh BTW, AMD Ryzen ultra mobile CPU's also have a lower IPC than the desktop variants... Just throwing it out there.

Now let's see some performance per watt specs.

 

[Squared]

Originally there wasn't supposed to be a Raptor Lake, nor a RPL refresh, nor a Rocket Lake for that matter. They were going to go AL -> Meteor Lake.

If you go back a little further these "supposed to bes" explain a lot. Alder Lake was supposed to be on Meteor Lake's node, which is why it was so power-hungry in laptops. Using Intel's new naming schemes:

Line	P Core µarch	Node	Tick/Tock
Skylake	Skylake	14nm	new µarch
Cannon Lake	Palm Cove	IFS 7*	die shrink
Ice Lake	Sunny Cove	IFS 7	new µarch
Tiger Lake	Willow Cove	IFS 4	die shrink
Alder Lake	Golden Cove	IFS 4	new µarch
Meteor Lake	Redwood Cove	IFS 20A	die shrink
Arrow Lake	Lion Cove	IFS 20A	new µarch

Now that I see this table, I get the impression that the die shrink microarchitectures were supposed to be trees and the new microarchitectures were supposed to be yellow. Also if Arrow Lake really comes out on IFS 20A then it will be the first line since Ice Lake to be on its originally intended node. It's just a few years later than once planned.

*Technically, what I'm calling IFS 7 here, Intel broke into 4ish names, 10nm, 10nm+ (renamed 10nm), 10nm++ (renamed 10SF), and 10ESF (renamed Intel 7). For simplicity I'm calling all of them IFS 7 here. Most of that table is basically a fairy tale anyway.

 

[kondamin]

Didn’t it take a while before software was ready to get everything out of alderlake and zen

 

[R0H1T]

Zen was fine till they let the TR's rip, this was in part because of optimizations that were needed(?) for servers & the odd configuration like TR 29xx with half the mem channels disabled.

 

[remixedcat]

We shall see what further testing reveals, but looks like at intel they're still actively digging their grave...

no they are digging for apples!

Pat gelsinger is digging for apples!

 

[InVasMani]

Hopefully Intel has done some refinement to push E core's peak performance a bit further by reducing some of the peak performance on P core's, but with some improved efficiency. It actually does make sense given you gain a lot more relative performance per E core ratio than per P core ratio. By making the P cores more efficient and upping the performance on E cores they can push more overall performance higher and improve efficiency. The ST performance will slightly drop, but seems like a very fair trade off to make and efficiency improving on P cores is certainly a plus those core are perfect for that role.

It seems like Intel is slowly coming to grips with the E-cores taking the fore front role and P cores having more of a back seat role. It's still going to take a bit of transitioning time, but most tasks are pretty multi-thread aware today and systems perform a lot more leisurely multi-tasking these days versus systems a decade ago.

The most notable thing to me is the graphics improvement and the SoC island looks like a promising addition. I think if you're on the fence though wait on the follow up generation because I'm sure they will iron out improvement upon it.

I'd like to see what Intel could do with a modern unlocked dual core and respectable binning. It would also be neat in that scenario to see them insert a larger iGPU and make it designed more as a buget minded HTPC/APU. It would be pretty great at it and low power.

 

[Denver]

It would have been fine to have a slight IPC regression in favor of a revolutionary low-TDP efficiency boost, but Meteor Lake didn't even deliver that. Low TDP performance is poor compared to 7840U, it will be very difficult to compete with Zen5 APUs.

 

[InVasMani]

Yeah and given it's mobile probably not the right call actually. I'd say they went wrong direction for mobile with this, but for desktop variants of this arch will probably make more sense. At least for now especially given cooling and battery life are important considerations and mobile devices are less workstation orientated in general in terms of multi-thread expectations.

 

[dyonoctis]

Hi,
No hate just facts
Intel knows people will buy anything they put out
That's why they refresh over and over.

Those refreshes are at least bringing some gains. Launching a CPU in late 2024 slower than a 2022 CPU will not play well in a market where AMD is getting more and more mindshare, and is VERY competitive. And with TSMC node pricing, they won't even have much wiggle room to play on the price.

If that happens, that would be an even worse launch than first gen P4

 

[pressing on]

Those refreshes are at least bringing some gains. Launching a CPU in late 2024 slower than a 2022 CPU will not play well in a market where AMD is getting more and more mindshare, and is VERY competitive. And with TSMC node pricing, they won't even have much wiggle room to play on the price.

If that happens, that would be an even worse launch than first gen P4

Yes, AMD is very competitive. However, there are factors that will help Meteor Lake compete. For example, every ML processor will have an NPU. Taking AMD's 88xxU range as an example, only the 8840U and 8860U will have this device. The 8540U and 8440U don't have an NPU. There is a large corporate market for mobile devices, if they are looking at 8440U level notebooks or laptops ML will likely be the default choice based purely on this one factor.

 

[BoggledBeagle]

Practical negative impact of slightly lower performance of P cores compared to previous gen on the actual users of the laptops is negligible, nobody is going to notice. It is no tragedy.

I wonder what is the real reason for this? One hypothesis could be that these P cores were a result of two development projects, one going really well, above expectations (Alder lake), and one having some problems (Meteor lake). That combined with possible delays of Meteor lake resulted in the lower performance in comparison with Alder/Raptor lake P core.

Or is something in Meteor lake P core simply broken and they could not make it to work in time?

 

[Tek-Check]

IPC regression is the main reason why Meteor Lake desktop line-up was cancelled and Raptor Lake refresh was introduced.

 

[RandallFlagg]

Practical negative impact of slightly lower performance of P cores compared to previous gen on the actual users of the laptops is negligible, nobody is going to notice. It is no tragedy.

I wonder what is the real reason for this? One hypothesis could be that these P cores were a result of two development projects, one going really well, above expectations (Alder lake), and one having some problems (Meteor lake). That combined with possible delays of Meteor lake resulted in the lower performance in comparison with Alder/Raptor lake P core.

Or is something in Meteor lake P core simply broken and they could not make it to work in time?

Taking a note from the post above (<- Link), there apparently was never any intention of ML being a new uArch.

Given that a uArch takes about 5 years start to finish, it should be no surprise that shifting to tiled design introduced a small performance penalty. This is Raptor Lake adapted to a tiled package and a die shrink.

There are four wins for Intel here. The new node, improved iGPU, inclusion of NPU, and the shift to tiled packaging. The user sees benefit from the iGPU, NPU, more cores, and the uArch of the E-cores is improved - so much improved multi-core performance.

Clearly, not so much benefit in single thread since the P-cores are likely raptor lake based.

People looking for big single thread bumps are looking for Arrow Lake.

The 'original' plan from the earlier post (no Raptor Lake here) :

 

[Denver]

Taking a note from the post above (<- Link), there apparently was never any intention of ML being a new uArch.

Given that a uArch takes about 5 years start to finish, it should be no surprise that shifting to tiled design introduced a small performance penalty. This is Raptor Lake adapted to a tiled package and a die shrink.

There are four wins for Intel here. The new node, improved iGPU, inclusion of NPU, and the shift to tiled packaging. The user sees benefit from the iGPU, NPU, more cores, and the uArch of the E-cores is improved - so much improved multi-core performance.

Clearly, not so much benefit in single thread since the P-cores are likely raptor lake based.

People looking for big single thread bumps are looking for Arrow Lake.

The 'original' plan from the earlier post (no Raptor Lake here) :

View attachment 328285

I don't see any of these gains mentioned, it falls behind the old AMD generation, below 30w. Chances are it would be a much more efficient product being a monolithic chip.

 

[FoulOnWhite]

Couple of gens of multi tile and intel will be killing it, wait and see.

 

[Tek-Check]

There are four wins for Intel here. The new node, improved iGPU, inclusion of NPU, and the shift to tiled packaging. The user sees benefit from the iGPU, NPU, more cores, and the uArch of the E-cores is improved - so much improved multi-core performance

Wins? Let's examine those "wins".

- new node brought compute regression in IPC. This is disappointing. It's not surprising that Intel cancelled desktop Meteor Lake. In 370 tests, Phoronix found that Phoenix 7840U is on average 28% faster, which is hugely embarrassing for Intel. Let's see whether any BIOS update brings less humiliation in compute workloads. Meteor Lake 155H is actually shockingly slow considering node shrink. It is what it is...

Intel Core Ultra 7 155H Meteor Lake vs. AMD Ryzen 7 7840U On Linux In 300+ CPU Benchmarks Review - Phoronix

www.phoronix.com

- NPU: Intel has never shared performance numbers of this AI engine. At best, it might perform as last year's XDNA, around 10 TOPS. In a few weeks, XDNA2 on Hawk Point will bring 16 TOPS

- iGPU is 8% faster on average in compute workloads (Phoronix big testing) than RDNA3 on 7840U. This is 1024 sp. (VEs) on 155H vs 768 sp. on Phoenix.

So, iGPU has 33% more silicon, but only 8% higher performance. Good catch-up, but not that impressive at all, which will be visible once Strix come out with 16CUs and 1024 sp.

- gaming is not clear. We need to see comprehensive, like-to-like tests in dozens of titles, with the same RAM, preferably on similar laptops.

But hey atleast their CEO is effective at begging for funds from government at taxpayers expense.

If you are a taxpayer, would you approve of this behaviour and then have Meteor Lake 155H delivered on your lap being over 20% slower in CPU compute than 7840U?

Or maybe they are going one step back, so they can sell higher IPC with their next series.

What would you say if any other company delivered a brand new CPU with IPC regression on a new node?

 

[Minus Infinity]

It would have been fine to have a slight IPC regression in favor of a revolutionary low-TDP efficiency boost, but Meteor Lake didn't even deliver that. Low TDP performance is poor compared to 7840U, it will be very difficult to compete with Zen5 APUs.

And efficiency was also being hyped all through 2022 and early 2023. Again MLisD was claiming 30-40% lower power at same performance. Let's not forget Meteor Lake was being shown off by Gelsinger in late 2021 saying silicon was fully working and being hyped for imminent release in early 2022 at latest and it took these clowns another 2 years to ship it and in a far worse state than planned and cancellation of desktop versions. IPC gains nil, efficiency improvements small. At least they got the iGPU up to speed. I just hope they can actually deliver worthwhile improvements in every area for Arrow Lake but I'm losing confidence it'll be the leap they will hype.

 

[Tek-Check]

Still glad I opted for 7840hs

Nothing to regret. Another variant, 7840U, is 28% faster in CPU compute tasks than 155H. It's a crushing defeat of Meteor Lake. So, yours is even faster.

 

[RandallFlagg]

I don't see any of these gains mentioned, it falls behind the old AMD generation, below 30w. Chances are it would be a much more efficient product being a monolithic chip.

No, it doesn't. Unless you're still talking about IPC and not overall performance.

Notebookcheck has it about 15-20% faster overall in their performance tests vs AMD 7840U.

CPU performance ratings :
The higher end AMD 7840U (15W) = 69.2
155H (28W) = 81.7
The 45W range 7840HS = 83.6
The 45W range 13700H = 84.8

The 155H isn't the top end ML, 165H is. Quite likely a 165H would match or beat the 45W parts above if they tested it.

Yes it's not a 15W chip, but neither is it a 45W chip. But it performs *very* close to those 45W range chips.

Ref:

https://www.notebookcheck.net/MSI-Prestige-16-B1MG-laptop-review-From-Core-i7-Xe-to-Core-Ultra-7-Arc.785587.0.html

https://www.notebookcheck.net/Lenovo-Yoga-Pro-7-14-G8-laptop-review-AMD-Zen4-isn-t-automatically-better.787943.0.html

 

[BoggledBeagle]

.... Let's not forget Meteor Lake was being shown off by Gelsinger in late 2021 saying silicon was fully working and being hyped for imminent release in early 2022 at latest and it took these clowns another 2 years to ship it and in a far worse state than planned and cancellation of desktop versions.

I would not call people clowns if they struggle or deliver something with delay in an industry making the most complex products a mankind ever produced using technology quite close to the edge of human knowledge.

On the other hand, what you wrote certainly does not sound good. Taking two steps forward and one step back is worse than taking three steps forward, that is for sure.

 

[FoulOnWhite]

I love the way people call them clowns, as if you could do better, maybe give intel a call, maybe you can help out.