Intel "Raptor Lake Refresh" Meant to Fill in for Scrapped "Meteor Lake" Desktop?

[btarunr]

Intel's 2023 roadmap for the desktop processor segment sees the company flesh out its 13th Gen Core "Raptor Lake" desktop family with 65 W (locked) SKUs, and the new i9-13900KS flagship; followed by a new lineup of processors under the "Raptor Lake Refresh" family, due for Q3-2023, with no mentions of a desktop "Meteor Lake" processor in the year. It turns out that "Raptor Lake Refresh" is being designed to fill in for these (i.e. there won't be any "Meteor Lake" desktop chips). This, according to OneRaichu, a reliable source with Intel leaks.

"Meteor Lake" is Intel's first client processor to fully incorporate the company's IDM 2.0 product development strategy of disintegrating the processor into multiple chiplets built on various foundry nodes based on design needs; and combining them onto a single package with a high-performance interconnect. "Meteor Lake" has just one problem and that is CPU core-counts, with rumors pointing to 6P+16E (6 performance cores + 16 efficiency cores) being the maximum core-count possible, something Intel probably feels won't be competitive in the desktop segment against AMD, which will probably have a lineup of "Zen 4" X3D processors out by Q3-2023, with up to 16 P-cores. The company will, however, give "Meteor Lake" a sizable launch in the various mobile segments.



"Raptor Lake Refresh" remains shrouded in mystery, particularly what Intel does with packaging it—whether it retains LGA1700 or uses the next LGA1851 package; or whether it is a speed-bump, or like "Coffee Lake Refresh," Intel could even increases the core-counts. Assuming Intel doesn't change the silicon from the present 8P+16E, the "Refresh" series could see incremental core-count uplifts among each Core brand extension (eg: Core i5 going from 6P+8E to 6P+16E); besides clock speed increases. Should Intel take the path of changing the socket to LGA1851, the company might change the branding to 14th Gen Core, release a new chipset, with the socket probably offering improved I/O, such as CPU-attached PCIe Gen 5 NVMe (currently Gen 4). These LGA1851 motherboards will come with preparation for next-generation "Arrow Lake" processors due in 2024.

View at TechPowerUp Main Site | Source

 

[usiname]

The 10nm disaster repeats again, this time with 7nm rebranded as intel 4.

 

[TheinsanegamerN]

I'm much more interested in meteor lake for mobile with its 128/192eu HPG-2 based iGPU and how close/how much faster it is then AMD's 680m.

 

[qubit]

6P16E? I'll pass. I really don't like this "efficiency" core crap for desktop. Intel is right that it wouldn't be competitive and hence not releasing it. AMD has a big chance here to really nail Intel with lots of full power cores for desktop CPUs.

 

[Deleted member 185088]

6P16E? I'll pass. I really don't like this "efficiency" core crap for desktop. Intel is right that it wouldn't be competitive and hence not releasing it. AMD has a big chance here to really nail Intel with lots of full power cores for desktop CPUs.

Could make sense for them, the e-cores take little place in the die, plus they offer 6/7th gen performance, helps them keep up in multi threaded applications, and for single threaded applications they just clock the hell out of the e-cores and keep up there as well, unless we start to see applications relaying on p-cores more then Intel would be in real trouble.

 

[watzupken]

6P16E? I'll pass. I really don't like this "efficiency" core crap for desktop. Intel is right that it wouldn't be competitive and hence not releasing it. AMD has a big chance here to really nail Intel with lots of full power cores for desktop CPUs.

I feel this efficiency core thing will not go away. In fact, AMD will at some point embrace this big/little core configuration. The reason is because if you think how much Intel is charging for essentially 6/8 performance core now, and spamming the CPU with lots of efficient cores that's cheap to produce, Intel is likely making a lot with the i7 and i9 chips. I do have to admit, this strategy works very well for Intel because it is clear that all they need to do is to focus on pushing P-core power limit so that it can perform very well in games/ applications that do not require so that many cores/ threads, and using the E-cores to bump up performance when high core counts are required. After all, physical cores are still better than hyper threading. Thus, AMD was not able to outperform Intel in both low and high threaded workload despite investing a lot in making those performance cores only CPU. So I guess at some point, I am pretty sure AMD will go down the same approach since they have already patented a big/little core config design about a couple of years back.

Could make sense for them, the e-cores take little place in the die, plus they offer 6/7th gen performance, helps them keep up in multi threaded applications, and for single threaded applications they just clock the hell out of the e-cores and keep up there as well, unless we start to see applications relaying on p-cores more then Intel would be in real trouble.

How applications utilize the cores is dependent on the software, or what MS calls the thread director. At no point will Intel allow themselves in this situation where applications will only rely on P-cores. Even if so, they can divert all the power to the P-cores to allow it to boost higher clockspeed, and parking the E-cores. In highly multi-threaded applications, the E-cores will make up for the lack of P-cores as you can observe now. Again, having a physical core is still better than a virtual one on the form of hyper threading.

 

[KrazyT]

the company flesh out its 13th Gen Core "Raptor Lake" desktop family with 65 W (locked) SKUs

Really curious to see how good (or not) they are !

 

[Dirt Chip]

So a full line of "ks" oc level is coming.

14100FKS to the rescue!

 

[john_]

Why change platform, when platform is the main advantage over AMD?
We are going back in stagnation for two reasons.

1) Zen 4 failed because of platform costs and not a clear win over Intel in single threaded benchmarks. And yes, who cares about power consumption? None.

2) Tech press and individuals supporting Intel swallowed/promoted Intel's marketing and now a 24 core Intel CPU is a 24 core CPU. Period. As long as Cinebench scores are better, that's what matters.
So, 8 P cores for the next 10 years. AMD will jump in this bandwagon soon, because it can't have meaningful profit margins trying to sell P cores against E cores. And even if they could enjoy some good profit margins, people will always consider 24 bigger than 16, 16 bigger than 12, 14 bigger than 8, 10 bigger than 6.

 

[ARF]

And yes, who cares about power consumption? None.

I care about power consumption. It is essential for low electricity expenses, for cool and quiet systems.
Next time you see someone screams "but it runs too hot", remember what you've said

 

[KrazyT]

I care about heat (often tied with power consumption).
And as a gamer, i found an i5 12400F (for exemple) very able to run good and chill.
My old physics teacher always told us that heat is the enemy of the computer chips ...
Can shortened the lifespan of the chips ...
Was he lying ?

 

[Dyatlov A]

Who needs 16 E Cores? They just adding heat unnecessarily. I don’t need any!

 

[The Von Matrices]

The 10nm disaster repeats again, this time with 7nm rebranded as intel 4.

I'm not surprised. Don't forget that Intel had the same situation with 14nm as well, so this would be the third node that can't achieve desktop performance in its initial release.

What I think is being missed here is that people are assuming that the 6P+16E die was intended for desktops and isn't performing as well as expected. I disagree; my reading indicates that the 6P+16E die was always intended to be the mobile chip, and the desktop would have had a different die with more P cores. Thus the reason for the cancellation of desktop Meteor lake is not because the 6P+16E die isn't achieving its expected performance but because the die with more P cores that was originally intended for desktop Meteor Lake was cancelled.

Zen 4 failed because of platform costs and not a clear win over Intel in single threaded benchmarks. And yes, who cares about power consumption? None.

Zen 4 is hardly a failure. It's currently selling just as well as comparable Intel systems. The failure was the initial CPU prices being too high for the performance offered, but once AMD dropped their prices to match Intel's price/performance then they became a compelling value and even I was convinced to buy.

The problem for AMD is that the platform is currently too expensive for low-end CPUs to make sense, but that is mainly due to DDR5 memory prices, of which AMD has little control.

 

[SL2]

The 10nm disaster repeats again, this time with 7nm rebranded as intel 4.

Obviously they haven't looked into rebranding it again to something like Intel 2+2, Intel <5, Intel not 7, or something else.
I mean, who knows, maybe the issues will go away..

I'm much more interested in meteor lake for mobile with its 128/192eu HPG-2 based iGPU and how close/how much faster it is then AMD's 680m.

Intel sure has great IGP's now, but by the time Meteor launches it will compete against the Zen 4 Phoenix APU.

I still doubt Zen 4 Dragon Range is a true APU, as 16 cores paired with fastest IGP doesn't make sense.

 

[Why_Me]

I care about heat (often tied with power consumption).
And as a gamer, i found an i5 12400F (for exemple) very able to run good and chill.
My old physics teacher always told us that heat is the enemy of the computer chips ...
Can shortened the lifespan of the chips ...
Was he lying ?

If the 13 gen locked cpu's are anything like the 12 gen locked cpu's then Intel has another hit.

Intel Core i7-12700 + Intel B660 Review

Today we have a combo review of all-new Intel hardware, our first look at the affordable Intel B660 platform, the non-K Core i7-12700 and Intel's latest box...

www.techspot.com

 

[docnorth]

the company flesh out its 13th Gen Core "Raptor Lake" desktop family with 65 W (locked) SKUs

Really curious to see how good (or not) they are !

You can check a leak or “leak” at videocardz. A 13500 in 65w (and 154w turbo bursts?) mode almost reaches 12600k MT performance.

What I think is being missed here is that people are assuming that the 6P+16E die was intended for desktops and isn't performing as well as expected. I disagree; my reading indicates that the 6P+16E die was always intended to be the mobile chip, and the desktop would have had a different die with more P cores. Thus the reason for the cancellation of desktop Meteor lake is not because the 6P+16E die isn't achieving its expected performance but because the die with more P cores that was originally intended for desktop Meteor Lake was cancelled.

Personally I disagree. My guess is most of us believe the same with you, i.e. that top Meteor Lake desktop chips were planned to feature at least 8 P cores, but Intel isn’t ready yet.

 

[john_]

The 10nm disaster repeats again, this time with 7nm rebranded as intel 4.

Considering Raptor Lake is a 7nm product and Meteor Lake is still happening in mobile, it's not a manufacturing problem. This is a decision from Intel to keep it's number of P cores at 8, to avoid negative reactions from consumers and also remain in the same platform for as long as it realises this is giving them a huge advantage over AMD's AM5.

I care about power consumption. It is essential for low electricity expenses, for cool and quiet systems.
Next time you see someone screams "but it runs too hot", remember what you've said

Most people look at benchmark charts. They don't care about power consumption as long as they buy the product will the longest line on those charts. And even if they build a system with a bad cooling solution that will prohibit their systems to reach those performance numbers, most of them will never realize it or care about it.

Who needs 16 E Cores? They just adding heat unnecessarily. I don’t need any!

Intel's marketing department needs them. The more E cores the more expensive the champagnes they are opening.

Zen 4 is hardly a failure. It's currently selling just as well as comparable Intel systems. The failure was the initial CPU prices being too high for the performance offered, but once AMD dropped their prices to match Intel's price/performance then they became a compelling value and even I was convinced to buy.

The problem for AMD is that the platform is currently too expensive for low-end CPUs to make sense, but that is mainly due to DDR5 memory prices, of which AMD has little control.

Having to drop prices so much and so early, for your brand new products that are based on your brand new architecture to just have some kind of parity with your competitor in retail sales, it's hardly a success. Having to also sell a 16 P core model at the price your competitor is selling an 8 P core model(plus some E cores), does make it a failure. Zen 4 P cores should have been at least 10-15% faster compared to Raptor Lake's P cores to make AMD look like a winner. AMD needs to look like a winner, because Intel's brand and ties with OEMs are extremely strong. Equal products always favor the stronger brand, especially when the stronger brand enjoys better marketing. Until Alder Lake AMD was selling more cores. Now Intel is selling more cores, at a lower price. Even with cheaper AM5 motherboards, MUCH cheaper AM5 motherboards, the core count still favors Intel. Let's not forget that the higher core count was the reason for AMD to gain market share while having lower IPC. Zen, Zen + and Zen 2 where selling because the consumer was seeing (much) higher number of cores in those CPUs compared to Intel's models.

Motherboard prices are a huge problem because consumers focus at the price, without considering the long term. To be fair, most of them don't know/care/even imagine the long rerm value of an AM5 motherboard.

I see you bought the ASUS STRIX X670E-F . So you invested $450 and that's probably the sum of your costs in the next 5-6 years, if you don't decide to upgrade latter to another mobo or the mobo doesn't die after the warranty period.
Someone buying an LGA1700 board will probably pay 2 times for a mobo in this period. That means $300-$800 depending the motherboard. Probably between $400-$600 most buyers. The advantages here are the option to go ultra cheap, the money someone will get from selling the old mobo, the better warranty period coverage (2 mobos, 2X2 or 2X3 years).
But in the end the price is the same or even favoring AM5. But people can't see that far, or don't care to see that far, or don't like staying with the same system for too long, or (and I mostly agree here) would prefer to get a new mobo in 2-3 years with any extra features that that future board will be offering, plus a new full warranty.

 

[TheinsanegamerN]

Intel sure has great IGP's now, but by the time Meteor launches it will compete against the Zen 4 Phoenix APU.

I still doubt Zen 4 Dragon Range is a true APU, as 16 cores paired with fastest IGP doesn't make sense.

I have no faith in AMD now. They sat on vega 7 for years and only begrudgingly moved to rDNA2 with zen3+. By all accounts the peonix point APUs (not the zen 2 or zen 3 models tho) will have the same core count as zen3+ for the GPU, just moving them to rDNA3. I doubt performance will be more then 5-10% faster. And given AMD's naming melarky it wouldnt surprise me if actual laptops sporting a zen4+rDNA3 APU are relatively rare. Intel stands a good chance of being faster then the majority of ryzen 7000 laptops, depending on how bad AMD's 7000 series pans out.

Why change platform, when platform is the main advantage over AMD?
We are going back in stagnation for two reasons.

1) Zen 4 failed because of platform costs and not a clear win over Intel in single threaded benchmarks. And yes, who cares about power consumption? None.

2) Tech press and individuals supporting Intel swallowed/promoted Intel's marketing and now a 24 core Intel CPU is a 24 core CPU. Period. As long as Cinebench scores are better, that's what matters.
So, 8 P cores for the next 10 years. AMD will jump in this bandwagon soon, because it can't have meaningful profit margins trying to sell P cores against E cores. And even if they could enjoy some good profit margins, people will always consider 24 bigger than 16, 16 bigger than 12, 14 bigger than 8, 10 bigger than 6.

So is zen 4 bad because its performance isnt impressive or is its performance not impressive because of tech press? You're contradicting yourself here.

I'll give you a hint: zen 4 was and own goal by AMD. Its too expensive (becaue AMD chose not to support DDR4) and it's performance isnt as much of a jump as many were hoping and its price is too high for what it offers. zen4x3d will hopefully perform better, but this is the same AMD that thought the likes of the 4100 and 4500 were good ideas after launching the rx 6400/6500.

 

[freeagent]

It would be nice if they could get creative with a new naming scheme. Core has been around since 775, i7 etc has been around since 1366, and these new CPUs have nothing to do with their past.

 

[john_]

So is zen 4 bad because its performance isnt impressive or is its performance not impressive because of tech press? You're contradicting yourself here.

I'll give you a hint: zen 4 was and own goal by AMD. Its too expensive (becaue AMD chose not to support DDR4) and it's performance isnt as much of a jump as many were hoping and its price is too high for what it offers. zen4x3d will hopefully perform better, but this is the same AMD that thought the likes of the 4100 and 4500 were good ideas after launching the rx 6400/6500.

No I am not. And you know it. YOU EVEN SAY IT. I thought you where missing the point, but YOU SAY IT. So, what exactly confuses you?

Zen 4 is bad because AMD's P cores don't beat Intel's P cores clearly. AMD selling CPUs with lower core count means they needed much faster P cores. As it was happening when AMD was selling Zen, Zen + and Zen 2. AMD had the core count advantage, but Intel had the single core performance advantage. Now Intel enjoys, lower platform costs, higher core count, while also been about the same in P core performance with AMD. I wrote it in my previous post.

As for tech press.
Tech press abandoned pretty quickly the idea of being detailed about Intel CPUs. 13900K is a 24 core CPU, period. It's cores can reach 5.8GHz period. No focusing on "Only 8 cores are P cores, only 8 cores are capable of reaching 5.8GHz". Just to give a couple of examples. The way tech press is covering Intel CPUs, make sure that the majority will understand "24 (equal) cores that (all) can reach 5.8GHz".
Please don't point at slides, pictures and stuff showing that 13900K is an 8P + 16E cores model. I am talking about the way people talk in videos, the way people write in their reviews in general, what the average viewer/reader in the end will keep.
In other words, what is happening in smartphones. That's the direction we are following. We are talking about desktop CPUs, but mainstream CPU models even for desktops, are going in that direction. Where smartphones are today. There are smartphone SOCs with only 1 P core and the rest of those, E cores or something in the middle. That's the future of mainstream desktop CPUs.

What probably you are missing is the fact that Zen 4 can be not enough at the same time that tech press is playing Intel's marketing game.

 

[ZoneDymo]

Why change platform, when platform is the main advantage over AMD?
We are going back in stagnation for two reasons.

1) Zen 4 failed because of platform costs and not a clear win over Intel in single threaded benchmarks. And yes, who cares about power consumption? None.

Idk if you live on planet earth, but power consumption has been on people's mind a lot for a while now.

 

[john_]

Idk if you live on planet earth, but power consumption has been on people's mind a lot for a while now.

Yeah, sure, of course, absolutely. No one is buying RTX 4000 and 13th gen products because of POWER CONSUMPTION.
Definitely.


I apologize for the irony in the post, but you have to be kidding me or being sarcastic, or while living on planet Earth, have no contact with humans.

Intel remained competitive until Alder Lake by ignoring power consumption and keeps doing so. RTX 4090 is a beast because Nvidia said "F___ power consumption, go for maximum performance".
AMD's Zen 4 is somewhat competitive thanks to the fact that AMD throw efficiency out the window by introducing 170W products, and RDNA3 is a fail, because they kept trying to keep it efficient and under 350W. We can expect RDNA4 to be a 500W product.

 

[ZoneDymo]

Yeah, sure, of course, absolutely. No one is buying RTX 4000 and 13th gen products because of POWER CONSUMPTION.
Definitely.


I apologize for the irony in the post, but you have to be kidding me or being sarcastic, or while living on planet Earth, have no contact with humans.

Intel remained competitive until Alder Lake by ignoring power consumption and keeps doing so. RTX 4090 is a beast because Nvidia said "F___ power consumption, go for maximum performance".
AMD's Zen 4 is somewhat competitive thanks to the fact that AMD throw efficiency out the window by introducing 170W products, and RDNA3 is a fail, because they kept trying to keep it efficient and under 350W. We can expect RDNA4 to be a 500W product.

That is not really how irony or sarcasm works.

 

[john_]

That is not really how irony or sarcasm works.

Please. You seem to know how much people care about power consumption in PC hardware and now,, you seem to know how irony and sarcasm works.
Please enlighten us.

 

[TheinsanegamerN]

Idk if you live on planet earth, but power consumption has been on people's mind a lot for a while now.

The high end doesnt care. The high end has NEVER cared. You seem to forget the high end market that drove 2kW PSUs, multi GPU rendering, and LN2 OCing isnt going to care about a few extra watts.

If an extra $2-5 a year worries you you are not in the market for a $1600 GPU in the first place.

No I am not. And you know it. YOU EVEN SAY IT. I thought you where missing the point, but YOU SAY IT. So, what exactly confuses you?

Well, you contradict yourself again here:

What probably you are missing is the fact that Zen 4 can be not enough at the same time that tech press is playing Intel's marketing game.

Zen 4 doesnt do enough and also the only reason people are negative on zen 4 is because of tech press sucking up to intel. You claimed earler one of the problems was that zen 4 failed to produce a clear win over intel, but then blame the tech media for why intel looks good, ignoring your previous point that intel does, in fact, outperform AMD again.

Zen 4 is bad because AMD's P cores don't beat Intel's P cores clearly. AMD selling CPUs with lower core count means they needed much faster P cores. As it was happening when AMD was selling Zen, Zen + and Zen 2. AMD had the core count advantage, but Intel had the single core performance advantage. Now Intel enjoys, lower platform costs, higher core count, while also been about the same in P core performance with AMD. I wrote it in my previous post.

Oh so now were back to "well actually AMD IS slower then intel". Let me throw you a bone here: you think that maybe the tech press is excited about intel because intel is making a comeback whereas AMD is starting to fall behind again?

Nah, its gotta be shilling.

As for tech press.
Please don't point at slides, pictures and stuff showing that 13900K is an 8P + 16E cores model. I am talking about the way people talk in videos, the way people write in their reviews in general, what the average viewer/reader in the end will keep.

":dont show me evidence that I am wrong".

Wow.