Intel Meteor Lake CPU Family Details Allegedly Leak
Intel readies multiple Meteor Lake SKUs
Intel has already confirmed that its codenamed Meteor Lake processors are set to release in the coming months, but the company has naturally kept all the details about the upcoming lineup under wraps. However, a new slide leaked by Moore's Law Is Dead uncovers some of the peculiarities surrounding this new product family, including core count and thermal design power. The information is described as preliminary, so discretion is advised.
Intel's next-generation mobile CPU lineup will consist of multiple Meteor Lake-based models belonging to entry-level, midrange, and high-end segments with TDPs ranging from 7W to 45W and core counts varying from 5 to 14 cores. Intel's highest-end CPUs with up to 24 cores for gaming systems and mobile workstations will reportedly continue to use Raptor Lake silicon with Raptor Cove and Gracement cores as well as Xe-LP graphics in S-BGA packaging.
Intel's next-generation mobile CPU lineup, in general, looks as follows:
PL1 TDP | Core Configuration | Total Cores | iGPU | Silicon | Package |
7W | 1P + 8E | 5 to 9 | 3 Xe - 4 Xe | Meteor Lake | M |
9W | 2P + 8E | 6 to 10 | 3 Xe - 4 Xe | Meteor Lake | M |
15W | 4P + 8E | 6 to 12 | 3 Xe - 8 Xe | Meteor Lake | M |
28W | 6P + 8E | 10 to 14 | 7 Xe - 8 Xe | Meteor Lake | P |
45W | 6P + 8E | 12 to 14 | 8 Xe | Meteor Lake | H |
55W | 8P + 16E | 14 to 24 | 32 EU | Raptor Lake | HX |
Intel's Meteor Lake system-in-package consists of four tiles: the compute tile carrying high-performance Redwood Cove and energy-efficient Crestmont CPU cores and made using its Intel 4 fabrication process; the graphics tile based on the Xe-LPG architecture and produced by TSMC presumably using its N3 or N5 node; the SoC tile; and the I/O tile. The tiles will be interconnected using Intel's Foveros 3D technology.
Such an arrangement gives Intel quite some flexibility when it comes to configuring its Meteor Lake processors. It looks like 7W and 15W CPUs will use a miniature compute tile with up to 2P and 8E cores, whereas 15W, 28W, and 45W models will rely on a larger tile featuring up to 6P and 8E cores. As for graphics tile, Intel will probably have two chiplets: one with up to four Xe clusters and another with up to 8 Xe clusters. Intel will be able to mix and match chiplets depending on exact SKUs.
One thing to remember about the slide leaked by Moore's Law Is Dead is that it is said to be preliminary, so many things may change before an official launch. Models of the upcoming CPUs are marked as Core i3, Core i5, Core i7, and Core i9. Meanwhile, Intel has officially confirmed plans to switch to Core Ultra branding for Meteor Lake in early May, so the slide is outdated.
Since we are dealing with an unofficial source, take the information with a grain of salt.
Stay On the Cutting Edge: Get the Tom's Hardware Newsletter
Get Tom's Hardware's best news and in-depth reviews, straight to your inbox.
Anton Shilov is a contributing writer at Tom’s Hardware. Over the past couple of decades, he has covered everything from CPUs and GPUs to supercomputers and from modern process technologies and latest fab tools to high-tech industry trends.
-
So you guys cover MLID’s leaks and rumors as well ? He has a solid track record for being unreliable. I mean don’t get me wrong though, but Tech outlets need to stop taking him seriously.Reply
This leaked INTEL slide/roadmap might be an exception, but it's still too hard to digest any of his leaks. Remember Moore's Law is dead once confirmed a 10-core 7800x with dual 5-core chiplets ? That never panned out.
This slide looks obsolete and confusing to me though. Some of the core count info/config doesn't make much sense. There is TOO much segmentation here.
Before continuing any further, I want to make this clear that I don’t have any personal vendetta against MLID or any YouTuber, nor there is any “fanboyism” here either, since I don’t take side of any camp be it AMD, Nvidia, Intel, HP, DELL, APPLE or Microsoft.
Anyway, MLID also deletes videos when his predictions are laughably wrong. He can get anything wrong because if people are stupid to believe everything he says in the first place, he can just make another lie up explaining why he was wrong and they'll eat it up.
Take what he and everyone like him says, with a truck load of salt. The problem is people like him create unfounded hype and spread questionable information. Even Intel had to come out and address his Arc 780 leaks indirectly - that it never was planned and some people on the internet just took it too far.
He claimed RDNA 3 performance would be 60-80% over RDNA 2 , but we know what happened to it.
Raptor Lake CPU pricing ranges - got *all* ranges wrong just before the release (all CPUs from the lineup could be bought outside his predicted ranges). Released a damage control video claiming he was misunderstood. These are the facts none can deny based on MLID’s past leaks/rumors. -
Amdlova This intel chart is accurate as he'llReply
12700t can go higher than 200w
13500t can go higher than 160w
The CPU at 35w it's a dream lol -
baboma >So you guys cover MLID’s leaks and rumors as well ? He has a solid track record for being unreliable.Reply
Think of it like this: Which would you rather read, an (unreliable) rumor about CPUs, or more pieces on "butt-warming RPi" or "how to install a CPU"?
The bane of ad-supported sites is that they need eyeballs everyday, which means they need article postings everyday. In the Internet economy, quantity matters much more than quality. There isn't enough actual (PC) hardware news for everyday, year-round coverage anyway.
The consequence is that there are lots of filler pieces to get the requisite quota of eyeballs that corporate overlords (https://futureplc.com in case of THW) demand. This is what powers the large part of the Internet. Your (our) opinion on its quality doesn't really matter, since your only value is your eyeball, and your eyeball has already voted 'yes' on this piece as you've read it--and better, have commented on this, generating yet more eyeballs for the piece.
It is what it is. Think of the trade-off: It's free.
As for MLID, read it for entertainment. There's no need to take it or any rumor seriously. -
Geef It would be great if rumor stories like this had a tag at the top showing RUMOR or UNOFFICIAL information.Reply
A single sentence saying "This RUMOR came from HungLo who is known for reliable leaks of information." -
doughillman Metal Messiah. said:So you guys cover MLID’s leaks and rumors as well ? He has a solid track record for being unreliable. I mean don’t get me wrong though, but Tech outlets need to stop taking him seriously.
Yeah, I quit watching his bull loooong ago. He just spews crap out, guarantees that he's got good intelligence, and then never mentions it again when he's shown to have been completely wrong.
Is he occasionally correct? Sure. But any of us here could probably make up a buncha tech rumor <Mod Edit> outta thing air and SOME of it would likely pan out. You throw a million darts at the wall, you're bound to hit the bullseye now and then.
No self-respecting tech site should be basing stories on anything he's said. -
bit_user
100%. I can't comment specifically on MLiD, but the general phenomenon of these folks posting leaks on social media is just bad news. There seems to be very few negative consequences for them spreading bad info. All the incentives seem pointed in the direction of more sensationalism, which only leads to more bad info, not less.Metal Messiah. said:Take what he and everyone like him says, with a truck load of salt. The problem is people like him create unfounded hype and spread questionable information. -
bit_user
That's a false dichotomy. It presumes they must post something. In the newspaper industry, it used to be poisonous to your reputation to traffic in gossip and sensationalism.baboma said:Think of it like this: Which would you rather read, an (unreliable) rumor about CPUs, or more pieces on "butt-warming RPi" or "how to install a CPU"?
It's not as if there's no other tech news for them to cover. How about the Qualcomm/Nuvia/Arm litigation or Arm's recent introduction of the X4, A720, and A520 cores? With AI heating up, do you have any idea how well or poorly Nvidia's competitors are doing at gaining any marketshare? Why don't we get coverage of Hot Chips talks, like Anandtech used to do? I can't remember the last time I saw good coverage of networking products on here, either.
And then there's always more reviews and shootouts they could be doing. In particular, they have seemed to get away from thermal paste and heatsink shootouts, or their selection is very limited when they do.
So, I think the criticism of covering rumors from unreliable sources is fair. About 4-5 years ago, they had a policy against covering any rumors, period. That was going too far, but I think it's reasonable to expect them not to publish just anything and everything that comes their way.
I never thought substack was for me, but the folks at SemiAnalysis seem to be doing some good work.baboma said:It is what it is. Think of the trade-off: It's free.
https://www.semianalysis.com
I haven't subcribed to SemiAnalysis yet, but I did just setup a Patreon subscription for these guys:
https://chipsandcheese.com
It's like the best of the deep dives Anandtech used to publish, but even deeper and without any pretense of appealing to a mass audience.
That's letting them off the hook. It's too easy for people to get away with anything if you can just say "it's for entertainment". The fact is that people don't watch or read it for a good laugh - they read it because they want to be informed! That makes it news, whether or not it meets any standards of accuracy or bias.baboma said:As for MLID, read it for entertainment. There's no need to take it or any rumor seriously. -
doughillman said:Yeah, I quit watching his bull loooong ago. He just spews crap out, guarantees that he's got good intelligence, and then never mentions it again when he's shown to have been completely wrong.
Is he occasionally correct? Sure. But any of us here could probably make up a buncha tech rumor <Mod Edit> outta thing air and SOME of it would likely pan out. You throw a million darts at the wall, you're bound to hit the bullseye now and then.
No self-respecting tech site should be basing stories on anything he's said.
Couldn't agree more. Yeah, any one of us here can make a wild guess about the next-gen GPU or CPU architecture design and specs, based on current and past architecture specs and performance. The analogy of a Million darts is 100% correct.
It's not a very big deal to do this. But I gotta give the guy a credit for spewing out dangerous predictions, sometimes even without providing a solid backup. Lot of guts required to do this.
I guess this is what some YouTubers do these days to get more views and clicks, and kids and teens are already ready to jump on this YT bandwagon hype. So hey, why not make a prediction video !
bit_user said:That's letting them off the hook. It's too easy for people to get away with anything if you can just say "it's for entertainment". The fact is that people don't read it for a good laugh - they read it because they want to be informed! That makes it news, whether or not it meets any standards of accuracy or bias.
100% this ! -
TerryLaze
What is this pap-per thing you are talking about?!bit_user said:That's a false dichotomy. It presumes they must post something. In the newspaper industry, it used to be poisonous to your reputation to traffic in gossip and sensationalism.
Also gossip and sensationalism has been a best selling thing in reporting and later TV for since forever.
https://en.wikipedia.org/wiki/Yellow_journalism -
bit_user
That depends on what sort of audience you have or are trying to cultivate.TerryLaze said:Also gossip and sensationalism has been a best selling thing in reporting and later TV for since forever.
https://en.wikipedia.org/wiki/Yellow_journalism