Intel "Bartlett Lake" Appears as a P/E-Core Hybrid, P-Core Only CPUs Could Soon Follow

[AusWolf]

Well although I'm not personally interested on a non ecore intel cpu, I can see that a lot of people would be highly interested simply because there is no other cpu like it. There is an alarming absence of a chip with over 8 big cores that aren't separated by an ocean. Literally, there are none. You either have to go with ecores or split CCDs

Exactly. If Intel had released the pure 12 p-core variant before I got my 7800X3D, I would have jumped on one.

 

[Dirt Chip]

12P with HT or it was axe here also?

 

[R0H1T]

I would have jumped on one.

Well Intel's had yield issues for better part of a decade now so I seriously doubt that.

Their priority is still server, like AMD, then mobile chips & finally scraps for DIY

 

[Nater]

Hopefully they clock them higher or add more cache to make them worth getting. It's kind of a lateral move. I'm sitting on 12600K's and 13600K's...don't really get anything moving to more cores, but it'd be nice to have.

My apps are all mostly single threaded. (CAD/CAM)

 

[GhostRyder]

I mean, they essentially are backtracking on this design a bit on the high end. Not sure how I overall feel about it, would like to see how these chips perform not only in performance, but also in power usage, heat, etc.

 

[Vayra86]

If that's still 10nm / "Intel 7", it's embarrassing.

Well if you consider the cost of smaller nodes, a big one might not be all bad. If the performance is good and perf/W is the same.... you'd be crazy not to bake on cheaper sand. You can also make larger chips at lower cost that way.

 

[wNotyarD]

Well if you consider the cost of smaller nodes, a big one might not be all bad. If the performance is good and perf/W is the same.... you'd be crazy not to bake on cheaper sand. You can also make larger chips at lower cost that way.

Let's just hope BTL doesn't have the same degradation issues that plague RPL.

 

[kondamin]

I mean, they essentially are backtracking on this design a bit on the high end. Not sure how I overall feel about it, would like to see how these chips perform not only in performance, but also in power usage, heat, etc.

AFAIK it's not meant for the wider market more embedded/edge

Let's just hope BTL doesn't have the same degradation issues that plague RPL.

they look like regular parts, not K parts meant for OC

 

[sudothelinuxwizard]

very meh launch. What exactly differentiates this from just, say, a 14900k?

 

[wNotyarD]

very meh launch. What exactly differentiates this from just, say, a 14900k?

No scheduling issues due to heterogeneous architecture (and hopefully no self-destructing). Possibly better thermals from not having that boatload of E-cores onboard.

 

[Wirko]

Could the 12-core part be reused silicon from some Rapids server chips?

A single 400 mm² die of the four-die big Sapphire Rapids processor would be a candidate... almost. It has 15 cores, DDR5 and PCIe5 controllers and everything. But Intel never mentioned that a single die can be made to work, and also the cores would be Golden Cove, not even Raptor Cove.

 

[JustBenching]

very meh launch. What exactly differentiates this from just, say, a 14900k?

Probably faster for gaming due to having 12 beefy cores, but that only applies to core heavy games I guess. Other than that, yeah well, not much.

 

[Ravenmaster]

Not gonna lie, I would totally buy a Core 9 with 12 P-cores. I still have a Z790 with a 12900K installed in it. Would be a nice little upgrade to my secondary rig.

 

[chrcoluk]

My only question is, why now? I would have been interested in getting one in LGA-1700's heydays, but now is a bit too late.

LGA1700 still intel's best socket/platform.
AMD did same kept releasing for old platform after new launched.

I think I might grab the 12 p core version myself.

 

[HisDivineOrder]

Intel's just cribbing off AMD's old playbook when they were behind. They know Arrow Lake is a dead end and know that anyone still on Alder Lake is looking for an upgrade, so 12P cores with a great clockspeed could be very tempting to me rather than a CPU+MB+Mem upgrade I'd have to do otherwise.

I saw rumors like this last year, but I figured they'd changed their minds. I'd be tempted by a well priced 12 Pcore part with great clockspeeds. My 12700k is already pretty great. A bit more oomph for a lot less money than a 9800x3D would be convenient.

 

[Cheeseball]

Perhaps they could upgrade the UHD 770 to a Xe² from LNL? Though that might be asking too much, this design was likely floorplanned and built alongside the 8+16 chips. If anything, I am buying the KF model this time anyway... unless it's the same price, I mean. My motherboard does not support graphics.

They should do this. The Xe² Arc cores along with the updated QuickSync (Media Engine) are so much better than what it was in the previous gens. Even if it will just be 4 Xe-cores that would still be better than the UHD770's 32 EUs and Iris Xe's (mobiles) 96 EUs.

 

[sudothelinuxwizard]

Probably faster for gaming due to having 12 beefy cores, but that only applies to core heavy games I guess. Other than that, yeah well, not much.

Sure, but if it's meant for embedded I doubt any gamers who don't also happen to work at say, an ATM manufacturer who uses them and steals them (which you should not do of course) will be able to use them until they start getting retired. As a non gamer myself it seems like a straight downgrade.

 

[JustBenching]

Sure, but if it's meant for embedded I doubt any gamers who don't also happen to work at say, an ATM manufacturer who uses them and steals them (which you should not do of course) will be able to use them until they start getting retired. As a non gamer myself it seems like a straight downgrade.

They won't be embedded. Read the whole article..
If you are a non gamer, yeah nothing to see here. 13/14900k will still be faster.

 

[AusWolf]

LGA1700 still intel's best socket/platform.
AMD did same kept releasing for old platform after new launched.

I think I might grab the 12 p core version myself.

I would love to as well, but having a 7800X3D, I believe a GPU upgrade would make more sense, and I don't have the spare cash for both.

 

[Onyx Turbine]

Intel's just cribbing off AMD's old playbook when they were behind. They know Arrow Lake is a dead end and know that anyone still on Alder Lake is looking for an upgrade, so 12P cores with a great clockspeed could be very tempting to me rather than a CPU+MB+Mem upgrade I'd have to do otherwise.

I saw rumors like this last year, but I figured they'd changed their minds. I'd be tempted by a well priced 12 Pcore part with great clockspeeds. My 12700k is already pretty great. A bit more oomph for a lot less money than a 9800x3D would be convenient.

instead of the 12p i hope they make it an deca p core with 2 durable advanced e cores. The p cores should have an insane high boost clock and run on optimal gaming base clock,
like a 12700k. Its multicore performance should outperform the 14900KS, when running in idle < 10 Watts.
If its a 12p core it could become to good of an cpu canabilising the arrow lake line or after, as such by 10p still coming in great with benchmarks somewhere even larger than what is released
than a 265k or even 285k would be it.

 

[Dr. Dro]

instead of the 12p i hope they make it an deca p core with 2 durable advanced e cores. The p cores should have an insane high boost clock and run on optimal gaming base clock,
like a 12700k. Its multicore performance should outperform the 14900KS, when running in idle < 10 Watts.
If its a 12p core it could become to good of an cpu canabilising the arrow lake line or after, as such by 10p still coming in great with benchmarks somewhere even larger than what is released
than a 265k or even 285k would be it.

Realistically speaking, a 12P+0E design will be a regression in multithreaded performance against the existing 8P+16E processors in controlled, benchmarked environments. However, the extra high performance execution units should prove very useful in many more real-world scenarios, without the drawbacks of the 12-core Ryzen processors' split CCDs or their X3D counterparts' imbalanced hardware configuration.

E-cores also come in clusters of 4, which roughly take the same die area as one P-core. I believe you wanted something like the Royal Core design, where there are P-cores and a couple of super big cores. This was supposed to be Beast Lake and to be marketed as Core Ultra series 4 (after Panther Lake/series 3, which would be Arrow Lake's refinement still on LGA 1851), but it was unfortunately canned last year.

 

[ratirt]

My only question is, why now? I would have been interested in getting one in LGA-1700's heydays, but now is a bit too late.

From all that Intel was saying and did regarding the hybrid. All the praise around the ecores and now, Pcore only which is not surprising and it is at the same time. Pcores are faster no doubt about it and it does not matter how efficient ecores are. As you know, 13th and 14th gen were not efficient in the top performance market. They did not deliver highest performance and they were still power hungry. I assume, considering what AMD has done, it kind of force Intel to reach for the performance crown again and forget about efficiency. Hybrid did not work they way Intel planned so why bother?
I wonder now, will AMD still go with the Hybrid in the top tier desktop market or this efficiency cores will only be for laptops.

If Intel decides to get this one released for current boards, Intel is pulling all the stops. I think, that would have been brilliant. I'm not in the market for a CPU nor boards etc. but i would definitely look closer to Intel 12c core if I will be changing my rig in a year or 2. Still think AMD would win since if Intel does release it for the current boards, these CPUs would have been more appealing for those who have alder 13th 14th gen. AMD is not going to resign from the AM5 support at that time. Either way, huge disruption which would be awesome from a consumer standpoint.

Realistically speaking, a 12P+0E design will be a regression in multithreaded performance against the existing 8P+16E processors in controlled, benchmarked environments. However, the extra high performance execution units should prove very useful in many more real-world scenarios, without the drawbacks of the 12-core Ryzen processors' split CCDs or their X3D counterparts' imbalanced hardware configuration.

E-cores also come in clusters of 4, which roughly take the same die area as one P-core. I believe you wanted something like the Royal Core design, where there are P-cores and a couple of super big cores. This was supposed to be Beast Lake and to be marketed as Core Ultra series 4 (after Panther Lake/series 3, which would be Arrow Lake's refinement still on LGA 1851), but it was unfortunately canned last year.

I think Intel is reaching our for gamers and the performance crown again. Take what was Intel's for so long before AMD took it.

 

[chrcoluk]

I would love to as well, but having a 7800X3D, I believe a GPU upgrade would make more sense, and I don't have the spare cash for both.

Yep of course, for me I already have the board so will just take advantage of it.

 

[JustBenching]

From all that Intel was saying and did regarding the hybrid. All the praise around the ecores and now, Pcore only which is not surprising and it is at the same time. Pcores are faster no doubt about it and it does not matter how efficient ecores are. As you know, 13th and 14th gen were not efficient in the top performance market. They did not deliver highest performance and they were still power hungry. I assume, considering what AMD has done, it kind of force Intel to reach for the performance crown again and forget about efficiency. Hybrid did not work they way Intel planned so why bother?

What are you even talking about? The 12P core will be much slower than their normal 13/14900k and the core ultra. Are you suggesting that Intel is releasing a SLOWER CPU to retake the performance crown? Does that make sense to you man?


Efficiency was never the issue on 13 14900. Limit them or get a locked 65 or 35w tdp model Most efficient chips out of the box, they probably still are. I know we like to clown on Intel and sure you can have your fun but some times we need to accept reality for what it is.

 

[AusWolf]

From all that Intel was saying and did regarding the hybrid. All the praise around the ecores and now, Pcore only which is not surprising and it is at the same time. Pcores are faster no doubt about it and it does not matter how efficient ecores are. As you know, 13th and 14th gen were not efficient in the top performance market. They did not deliver highest performance and they were still power hungry. I assume, considering what AMD has done, it kind of force Intel to reach for the performance crown again and forget about efficiency. Hybrid did not work they way Intel planned so why bother?
I wonder now, will AMD still go with the Hybrid in the top tier desktop market or this efficiency cores will only be for laptops.

That's because e-cores aren't efficient with power, only with die area. One p-core takes up the same space as a cluster of 4 e-cores, so e-cores are a nice way to increase core count and MT performance, but they throw power economy out of the window. I have a feeling that the 12 p-core design will consume less power than Raptor Lake does, unless Intel decides to overvolt it to the moon.