Monday, December 6th 2021
Intel Prepares Raptor Lake Designs With 24 Cores and 32 Threads, More E-Cores This Time
With the launch of Intel's Alder Lake processors, Intel has switched from a homogeneous to a heterogeneous design of processors, where smaller, high-efficiency cores are mixed with high-performance cores to create a highly efficient and high-performance processor for all kinds of workloads. And it seems like Intel is not over with adding more E-cores to its future products, as the latest leaks suggest. According to the BAPCO's Crossmark benchmark database, Intel's upcoming Raptor Lake processors will feature more E-cores than the high-performance P-cores in the SoC design. As to why this design choice is present, we are not sure and don't have a definitive answer.
E-Cores are suitable for background tasks, and adding more would potentially leave space for P-cores to do heavier workloads. In the benchmark submission, which is now offline, the samples used were a configuration with eight P-cores and sixteen E-cores. Since the big cores are hyperthreaded, it makes up for a total composition of 24 cores with 32 threads. The platform "RPL-S ADP-S DDR5 UDIMM OC CRB" was used with DDR5-4800 memory, indicating an early stage engineering sample with a probably unfinished memory controller. The Raptor Lake generation will also use LGA 1700 socket, DDR5 memory and be present in the desktop and mobile sector once it launches in Q4 of 2022. It will also use Intel's 7 semiconductor manufacturing process, similar to Alder Lake. The only difference with the next-generation design is the updated Raptor Cove core design that brings a significant IPC uplift.
Sources:
Tom's Hardware, KOMACHI_ENSAKA (Twitter), via VideoCardz
E-Cores are suitable for background tasks, and adding more would potentially leave space for P-cores to do heavier workloads. In the benchmark submission, which is now offline, the samples used were a configuration with eight P-cores and sixteen E-cores. Since the big cores are hyperthreaded, it makes up for a total composition of 24 cores with 32 threads. The platform "RPL-S ADP-S DDR5 UDIMM OC CRB" was used with DDR5-4800 memory, indicating an early stage engineering sample with a probably unfinished memory controller. The Raptor Lake generation will also use LGA 1700 socket, DDR5 memory and be present in the desktop and mobile sector once it launches in Q4 of 2022. It will also use Intel's 7 semiconductor manufacturing process, similar to Alder Lake. The only difference with the next-generation design is the updated Raptor Cove core design that brings a significant IPC uplift.
81 Comments on Intel Prepares Raptor Lake Designs With 24 Cores and 32 Threads, More E-Cores This Time
Personally I'm not impressed with the e-cores in my 12700k. I'd rather have more P cores. I don't think the P cores use too much energy at reasonable clock speeds. In fact I was using all cores at 4.5ghz and only 120 watts. Plenty of room for 16 P cores at 250W. I don't find E cores compelling in a desktop. They only work well in the most ideal situations (like Cinebench). As long as E cores are cheap I suppose they make sense, but I'm not paying a lot for them. I don't think the i9 makes any sense. I would have bought a 12 P core i9 though. Not an 8.
In the meantime, my rocket lake system is good enough for my gaming needs. Added benefit of maturing bios for my mobo & I'm finding even more OC headroom for the ram kits I have atm + the ease with which Intel extreme tuning app in windows for rocket lake is a great piece of software imo. The IMC on rocket lake is fantastic if tuned well!
2nd gen DDR5 IMC with raptor lake should be something certainly nice to look forward too in combo with the increase in ipc of its P-cores. :)
As to how much power the chip consumes, it really depends on the workload. I believe if you really push the cores hard, the power consumption will increase drastically, especially when you allow the chip to run infinite PL2 under load. I suspect 120W applies when you leave it to the Intel's recommended boost timing.
As demonstrated in the Xeon e5 v4 series.
And exactly the reason why they went for mesh architecture.
So they've made this P/E-core config to get "16 cores" within the architectural limit.
I've tested my max overclock on the 12600k at 5.4 ghz and 47 ring with e-cores off and my 24/7 5.3 ghz with 43 ring with e-cores on (overclocked to 4.3ghz) and having them on very noticeably eliminates intermittent stutters in cyberpunk and far cry 6. Could be just my setup but they seem to really work (especially because I'm too lazy to shut down all my background stuff).
Also I get 90% of 12700K multithread perf at ~186W which is pretty nice - not something 8 P cores by themselves can do afaik. 12900K with 10P cores would probably get lower multithreaded performance than current 12900k 8p/8e for 100W more draw, and 0 benefit in virtually any current real-world application.This is probably true. Stacked cache looks insane.
Hoping they start leap-frogging each other and we get sweet gear for cheap.
If not for AMD coming up would still be sitting on quad cores (maybe 6 cores on the i7), and if not for Intel coming back the 6600X would have been a 6 core zen 4 that cost $450.
Increase in the quantity of the stronger cores either way would make likely a bigger impact on performance.
Still, I would pick pure P-core config all-day.
AMD need to give them a good kicking with all-performance cores then Intel will stop with these games.
24 Zen4 cores up against 8 Intel P cores, no chance. And we all know that AMD could put 32 cores on the desktop if it's needed.
It's also confirmation that Intel can't use more than 8 P cores, due to thermal limits. Oops Intel, looks like your new architecture is a hot bust.
I had to laugh when I saw LTT's Intel 12900 shill chiller video, that CPU was sucking down well over 420 Watts, and was still thermal throttling during the benchmark, whilst using a 3KW Aircon unit to cool it! They only got it to 5.3GHz all P core OC. The crappy E cores were at 4.2GHz I believe. Yes it performs well, but when you look at it, it's an architectural dead end, unless they can get it on a more advanced production node. Too hot, to power hungry.
Speaking about production applications.
Their CPU for production applications a.k.a Sapphire Rapids retains a pure P-core design.
I don't think intel's p+e-core design is ready for production applications right now.
Since they are not going to do it in the server market , hence lack of confidence / lack of optimization.
I won't put any risk running my production software on beta testing hardware and software, which this AlderLake +DDR5+WIN11 combo really is.