Intel's next-generation desktop processors codenamed Arrow Lake-S will not increase core count compared to existing 14th Generation Core 'Raptor Lake Refresh' processors, according to excerpts from an alleged Intel document published by a renowned hardware leaker @YuuKi_AnS (who removed the post, but VideoCardz has re-published them). Surprisingly, these documents point to Intel's Hyperthreading being absent from the new processors, perhaps confirming long-running rumors. This source has an impeccable record with leaked Intel documentation but be aware that these specifications could be subject to change as they outline pre-Alpha silicon.
The documents indicate that Intel is set to offer three types of silicon within its Arrow Lake-S lineup for desktops (in LGA-1851 form factor) featuring high-performance (P) and efficiency (E) cores, including three die configurations: 8P + 16E cores, 6P + 16E cores, and 6P + 8E cores.
Assuming that the information is accurate, Intel's next-generation desktop CPUs will not increase the core count compared to Raptor Lake Refresh-S CPUs but will also come without hyper-threaded P-Cores, a staple with Intel's existing processor lineups. There have long been rumors that Intel will move to a new approach with its P-Cores that discards hyperthreading, which allows two threads to run on a single CPU core, for a new approach that it has outlined in a patent. This listing will surely add fuel to those predictions, but this could merely be due to this being extremely early pre-Alpha silicon. It's best to take this with a grain of salt; the documents point to the P-Cores being disabled in the BIOS due to initialization issues, so anything is possible.
The Arrow Lake processors are set to use a new microarchitecture and faster DDR5-6400 memory. Notably, this series will also end Intel's support for DDR4 memory, meaning that both of the world's largest x86 chipmakers will be all-in with DDR5.
When we talk about Intel's Arrow Lake products, we should remember the process technology that Intel is set to use to make these products: the Intel 20A process technology (2nm-class). This fabrication node introduces RibbonFET gate-all-around transistors and the PowerVia backside power delivery network. Both innovations can enable Intel to optimize power consumption, increase performance, and boost transistor density. How this will affect the performance of socketed Intel's Arrow Lake processors remains to be seen. Still, the new CPUs will clearly have some advantages over existing products due to the innovations brought by the process node.
Also, the next-generation desktop CPU is set to support 16 PCIe lanes for graphics cards and two x4 lanes for solid-state drives. Meanwhile, the whole platform supports a DisplayPort output at a UHB20 rate and Thunderbolt 4 connectors, assuming that the motherboard has an appropriate Hayden Bridge retimer (which will not be cheap).
As far as the chipset side goes, Intel's 800-series platform controller hub supports almost everything one can imagine, including NVMe, SATA, PCIe, and USB 3.2 Gen2x2.
It's noteworthy that we don't know whether excerpts from the leaked document are up-to-date or outdated, but with Arrow Lake moving closer to market, we expect to learn more in the coming months.
