You won’t typically find a Hygon processor among the list of best CPUs for gaming or other applications. Nevertheless, Hygon stands as one of China’s most important fabless semiconductor companies. Hardware investigator HXL recently shared an ambitious roadmap detailing Hygon’s forthcoming initiatives, including a flagship 128-core, 512-thread processor.

Arrow Lake might have some fuel left in the tank, as Intel’s latest IPO (Intel Performance Optimizations) tech is reportedly hitting pre-builts in China, per UNIKO’s Hardware at X. Intel recently unveiled the IPO program in China. It is supposedly a suite of tuned settings or profiles that balance overclocking and stock settings. It is important to note that Intel has not detailed or announced an IPO for the global market.

AMD’s Zen 5 architecture has officially entered the embedded market with the introduction of the EPYC Embedded 9005 series. The new chips take advantage of AMD’s compact Zen 5c architecture, which is optimized for density-focused computing and edge tasks. AMD is offering a wide gamut of SKUs to serve as many customers as possible, similar to previous generations. Core counts range from “entry-level” models offering just 8 CPU cores, all the way to its flagship Embedded 9965, offering 192 cores.

Preliminary benchmarks at PassMark put Intel’s new Arrow Lake-based Core Ultra 9 275HX 7% ahead of AMD’s Ryzen 9 7945HX3D processor in CPU Mark. Moreover, despite a 400 MHz clock deficit, the 275HX manages to land 34% faster than last generation’s i9-14900HX, with a solid 9% lead when looking at single-threaded performance.

G.Skill has announced new DDR5-6400 CL30 RAM kits that combine high memory capacities with fast speeds and low latency to compete with the best RAM. The kits, launched under the Trident Z5 RGB and Trident Z5 Royal series, will feature 96GB (2x48GB) capacities, with smaller 64GB and 48GB variants also in the works. The DRAM manufacturer promises DDR5-6400 CL30 specs using Intel XMP 3.0 and has also showcased how the memory performs on an equivalent AMD Ryzen 9000 system.

New e-mails from Huawei engineers regarding Linux kernel development suggest that HiSilicon – Huawei’s chip subsidiary – is prepping a new Kunpeng SoC with HBM (High Bandwidth Memory) technology – as highlighted by Phoronix. This will likely be the the first significant release from HiSilicon in a while but don’t get your hopes up as it could also just be a rebranded older model with a slight pinch of HBM.

Some purported specifications of Intel’s upcoming Core Ultra 200U or Arrow Lake-U mobile processors have been surfaced by Jaykihn at X. It turns out that despite the “Ultra” moniker; the Core Ultra 200U CPUs will utilize an updated version of the Meteor Lake architecture – reportedly using the Intel 3 node.

TeamGroup has introduced its family of clock-unbuffered memory modules (CUDIMMs) that promise to significantly enhance memory performance on Intel’s latest Core Ultra 200S-series platforms. TeamGroup’s flagship T-Force Xtreem CKD DDR5 modules boast a data transfer of DDR5-9000 and overclocking potential up to DDR5-9600.

Golden Pig Upgrade Pack, a well-known leaker, has published preliminary details about the potential configurations of AMD’s Ryzen AI Max 300 (codenamed Strix Halo) processors, which are expected to be released in early 2025. In addition, the leaker revealed the official name of these CPUs with an ultra-high-end integrated Radeon GPU.

Hardware leakers Jaykihn and Harukaze5719 have reportedly leaked the power profiles for Intel’s Arrow Lake and Panther Lake processors. These next-generation chips will arrive to rival the best CPUs on the market. The leak details Intel’s Baseline, Performance, and Extreme power profiles for five distinct variants of Arrow Lake-S processors. The core configurations and TDPs include: 8+16 at 125W, 8+12 at 125W, 6+8 at 125W, 6+8 at 65W, and 6+4 at 65W. The former number depicts the number of P-cores, while the latter depicts the chip’s number of E-cores.