New 3D-stacked memory tech seeks to dethrone HBM for AI inference — d-Matrix claims 3DIMC will be 10x faster and 10x more efficient
Memory startup d-Matrix is claiming its 3D stacked memory will be up to 10x faster and run at up to 10x greater speeds than HBM. d-Matrix's 3D digital in-memory compute (3DIMC) technology is the company's solution for a memory type purpose-built for AI inference.
China gets Intel Xeon 6P Granite Rapids CPUs repackaged with added security and surveillance features — Jintide C6P CPUs with up to 86 P-cores
Montage Technology, one of China's many fabless semiconductor companies, has launched the new Jintide C6P series of server processors. The C6P chips, based on Intel's Granite Rapids platform with additional security and tracking features, are designed explicitly for the Chinese market.
New Ryzen 9000X3D CPU could deliver EPYC levels of game-boosting L3 cache — rumored chip reportedly sports 16 Zen 5 cores, 192MB L3 cache, 200W TDP
AMD already sells some of the best CPUs on the retail market. However, hardware leaker chi11eddog claims that the chipmaker might capitalize on the success of the Ryzen 9000X3D series and introduce two more models to the already diverse family.
Intel Core Ultra Series 3 CPUs could finally answer AMD’s V-Cache — Nova Lake could boast massive 144MB L3
Intel is rumored to be introducing a new large L3 cache pool that will rival AMD's 3D V-Cache chips with its next-gen Core Ultra (likely Core 400, if Core 300 is a refresh of Arrow Lake) series CPUs. Leakers Haze and Raichu on X believe this new last-level L3 cache will feature a 144MB capacity.
G.Skill introduces new high-capacity, high-speed kits — now offers 256GB (64GBx4) DDR5-6000 and 128GB (64GBx2) DDR5-6400 modules
Memory manufacturer G.Skill just announced two new high-capacity memory modules overclocked to DDR5-6000 CL32 and DDR5-6400 CL36. Previously, high-performance memory kits were limited to smaller modules.
AMD reveals benchmarks of Ryzen Threadripper 9000 — claims it’s up to 145% faster than rival Xeon in some tests
Although AMD formally introduced its Ryzen Threadripper 9000-series processors at Computex and disclosed their specifications, the company did not reveal their performance compared to predecessors and rivals, or their pricing.
Intel and SoftBank collaborate on power-efficient HBM substitute for AI data centers, says report
American chip giant Intel has partnered with Japanese tech and investment powerhouse SoftBank to build a stacked DRAM substitute for HBM. According to Nikkei Asia, the two industry behemoths set up Saimemory to build a prototype based on Intel technology and patents from Japanese academia, including the University of Tokyo.
TSMC reiterates it doesn’t need High-NA EUV for 1.4nm-class process technology
TSMC reiterated its long-known stance on next-generation High-NA EUV lithography tools at its European Technology Symposium in Amsterdam. The company does not require these highest-end lithography systems for its next-generation process technologies, including A16 (1.6nm-class) and A14 (1.4nm-class) process technologies.
Intel officially cuts Core Ultra 7 200-series desktop CPU prices by up to 25%
On Tuesday, Intel notified Tom's Hardware that it officially cut the suggested retail price (SRPs) of its boxed Core Ultra 7 200-series processors for desktops by around $100. It is common for CPU developers to cut the prices of their products, but it is uncommon for companies like Intel to formally confirm such moves. However, there is a small catch: since we are dealing with SRPs, actual prices may vary.
AMD 16-core Zen 5c die shots show long, narrow CCX, all 16 cores sharing a single L3 cache
Die shots of AMD's 16-core Zen 5c CCD used in its latest EPYC 9005 series server processors have been exposed, revealing clear differences compared to AMD's previous-generation Zen 4c CCDs. Posted by HXL on X, photos of AMD's 16-core 3nm Zen 5c CCD expose a long row of two banks of cores flanking a 32MB layer of L3 cache in the middle of the die.
