Entry-Level GPUs Are In Trouble As DRAM Prices Go Nuclear
DRAM has evolved from a boring background commodity to the part that dictates who even bothers to build affordable GPUs. AI data centers are consuming […]
GDDR7
NVIDIA Rubin CPX Explained: Disaggregated Inference And The Cost Of Million-Token Context
NVIDIA’s Rubin platform splits long-context prefill from token decode. Rubin CPX handles the compute-heavy front half, standard Rubin handles bandwidth-heavy generation. The NVL144 CPX rack […]
SK hynix’s AI memory roadmap: HBM5, AI DRAM, AI NAND and what it means for GPUs
Fresh off record quarterly profits, SK hynix has laid out a long-range memory roadmap that doubles down on HBM and AI-oriented DRAM and NAND. At […]
NVIDIA’s first U.S. Blackwell wafer: signal or showpiece?
NVIDIA and TSMC stood on a stage in Arizona and showed off something real: the first U.S.-made Blackwell wafer. On the surface, that’s a headline […]
Micron Says HBM4 Samples Are Hitting 2.8 TB/s; GDDR7 at 40 Gbps
Micron is openly talking HBM4 now—and the numbers are punchy: over 2.8 TB/s per stack at pin speeds north of 11 Gb/s, plus GDDR7 binned to 40 Gb/s. […]