See how the DRAM memory shortage links to HBM-first focus and consumer supply cuts, helping you avoid overpaying and time ...

Memory makers just can't churn out their DRAM fast enough. On the heels of an AI-driven shortage, SK Hynix on Tuesday announced a new 19 trillion Korean won (about $13 billion) advanced packaging and ...

TL;DR: SK hynix has improved its 1c DRAM yields from 60% to over 80%, focusing on HBM for AI GPUs. The company developed the first 1c process-based 16GB DDR5 DRAM and will lead in mass-producing HBM4 ...

HBM4 shifts from component upgrade to system driver in Nvidia and Samsung’s Rubin strategy for large-scale AI compute ...

TL;DR: NVIDIA is transitioning to SK hynix GDDR7 memory modules for its GeForce RTX 50 series GPUs, starting with the RTX 5070, moving away from Samsung. SK hynix, known for its HBM technology, is ...

SEOUL (Reuters) -The global rush by chipmakers to produce AI chips is tightening supply of less glamorous chips used in smartphones, computers and servers, spurring panic buying by some customers and ...

Micron is one of the three market-defining providers of HBM modules, and is gaining market share against South Korean rivals ...

Nvidia has reportedly tapped Micron Technology as the first supplier of its next-generation small outline compression attached memory module (SOCAMM) chips, putting the US-based memory maker ahead of ...

Micron Technology, Inc.’s stock is up 180% YTD – or 4X more YTD than AI heavyweight Nvidia. MU is expanding its presence within high bandwidth memory, or HBM, stating in Q4 that it has expanded its ...

A new KAIST roadmap reveals HBM8-powered GPUs could consume more than 15kW per module by 2035, pushing current infrastructure, cooling systems, and power grids to breaking point. The next generation ...