Korean memory giant Samsung has been working hard to prepare for supplying HBM3E to NVIDIA, but now it appears almost impossible to supply to NVIDIA in 2024. According to Wccftech's report, Samsung failed to secure a position in NVIDIA's supply chain because it couldn't meet NVIDIA's qualification t...

Marvell announces new custom HBM compute architecture that enables XPUs to achieve greater compute and memory density. According to its press release, this new AI accelerator (XPU) architecture enables up to 25% more compute, 33% greater memory while improving power efficiency. Notably, the new c...

Following the footsteps of semiconductor bellwethers TSMC and Intel, U.S. head-quartered Micron becomes the first memory giant to secure the funding under the CHIPS Act. According to U.S. Department of Commerce, it has finalized a USD 6.165 billion government subsidy for Micron to support semiconduc...

According to a report from Economic Daily News, at the launch event for the second volume of his autobiography yesterday, TSMC founder Morris Chang shared insights about TSMC's two major rivals, Intel and Samsung. Chang once referred to these two competitors as “700-pound gorillas.” However, bot...

Though still working towards the last mile on its 12hi HBM3e qualification with NVIDIA, Samsung seems to be advancing steadily on its HBM4 roadmap. According to South Korean media outlet ZDNet, Samsung is gearing up for mass production of 1c DRAM, a critical component that will determine the company...