中文網
  • 繁體
  • 简体
    •
    Trending
    CONNECT WITH US
    NEWS TAGGED BANDWIDTH
    Wednesday 1 April 2026
    Micron reportedly developing stacked GDDR to address shifting AI memory demand

    Micron is reportedly developing vertically stacked graphics DRAM (GDDR) products to address shifting AI memory demand, according to...

    Monday 30 March 2026
    AI compute shifts to inference, reshaping data center bottlenecks

    The focus of artificial intelligence computing is set to shift from training to inference beyond 2025, a transition that will also redefine...

    Tuesday 24 March 2026
    Huawei's Ascend 950PR debuts with nearly 3x H20 performance, targets China's AI compute gap
    Huawei has officially launched its Ascend 950PR processor and introduced the Atlas 350 AI accelerator card equipped with this chip, marking the commercial debut of its next-generation...
    Monday 16 March 2026
    HyperLight, UMC, and Wavetek form foundry partnership for high-volume TFLN chiplet production
    HyperLight Corporation, United Microelectronics Corporation (UMC), and Wavetek Microelectronics Corporation, a wholly owned UMC subsidiary, have announced a strategic manufacturing...
    Thursday 12 March 2026
    UMC, HyperLight to mass-produce TFLN photonics chiplets for AI data centers
    United Microelectronics Corporation (UMC) has partnered with startup HyperLight to mass-produce thin-film lithium niobate (TFLN) chiplets, aiming to improve data transmission speeds...
    Thursday 5 March 2026
    Micron unveils 256GB SOCAMM2, scaling AI server memory to 2TB per CPU

    Artificial intelligence (AI) infrastructure is evolving faster than traditional server hardware cycles can accommodate. As AI workloads...

    Thursday 5 March 2026
    Commentary: HBM hybrid bonding race heats up, but memory makers slow to adopt
    Competition in the high bandwidth memory (HBM) market is intensifying, prompting equipment makers such as Hanwha Semitech and Semes to accelerate...
    Tuesday 3 March 2026
    FIT unveils 102.4T CPO external laser solution for AI data centers at OFC 2026
    As AI model training scales up and cloud service providers (CSPs) accelerate the buildout of high-performance data centers, switch bandwidth is moving from 51.2Tbps to the 102.4Tbps...
    Monday 2 March 2026
    Commentary: Nvidia sparks silicon photonics race, yet copper still anchors data centers?
    As data center computing demand expands, high-speed transmission architectures are under pressure to upgrade. Scaling high-performance computing platforms is exposing bandwidth and...
    Monday 2 March 2026
    Nvidia's Vera Rubin platform faces HBM4, cooling, and software hurdles ahead of ramp
    Nvidia's next-generation Vera Rubin platform has moved from public unveiling to early customer sampling, with the company projecting a broader production ramp later this year. Both...
    Wednesday 18 February 2026
    Samsung advances LPDDR5X-PIM, signaling memory's new role in AI computing

    Artificial intelligence is shifting the center of gravity in semiconductors. For decades, processors defined performance. Now memory bandwidth,...

    Friday 13 February 2026
    Citi flags post-HBM shift as edge memory, HBF advance
    A Citi analyst said the next phase of artificial intelligence (AI) memory demand will extend beyond high-bandwidth memory (HBM), pointing to emerging opportunities in physical AI devices...
    Friday 13 February 2026
    SEMICON Korea: Samsung pushes hybrid bonding HBM, SK Hynix bets on platform AI
    SEMICON Korea 2026, held in Seoul from February 11–13, brought senior executives from Samsung Electronics and SK Hynix to outline their AI-era memory R&D strategies. Samsung...
    Saturday 7 February 2026
    Microsoft's Maia 200 chip targets AI inference cost advantage, not Nvidia rivalry
    Microsoft recently unveiled its second-generation custom AI chip, Maia 200, more than two years after its previous generation. The Maia chip's update cycle lags behind the rapid iterations...
    Tuesday 3 February 2026
    SK Hynix achieves major breakthrough in HBM4 quality tests, eyes stable supply for Nvidia Rubin
    SK Hynix has reportedly made significant progress in the quality testing of Nvidia's sixth-generation high-bandwidth memory (HBM4) products, boosting the likelihood that it will become...
    1/20 pages
    Research Report Database
    BIZ FOCUS
    Feb 9, 08:00
    iCatch and DXOMARK Establish Taiwan's First Next Generation Imaging Laboratory
    Friday 6 February 2026
    2026 Market Outlook: Key Opportunities & Risks to Watch
    Thursday 5 February 2026
    Attopsemi Scales I-fuse Technology to 7nm following 12nm Silicon Success
    Wednesday 4 February 2026
    Teaching Machines to Feel the World TOUCH Lab Redefines Sensing
    marketing
    RESEARCH INSIGHTS
    Weekly research roundup: Qualcomm's AI strategy, memory giants and updated 2026 smartphone outlook