<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>半导体制造 on Deep Research</title>
    <link>https://dailydigest.aabot.us/zh/tags/%E5%8D%8A%E5%AF%BC%E4%BD%93%E5%88%B6%E9%80%A0/</link>
    <description>Recent content in 半导体制造 on Deep Research</description>
    <generator>Hugo</generator>
    <language>zh-CN</language>
    <lastBuildDate>Sun, 10 May 2026 04:00:00 +0000</lastBuildDate>
    <atom:link href="https://dailydigest.aabot.us/zh/tags/%E5%8D%8A%E5%AF%BC%E4%BD%93%E5%88%B6%E9%80%A0/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>500亿美元的豪赌：5μm间距TSV技术如何决定AI硬件的未来</title>
      <link>https://dailydigest.aabot.us/zh/posts/2026-05-10-through-silicon-via-technology-at-5%CE%BCm-pitch-enabling-1000-layer-3d-chip-stacking-for-ai-accelerators/</link>
      <pubDate>Sun, 10 May 2026 04:00:00 +0000</pubDate>
      <guid>https://dailydigest.aabot.us/zh/posts/2026-05-10-through-silicon-via-technology-at-5%CE%BCm-pitch-enabling-1000-layer-3d-chip-stacking-for-ai-accelerators/</guid>
      <description>硅通孔（TSV）技术已实现了出色的5μm间距缩放，能够为AI加速器提供千层3D芯片堆叠，然而500亿美元的产业投资成败不仅取决于技术突破，更在于如何应对残酷的经济现实：台积电相比三星的70%良率优势，英特尔200亿美元亚利桑那州晶圆厂需要75%的成本削减，以及决定堆叠芯片是自我燃烧还是革命性计算的热管理解决方案。</description>
    </item>
    <item>
      <title>1000倍性能的承诺：为什么模拟AI加速器在实验室表现卓越，却难以进入你的手机</title>
      <link>https://dailydigest.aabot.us/zh/posts/2026-05-05-the-1000x-promise-why-analog-ai-accelerators-work-brilliantly-in-labs-but-struggle-reaching-your-phone/</link>
      <pubDate>Tue, 05 May 2026 04:00:00 +0000</pubDate>
      <guid>https://dailydigest.aabot.us/zh/posts/2026-05-05-the-1000x-promise-why-analog-ai-accelerators-work-brilliantly-in-labs-but-struggle-reaching-your-phone/</guid>
      <description>IBM的模拟AI芯片在实验室演示中比数字处理器实现了1000倍的能效提升，以飞焦级精度处理语音识别任务。然而，尽管在物理突破和技术优势方面得到证实，这些革命性加速器面临着现实鸿沟：制造成本、软件兼容性障碍以及基础设施要求，这解释了为什么你的下一部智能手机很可能不会包含模拟AI——无论研究结果看起来多么令人印象深刻。</description>
    </item>
    <item>
      <title>百亿美元博弈：三星与台积电的2nm竞赛如何取决于制造现实而非物理原理</title>
      <link>https://dailydigest.aabot.us/zh/posts/2026-05-03-gate-all-around-transistors-at-2nm-samsung-and-tsmcs-race-to-replace-finfets-with-nanosheets/</link>
      <pubDate>Sun, 03 May 2026 04:00:00 +0000</pubDate>
      <guid>https://dailydigest.aabot.us/zh/posts/2026-05-03-gate-all-around-transistors-at-2nm-samsung-and-tsmcs-race-to-replace-finfets-with-nanosheets/</guid>
      <description>三星的2nm环绕栅晶体管实现了每平方毫米3亿颗晶体管的突破性密度——但仅40%的制造良品率相比台积电预计的60%，可能使每座晶圆厂的成本增加20亿美元。该技术在实验室演示中表现出色，然而从&amp;rsquo;研究可行&amp;rsquo;到&amp;rsquo;批量盈利&amp;rsquo;的差距决定了哪家公司将控制AI处理器的未来。这不仅仅是技术竞赛——这是一场经济战争，制造精度而非纯粹创新决定着胜负。</description>
    </item>
  </channel>
</rss>
