event-trade登录编辑
📰 DigiTimesDigiTimesDoubao · 实盘不做把握 7/10不做原因:无增量信息
北京时间·🕒 发布 09-14 17:04:30·⟳ 抓取 09-14 18:04:44 +1h00m·⚡ 决策 09-15 13:11:22 +19h06m·42.5 天前

DigiTimes 转述日经:富士通 Monaka AI CPU 采用台积电 2nm、2027 年量产、Supermicro 为首个客户

DigiTimes 该文是对日经(9/13)、Diamond Weekly、彭博旧报道的付费墙内二次汇编,正文多次以‘Nikkei reported / Diamond Weekly reported / Bloomberg reported’转述,无任一新增确认事实或数字;底层催化(2027 量产、FY2030 累计 2500 亿日元目标)时点遥远、对台积电/Supermicro/Nvidia 无可量化即期弹性,对富士通亦属旧战略的媒体重述,不构成可交易的增量预期差。

📄 原文DigiTimes↗ 原文
Fujitsu's Monaka targets 2027 mass production with TSMC 2nm boost
中文網 繁體 简体 SUBSCRIBE My account Sign out SUBSCRIBE Trending Apple Huawei Nvidia AI Search Scaling AI Glasses Inference E…
不做 富士通 Fujitsu(6702.JP)
DigiTimes 转述日经 9/13 旧闻,无新增事实;2500 亿日元为 FY2030 累计目标、2027 年才量产,无即期预期差,消息已扩散逾一天
不做 台积电 TSMC(2330.TW)
单一日本 CPU 系列 2027 年才上量,对其千亿美元级营收无可量化弹性,且为已报道路线确认
不做 Supermicro(SMCI.US)
仅列为首个客户、无约束性订单量价,2027 年前无收入贡献,属旧闻转述,edge 不足
展开详情(成交明细 / 研究档案 / 原文全文 / 决策记录链)

研究档案

结论摘要

这是 DigiTimes 对日经 9 月 13 日旧闻的转述汇总,核心事实(台积电 2nm、2027 量产、Supermicro 首家客户、FY2030 目标)此前已广泛传播;且量产在一年多以后、收入到 FY2030 才累计兑现,对所有可交易主体均无即期 materiality,不做。

🔎 新鲜度:最早出现 2026-09-13 · 扩散 high · price-in 风险 high ↗ ↗

消息传播链

  1. 2026-09-13首发Nikkei Asia(日经亚洲)
    多家媒体引述日经亚洲当地时间 9/13 报道:Monaka 用台积电 2nm、2027 年起在日美亚销售、Supermicro 为首个海外客户;本次未直接取到日经付费原文 URL,故链接留空并标注为转述核证
  2. 2026-09-13 17:18 +09:00财经媒体扩散朝鲜日报(ChosunBiz 英文版)非一手原文 ↗
    9/13 即转载 Monaka 由台积电生产、2027 年在日美亚销售,证明核心消息在 DigiTimes 前一天已广泛传播
  3. 2026-09-14 17:04 +08:00行业媒体转载DigiTimes非一手原文 ↗
    本文为付费墙内汇编,正文明确逐条转述 Nikkei、Diamond Weekly、Bloomberg,未提供任何新增确认事实或数字

关键理由

  • 若硬要做多,唯一直接受益方是富士通,但 2500 亿日元为 FY2030 累计目标、量产 2027 年才开始,属远期且已被日经报道过的路线
  • 对台积电、Supermicro、Nvidia 均无约束性新订单或金额,事件日无可交易的正 edge

主要风险

  • 若富士通后续公布具约束力的 Supermicro 大单或提前量产/上修目标,可能构成新的独立催化,需重新研判
  • DigiTimes 付费正文未完全获取,不排除存在未在免费摘要显示的新细节;但免费正文已明确全部核心信息来自日经等旧报道

反向证据

  • 做多前查反向 · 来源 web · 日期 2026-09-14 · 影响 weakens
    公开报道仅称 Supermicro ‘first in line’、24 家试用/40 余家洽谈,无约束性采购量或金额,无法坐实即期收入
    检索:Fujitsu Monaka Supermicro binding order volume revenue commitment 原文 ↗
  • 做多前查反向 · 来源 web · 日期 2026-09-13 · 影响 invalidates
    核心事实在 9/13 已由日经首发并多媒体扩散,DigiTimes 9/14 文为二次汇编,未发现新催化
    检索:Fujitsu Monaka TSMC 2nm new announcement September 2026 原文 ↗

先例如何影响了这次决策

对本次决策的影响:未改变结论;AVOID 基于陈旧性与低 materiality,不依赖先例涨幅

方向:未运行先例统计:结论由‘无新增事实+远期低 materiality’决定,同市场同相位先例不足以把一条二次转述旧闻变成正 edge

涨幅量级:不适用,无 ENTER 标的,不需校准止盈幅度

形态:不适用

工具执行结果 · 成功 2 · 降级 0 · 失败 0 · 取消 0
  1. search_processed_events 成功 stock_tools HTTP · 17 ms
    {"result": "search_processed_events:agent query=`fujitsu monaka cpu supermicro customer | fujitsu monaka tsmc 2nm mass production 2027 | 富士通 monaka 2nm 台积电 量产`;结构化索引覆盖完整 72 小时;旧记录兼容扫描近 0.0/72 小时、0 条本臂决策(完整),返回 0 条候选。未命中只表示这些 agent query 表达式没有召回候选,不能据此证明消息首发或不是重复;按当前事件继续正常研判。"}
  2. web_search 成功 doubao_search · 1,468 ms
    豆包搜索结果(原始查询:Fujitsu Monaka TSMC 2nm Supermicro first customer Nikkei mass production 2027;API SearchType:web)

原文全文

展开原文 · 5,167 字
中文網

繁體

简体

SUBSCRIBE

My account

Sign out

SUBSCRIBE

Trending Apple
Huawei
Nvidia
AI Search
Scaling AI Glasses
Inference Era AI
Advanced Packaging

Trending

SUBSCRIBE

CONNECT WITH US

Sign out

REALTIME NEWS

Corning sees CPO and glass core as new semiconductor battlegrounds

Tomorrow's Headlines

Sep 14, 18:47

China optical transceivers avoid US FCC curbs as NPO and CPO compete in high-speed links

Tomorrow's Headlines

Sep 14, 18:46

China's AIDC token costs hinge on access to advanced GPUs

Tomorrow's Headlines

Sep 14, 18:46

Home

Tech

Semiconductors

Fujitsu's Monaka targets 2027 mass production with TSMC 2nm boost

Jenchieh Chiang, Taipei

Sep 14, 2026, 17:04

0

Credit: AFP

Fujitsu is preparing to take its advanced AI CPU overseas, with sales set to begin as early as 2027 in Japan, the US, and Asia. The company is betting on Monaka, a high-efficiency, power-saving chip developed with technologies refined through its work on the Fugaku supercomputer, to regain competitiveness in advanced logic ICs as demand for AI inference surges.

Nikkei reported that Monaka will use TSMC's 2nm process technology and is expected to enter mass production as early as 2027, with Supermicro first in line as a customer before Fujitsu pushes the chip to US cloud providers and others. It will be the first time Fujitsu has formally turned an advanced chip originally used only in its own products into an external commercial offering, marking a major turning point for its semiconductor business.
Early traction with AI servers
Fujitsu will also begin selling AI servers equipped with Monaka in Japan and overseas. The server has already been trialed by 24 companies in the financial and telecommunications sectors, among others, while more than 40 companies are in talks to introduce it.
Revenue targets and platform ambitions
The company aims to generate cumulative revenue of JPY250 billion (about US$1.6 billion) from external CPU sales in fiscal 2030 (2030/4~2031/3). It also plans to sell 60,000 AI servers, while pitching platforms that combine CPUs and servers to support AI development and operations.
Diamond Weekly reported that Monaka is expected to begin mass production at TSMC as early as March 2027.
Nvidia partnership broadens Monaka's reach
Fujitsu was selected in 2025 as a partner in Nvidia's "NVLink Fusion," enabling Nvidia GPUs to connect at high speed with Fujitsu CPUs.
That collaboration is expected to broaden the use case for Monaka as a data-center CPU, with the combination of Monaka and Nvidia GPUs slated for Fujitsu's next-generation "Monaka-X" product in fiscal 2029.
Eyes on 1.4nm and Rapidus
It is still unclear which foundry will manufacture Fujitsu's next-generation chip using 1.4nm process technology. The company may consider Rapidus, Japan's national foundry champion, as one of the candidates for outsourced production.
Bloomberg reported that Rapidus president Atsuyoshi Koike said the company is discussing partnerships with its investors, including Fujitsu, Denso, NTT and Kioxia.
Along with continued financial support for Rapidus, Japan's Ministry of Economy, Trade and Industry wants Japanese companies such as Fujitsu to entrust production of advanced AI chips to Rapidus, in an effort to build a domestically made semiconductor supply chain.

Article
translated by Charlene Chen and edited by Jerry Chen

AI Search

Related stories

Japan funds AI chip ecosystem around Rapidus

Fujitsu plans 1.4nm AI chip Japan-based production with Rapidus

Rapidus raises US$1.7 billion to support 2nm mass production by 2027

Categories
Chips + components IC manufacturing

Tags
1.4nm 2027 2nm CPU Fujitsu Japan Nvidia production Rapidus sales technology TSMC

Companies
Taiwan Semiconductor Manufacturing Company

Share this article

Other links

Print

Related topics
Global supply chain: Advanced economies

Asia

TSMC updates

Regions

BIZ FOCUS
Sep 14, 08:00
Hauman Unveils Micro LED Testing Breakthrough for AI Optical Interconnects

Monday 14 September 2026
Scaling Photonics: Assembly, Test and Fiber Automation at SEMICON Taiwan
Friday 11 September 2026
DAS Expands in Taiwan and Unveils World First at SEMICON
Friday 11 September 2026
Toward Technologies Redefines Relay Solutions at SEMICON Taiwan 2026

MOST-READ
7 DAYS NEWS
Google's v10 TPU opens door to AMD without sidelining Broadcom, MediaTek
Update: ABF substrate maker Unimicron searched in origin-labeling probe
Qualcomm's HBC play pulls Samsung and SK hynix into a new race beyond HBM
SK keyfoundry expands 8-inch capacity as China demand strains supply
Samsung, SK Hynix step up China NAND upgrades as US tool risks grow

Column: When digital trust breaks — quantum risk rewrites supply chain resilience
Wuliangye skips carmaking, turns liquor waste into silicon-carbon anodes
The 1,000-watt problem: a TSMC veteran on the brute force forging Nvidia's AI chips
eCloudvalley expands enterprise AI Agent portfolio
CXMT rides AI boom to 80% operating margin
Full list

Members only

×

Sorry, the page you are trying to open is available only for our paid subscribers.
Please login to read more
New users, please register first

Notification

×

决策记录链 · 只追加

每次领取产生一条记录,重新决策追加一条并以 supersedes 指向被替代的记录(ADR 0014)。

这条决策早于决策账本(2026-10-08 起逐条追加),没有追加记录。