以下文章来源于腾讯科技 ，作者值得关注的 腾讯科技 . 腾讯新闻旗下腾讯科技官方账号，在这里读懂科技！ 本文来自微信公众号： 腾讯科技 ，编辑：徐青阳，作者：值得关注的 比黄仁勋激进，比苏姿丰强硬，这是硅谷芯片领袖中一张独特的华人面孔。 他曾裁掉英特尔2万人，几乎每天超200人离职，但对58岁的"老英特尔"来说，掌握什么底牌，才能 继续留在牌桌上？ 1月6日，在拉斯维加斯的CES展台上，英特尔CEO陈立武意气风发。 他回顾了过去一年英特尔团队在架构创新、制造工艺及跨软硬件协同优化等方面取得的突破性进展， 并正式宣布：英特尔已兑现承诺，在2025年底前完成了首批基于18A制程产品的出货。 陈立武还表示，英特尔实际上超额完成了目标。他现场宣布，酷睿Ultra系列3的所有三款芯片目前已 进入同步扩产阶段。 台下掌声不断，但只有陈立武知道，这一刻的荣光背后，是过去一年惊心动魄的救亡图存。 对英特尔来说，2025年热点不断：新CEO上任、大规模裁员、政府近百亿美元投资、英伟达50亿砸 进来、股价涨超80%。 热闹归热闹，英特尔还是那个英特尔。 换了CEO，裁了人，通过"混改"拿了钱，技术也有进展，可最核心的问题——代工 ...

2025年底，三星终于亮出了全球首颗2nm芯片。 这颗芯片就是三星自己的猎户座手机Soc Exynos2600，它采用三星自己的GAAFET 2nm技术制造，并且 已经是第二代GAA技术了。 按照三星的说法，这颗芯片采用10核设计，虽然这些CPU的主频并不高，但性能涨了39%，同时在采用 了HPB技术（Heat Pass Block）后，温效降低了16%。 当然，目前其它参数还差不清楚，比如晶体管密度等等。 但不管怎么样，三星的这颗芯片，意味着全球芯片工艺，正式进入了2nm时代。 而这也意味着，我们与国际顶尖的芯片制造工艺，又多落后了一代了，差距又稍微的拉大了一点点。 如今三星到达2nm，台积电也很快就要到达2nm，那么中间就相差了5nm、3nm，算是落后三代了。 很多人可能会说，我们很快就会追上的，浸润式光刻机也能够制造5nm芯片。 但是，大家要清楚的是，没有EUV光刻机，其实是非常难制造7nm以下芯片的，理论上浸润式DUV是能 够制造5nm，甚至3nm的芯片。 但要经过更多重的曝光之后，先不说技术上究竟能不能真正实现，因为还从来没有厂商，用浸润式 DUV光刻机去制造5nm芯片，台积电当年也只拿浸润式制造7 ...

Core Insights - Intel has begun mass production at its new $32 billion chip factory in Arizona, aiming to regain leadership in advanced manufacturing [1][4] - The company's 18A semiconductor production technology is touted as the most advanced globally, but Intel acknowledges the need to earn customer trust [1][5] - Key customers like Apple, Nvidia, and Qualcomm will test Intel's new chips in the next 6-8 months, which will be crucial for the company's future [1][4] Investment and Financial Implications - The investment in the Arizona factory exceeds half of Intel's projected revenue for 2024, raising concerns about the valuation of its loss-making foundry business [4] - Intel's foundry division is currently losing over $10 billion annually and carries a net debt of $20 billion [1][4] - The U.S. government has converted planned subsidies into equity investments, acquiring a 10% stake in Intel, which has boosted investor confidence [5] Production and Technology Challenges - The Fab 52 facility has commenced operations, while the Fab 62 facility is still under construction, requiring significant resources [4] - Early yield issues have reportedly been resolved, and Intel plans to launch two new chips: Panther Lake for PCs and Clearwater Forest for servers [4] - Intel must demonstrate the scalability of its 18A process to secure pre-orders for its planned 14A process set for 2028 [5] Market Position and Competitive Landscape - Intel faces the challenge of competing with TSMC, which has long dominated the high-end chip foundry market [5] - The company's ability to attract customers hinges not only on technology but also on establishing trust in its execution capabilities [5]

Group 1 - Tokyo Electron announced the dismissal of an employee from its Taipei branch following the arrest of six individuals suspected of stealing TSMC's trade secrets [1] - TSMC confirmed the leak of trade secrets related to its 2nm process, resulting in the dismissal of multiple employees and legal actions against them [1][2] - The investigation revealed that approximately 10 individuals were involved, including a former TSMC employee who contacted R&D personnel to obtain confidential information [1] Group 2 - Three individuals from the 2nm trial production team were found to have leaked over 400 images of process integration technology to Tokyo Electron employees while working remotely [2] - TSMC maintains a zero-tolerance policy towards violations of trade secret protection and has implemented strict monitoring mechanisms to detect such breaches [1][2] - In the global foundry market, TSMC holds a dominant position with a 67.6% market share, while Samsung and SMIC follow with 7.7% and 6.0% respectively [3]

Core Viewpoint - TSMC is facing a significant security breach involving former employees leaking sensitive information about its 2nm process technology, which has raised national security concerns in Taiwan [4][5]. Group 1: Incident Details - Multiple former TSMC employees are accused of illegally obtaining sensitive information related to the 2nm process, leading to their dismissal and legal action [4]. - The case is being investigated by Taiwan's High Prosecutors Office under the National Security Law, marking it as a major case involving chip technology [4][5]. - TSMC confirmed the incident and stated that it has taken strict measures against the involved personnel and is pursuing legal action [4]. Group 2: Technology and Market Implications - TSMC plans to begin mass production of 2nm chips by the end of this year, with major clients including Apple and Qualcomm [4][5]. - The 2nm technology is currently the most advanced semiconductor manufacturing process, with only a few companies, including TSMC, Samsung, Intel, and Japan's Rapidus, continuing its development [4][5]. - Rapidus, established by major Japanese companies, is in competition with TSMC, Samsung, and Intel, aiming to launch its 2nm process by 2025 [5]. Group 3: Potential Consequences - The leak not only involves trade secrets but could also impact Taiwan's technological advantage in the global semiconductor industry and its regional security [5]. - Rapidus has already begun testing 2nm GAA transistor prototypes, indicating its progress in the competitive landscape [5].

Core Insights - The article discusses the launch of IBM's Power11 processors, which are designed to support large-scale computing environments with high memory and I/O capabilities [4][10] - It highlights the evolution of IBM's Power processors, particularly the transition from Power10 to Power11, and the challenges faced during the manufacturing process [5][7] - The article emphasizes the competitive landscape in the server market, particularly against AMD and Intel, and how IBM aims to maintain its position through advanced technology [6][19] Summary by Sections Power11 Launch - Power11 processors were launched and are set to be available on July 25, featuring enhancements over the previous Power10 generation [4] - The processors are designed to handle extensive memory and I/O requirements, making them suitable for enterprise-level applications [3] Evolution of Power Processors - Power11 is a refined version of the Power10, which faced manufacturing delays due to issues with GlobalFoundries' process technology [5][6] - The transition to 7nm technology was necessitated by delays in the 10nm process, leading to a redesign of Power10 [7][8] Competitive Landscape - The article notes that AMD has gained market share due to Intel's manufacturing challenges, which parallels IBM's own struggles with Power10 [5][6] - IBM's strategy includes leveraging its long-standing expertise in chip design and manufacturing to compete effectively in the high-performance computing market [8][19] Technical Specifications - Power10 and Power11 both feature 16 cores and 18 billion transistors, with Power11 offering improved clock speeds and core utilization [10][11] - The Power11 architecture supports advanced memory configurations, including DDR5, which enhances performance and efficiency [12][20] System Architecture - The Power E1180 server, based on Power11, can support up to 256 cores and 64TB of memory, showcasing a balance between I/O and computational power [18][19] - The architecture allows for flexible memory configurations, enabling users to optimize performance based on their specific workloads [21][22]

Core Insights - Intel is facing significant challenges, including plans to lay off 15%-20% of its global workforce and specific layoffs at its Santa Clara headquarters [3][4] - The company has already undergone large-scale layoffs in May, affecting various departments, particularly engineering, and has closed its automotive chip division in Germany [3][4] - Intel's manufacturing division is expected to see up to 20% layoffs starting July 2025, and the company plans to shut down its Automotive Group, discontinuing its automotive business entirely [4] - Recent executive turnover includes the departure of several senior executives and changes in the board of directors, indicating a restructuring effort to reduce costs amid declining performance [4] - Intel's market share is being eroded by competitors like AMD and NVIDIA, with a reported net profit of only $292 million in 2023 and a net loss of $630 million in Q1 2024 [5] - Revenue declines were noted in key segments, with a 20.8% drop in the Client Computing Group and a 7.6% decrease in the Data Center and AI segment [5] - Despite these challenges, Intel retains strong R&D capabilities and brand influence in the semiconductor market, and the restructuring plan aims to lower operational costs and improve efficiency for future growth [5]

如果您希望可以时常见面，欢迎标星收藏哦~ 在台积电早起那的北美技术研讨会上，台积电业务发展和海外运营办公室高级副总裁兼副联 席首席运营官Kevin Zhang称其为"最后也是最好的 FinFET 节点"。台积电的战略是开发N3 工艺的多种变体，打造一个全面可定制的硅片资源。 Kevin Zhang表示："我们的目标是将集成硅片性能打造成为一个平台。" 截至目前，N3 现有 或计划推出的版本包括N3B、N3E、N3P、N3X、N3S、N3RF、N3A和N3C。 换而言之，自英特尔在2009 年的开发者大会上推出了 22 纳米 FinFET 晶圆后，这个改变了 芯片行业的设计，从某种程度看，走到了尽头。 FinFET，英特尔掀起革命 对芯片制造行业的读者应该知道，在过去几十年里，芯片的晶体管已经从planer走向了FinFet。至 于为什么用FinFet，以及这个进步的意义，我们需要从晶体管的原理说起。 本质上，在一颗芯片中，晶体管的目标是充当一个高速电子开关。导通时，电流从晶体管的源极流 向漏极。截止时，电流停止。反型层（上图蓝线）是电流实际流动的地方。 在理想情况下，晶体管需要做三件事： 1) 开启时允许尽可 ...

Core Viewpoint - Intel's Panther Lake chips are likely to be delayed beyond 2025 due to slow yield improvements in the Intel 18A process, which may impact the company's revenue and profit in the second half of 2025 [1][2]. Group 1: Chip Production and Delays - Analyst Ming-Chi Kuo indicates that the production timeline for the Panther Lake (PTL) series has been pushed from early September 2025 to mid-Q4 2025, meaning devices using these chips may not be available until 2026 [1][2]. - Intel's initial production schedule communicated to partners was for the second half of 2025, but the delays could negatively affect the company's revenue, profit, and supply chain trust [2]. Group 2: Manufacturing Process Challenges - The Intel 18A manufacturing process, which utilizes RibbonFET and PowerVia technologies, has not achieved the expected yield rates, currently estimated at 20%-30%, hindering the company's ability to enter the shipping phase [2]. - Internal investigations suggest that there will be no improvement in yield rates by Q3, making it nearly impossible for Intel to achieve mass production within the anticipated timeline [2].