②TMT领涨板块，SW营销代理(+5.75%)/SW军工电子Ⅲ(+5.41%)/SW LED (+5.18%)/ ③TMT领跌板块，SW通信网络设备及器件(-3.57%)/SW印制电路板(-3.18%)/SW数字芯片设计(-1.14%)/ 国内新闻 / Part 02 ①艾邦ARAI眼镜资讯，炬光科技与 BrightView 宣布建立战略合作伙伴关系/通过整合互补能力/炬光科技与 BrightView旨在加速光学元器件、模块及系统级解决方案的开发与商业化落地/以满足相关行业在性能、可靠性及规 模化生产方面的严格要求/ 行情速递 / Part 01 ①大盘指数，北证50 (+3.82%)/深证成指 (+0.79%)/科创50 (+0.78%)/创业板指(+0.63%)/上证指数(+0.33%)/ ②半导体封测，北大团队成功设计并研制出一款专用于非负矩阵分解计算的模拟计算芯片/该芯片在核心性能上实现 跨越式提升/相较于当前先进数字芯片/计算速度提升约12倍/能效比更是突破228倍/ ③大话芯片，Counterpoint Research表示/台积电预计2028年前下调晶圆14厂成熟制程产能15%–20%/以因应市 ...

Core Insights - The research team from Peking University has developed a new analog computing chip for non-negative matrix factorization, significantly improving processing speed and energy efficiency compared to current digital chips [1][2]. Group 1: Technology Overview - Non-negative matrix factorization (NMF) is a powerful data dimensionality reduction technique used in various fields such as recommendation systems, bioinformatics, and image processing [1]. - Traditional digital hardware struggles with real-time processing of large datasets due to computational complexity and memory limitations [1]. - The new chip utilizes resistive random-access memory (RRAM) and features a reconfigurable compact generalized inverse circuit, optimizing the core computation steps of NMF [1]. Group 2: Performance Validation - The research team built a testing platform to validate the chip's performance in typical scenarios, achieving minimal image quality loss in image compression while saving 50% of storage space [2]. - In recommendation system applications, the chip demonstrated a 212-fold speed increase and a 46,000-fold energy efficiency improvement compared to mainstream programmable digital hardware on the MovieLens 100k dataset [2]. - For the Netflix dataset, the chip achieved a speed improvement of approximately 12 times and an energy efficiency increase of over 228 times compared to advanced digital chips [2]. Group 3: Implications for Industry - This research opens new pathways for real-time solutions to constrained optimization problems like NMF, showcasing the potential of analog computing in handling complex real-world data [2]. - The advancements could lead to innovations in real-time recommendation systems, high-definition image processing, and genetic data analysis, contributing to more efficient and lower-power artificial intelligence applications [2].

Group 1: Simulation Chips - A research team from Peking University has developed a high-precision, scalable analog matrix computing chip based on resistive random-access memory, achieving 24-bit fixed-point precision for the first time globally [1] - This new chip significantly improves computing precision from 1% to one part in ten million, with throughput and energy efficiency enhanced by 100 to 1,000 times compared to current top digital processors (GPUs) [1] - The breakthrough in analog computing chips marks a significant advancement in China's computing paradigm shift in the post-Moore era, opening new pathways for AI and 6G communication [1] Group 2: AI Education - The Ministry of Education plans to continue advancing the AI Empowerment in Education initiative, with policies to be introduced in the coming year to systematically promote AI education and applications [2][4] - From 2024, the initiative will explore AI empowerment in education, with a goal to establish a future-oriented educational system by 2025 [2] - The introduction of these policies signifies a transition from pilot programs to a systematic approach in AI education, which will have a profound impact on educational philosophy, teaching models, and talent cultivation [2] Group 3: Macro and Industry News - The first batch of funding amounting to 62.5 billion yuan has been allocated under the "Two New" policies, which include subsidies for smart glasses and equipment upgrades in various sectors [3] - The Ministry of Industry and Information Technology has issued an implementation plan for the digital transformation of the automotive industry, focusing on the large-scale application of intelligent robots in manufacturing processes [3]

Group 1 - A research team from Peking University has developed a high-precision, scalable analog matrix computing chip based on resistive switching memory, achieving a 24-bit fixed-point precision for the first time globally [1] - This new chip significantly improves computing precision from 1% to one part in ten million, with throughput and energy efficiency enhanced by 100 to 1,000 times compared to current top digital processors (GPUs) [1] - The analog computing chip can be mass-produced using mature processes of 28 nanometers and above, circumventing the bottleneck associated with photolithography machines [1] Group 2 - The analog computing chip features high precision, low power consumption, and strong real-time performance, marking a significant advancement for China in the post-Moore era of computing paradigm transformation [1] - As technology matures and industrialization progresses, the analog computing chip is expected to further enhance computing precision, expand application areas, and meet more complex task requirements [1] - This technology represents a major breakthrough in computing architecture, laying a crucial foundation for building a self-controlled and efficient computing ecosystem [1]

Group 1 - The core point of the article highlights that despite the U.S. tariff policy aimed at Chinese mature chips, over 70% of global orders are still flowing to China, indicating a contradiction between policy pressure and market dynamics [1][3][21] - The U.S. government announced additional tariffs on mature chips from China, set to take effect in 2027, primarily targeting chips of 28nm and above, which are essential for everyday applications [3][5] - The demand for mature chips is driven by the need for reliability and stability rather than extreme performance, making them a crucial component in various industries [5][19] Group 2 - China's mature chip industry has developed a complete supply chain, including production, packaging, design, and delivery, which is difficult for U.S. policies to disrupt [5][21] - The U.S. policy aims to limit China's position in the chip supply chain, but the actual effect may catalyze companies to further invest in Chinese supply chains due to cost and stability considerations [7][23] - Chinese manufacturers have rapidly expanded production capacity, allowing them to respond quickly to market demands, which is a core capability valued by global customers [17][25] Group 3 - Recent technological breakthroughs in China, particularly by a team from Peking University, have enhanced innovation capabilities in mature chips, supporting AI and next-generation communication technologies [9][11] - The competitive edge of China's mature chips is not solely based on low prices but on a comprehensive advantage derived from a complete industrial chain [13][19] - The shift in global semiconductor orders from a few dominant firms to a more distributed model reflects changing market dynamics, with China establishing a leading position in the mature chip market [21][26] Group 4 - U.S. automotive and electronics companies have expressed concerns about the impact of tariffs on costs and delivery times, indicating a shift in the industry focus towards increased procurement from China [23][30] - The article emphasizes that the future competition in the semiconductor industry will depend on understanding market needs, with China redefining the global semiconductor logic through its actions [29][30]

Core Insights - The research team from Peking University has achieved a breakthrough in high-precision, scalable analog matrix equation solving, published in Nature Electronics, marking a significant advancement in analog computing technology [1][2] - This innovation demonstrates that analog computing can efficiently and accurately address core computational problems in modern science and engineering, potentially disrupting the long-standing dominance of digital computing [2][3] Group 1: Key Innovations - The first key innovation is the use of resistive random-access memory (RRAM), which allows for precise control of resistance states and retains data without power, enabling it to function as both a memory and a computing unit [4] - The second key innovation stems from a foundational discovery in 2019, where the team designed an analog circuit capable of solving matrix equations in a single step, significantly compressing traditional iterative algorithms [5] - The third key innovation is the "bit slicing" technique, which breaks down 24-bit precision into multiple 3-bit segments for processing, allowing for a more sophisticated and efficient analog computation [5] Group 2: Practical Implications - The breakthrough allows for solving matrix equations with 24-bit precision in just a few iterations, drastically reducing the computational steps required for complex tasks, such as 6G signal detection [7] - In the AI field, this advancement could alleviate the "computational bottleneck" faced by large models, enabling faster and more efficient training processes [7] - The technology also addresses critical challenges in 6G communication, enhancing signal detection capabilities while significantly reducing energy consumption [8]