[SMM Tin Express] China's Scientists Develop High-Precision Scalable Analog Matrix Computing Chip

Published: Oct 14, 2025 09:56

Save

Yesterday, a research team led by Sun Zhong from the Institute of Artificial Intelligence at Peking University, in collaboration with other researchers, published a paper in the international academic journal *Nature Electronics*, achieving a major breakthrough in new-type computing architectures. The team successfully developed a high-precision, scalable analog matrix computing chip based on resistive random-access memory, realizing for the first time an analog computing system whose accuracy rivals that of digital computing. This innovation remarkably improves the precision of traditional analog computing by five orders of magnitude. Relevant performance evaluations indicate that, when solving key scientific problems such as large-scale MIMO signal detection, the chip achieves a hundredfold to thousandfold increase in computational throughput and energy efficiency compared to current top-tier digital processors (GPUs). This achievement marks China’s breakthrough in solving the century-old challenge of analog computing, represents a significant advance in the transformation of computing paradigms in the post-Moore era, and opens up a new pathway to address computational challenges in fields such as artificial intelligence and 6G communications.

Data Source Statement: Except for publicly available information, all other data are processed by SMM based on publicly available information, market communication, and relying on SMM‘s internal database model. They are for reference only and do not constitute decision-making recommendations.

For any inquiries or to learn more information, please contact: lemonzhao@smm.cn

For more information on how to access our research reports, please contact:service.en@smm.cn

Related News

1 hour ago

Space Computing Power Professional Committee Established; 10 Key Projects Unveiled

On April 3, at the 2026 Space Computing Power Industry Conference, the "Space Computing Power Professional Committee" of the Computing Power Industry Development Consortium was established. Yicai reporters learned on site that the conference released ten key research projects that day, covering critical technologies at the satellite platform level, chip level, payload board level, networking and operations level, hardware system level, system software level, industrial application level, cross-structural level, cross-launch vehicle level, and cross-thermal management level, and that joint research collaboration will be carried out with industry participants.

1 hour ago

2 hours ago

Data: SHFE, DCE market movement (Apr 03)

The following table shows the ferrous and nonferrous metals movement on the SHFE and DCE on 03 Apr , 2026

2 hours ago

Nvidia Invests $2 Bln in Marvell for Silicon Photonics Tech Cooperation

Nvidia (NVDA.O) said it has invested $2 billion in Marvell Technology (MRVL.O), and the two sides plan to cooperate in the field of silicon photonics technology. Silicon photonics technology enables high-speed, energy-efficient data transmission, with its two common application fields being artificial intelligence and cloud computing. Nvidia Chief Executive Officer Huang Renxun said in a statement on Tuesday: “By joining forces with Marvell, we enable clients to leverage Nvidia’s AI infrastructure ecosystem and build dedicated AI computing capabilities at scale.”

2 hours ago