TSMC利用AI设计节能芯片,提高每瓦的性能,支持可持续的AI增长。 TSMC uses AI to design energy-efficient chips, boosting performance per watt and supporting sustainable AI growth.

TSMC正在利用AI设计更节能的芯片, 目标是每瓦的性能达到10倍以上, 解决AI服务器不断增长的电力需求, TSMC is using AI to design more energy-efficient chips, aiming for up to 10 times better performance per watt, addressing the rising power demands of AI servers that can consume up to 1,200 watts. Nvidia公司是Nvidia公司的关键供应商,该公司利用人工智能工具与Cadence和催眠术一起加速芯片设计,将任务从几天到几分钟地削减。 The company, a key supplier to Nvidia, leverages AI tools with Cadence and Synopsys to accelerate chip design, cutting tasks from days to minutes. 通过推进3D芯片集成和探索光学连接,TSMC解决了数据传输的物理限制,支持可扩展、可持续的AI基础设施。 By advancing 3D chip integration and exploring optical interconnects, TSMC tackles physical limits in data transfer, supporting scalable, sustainable AI infrastructure.