Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Story

Lingcun Unveils Integrated Power and Interconnect Solution for AI Chips

AI summary

MIT Technology Review China reports that Lingcun Technology is integrating early-stage simulation, high-capacitance silicon capacitors, silicon interposers, silicon bridges and advanced-packaging interconnects into a system-level delivery model for AI compute chips. The company says its approach, which can begin during a customer's initial design phase, reduces multi-supplier coordination and compatibility issues while addressing power integrity and signal integrity together. Performance, reliability and delivery benefits have not been independently established in the report.

In-depth insight

The offering reflects a broader shift in AI-chip development toward integrating power delivery, packaging and interconnect decisions earlier in the design cycle, where vendors can address compatibility and supply-chain coordination as one engineering problem.

随着算力芯片的功耗越来越高,供电这件事也变得越来越难办。电流的需求持续增加,去耦电容必须无限贴近芯片才能够改善供电性能。

企业研发团队面临的麻烦远不止于此,电容、基板和仿真往往要分别找不同厂商对接,各家接口互不兼容,出了问题有时还会互相推诿,导致工程师的精力大量消耗在沟通和适配上。

中介层产能紧缺、排期时间长,更是让项目进度充满了不确定性。既然电容必须贴近芯片,甚至要埋进封装内部,为何不能有一家供应商,把电容、中介层以及前期仿真一起打包做了?

破局之道:从器件供应商到系统级交付商

为何需要打包做?这是因为在之前的芯片研发模式里,产业链的分工十分明确。做逻辑芯片设计的公司只负责 GPU 和 SOC 芯片的设计,做封装测试的负责把芯片给包起来,各种电子元器件则由专门的供应商负责提供。这种模式在芯片制程比较宽裕的时代运转得很好,但是随着 AI 算力芯片的快速迭代,传统边界正在被打破。

#Lingcun Technology#HD Series#Silicon Interposer#Silicon Bridge#Integrated Voltage Regulator

Original · 12 min readRead the original
Group chat

No comments yet. Start the conversation.