The new process uses nanostack transistors and delivers significant performance and energy efficiency gains over the 2nm node.
IBM has announced the development of a new semiconductor manufacturing technology in the 0.7-nanometer class. The announcement marks the company's entry into sub-nanometer production scales, aiming for substantial advancements in processor capabilities.
According to the manufacturer, the technology employs nanostack transistors. This architecture requires twice as many manufacturing steps in the initial phase of the process, known as FEOL (Front-End-Of-Line), compared to previous methodologies.
The projected gains from the new architecture are significant. The 0.7nm process offers up to a 50% increase in performance when compared to IBM's own 2nm node. Beyond processing power, the technology also focuses on optimizing resource consumption.
According to the company, the energy efficiency of the new process is up to 70% higher than that observed in the 2-nanometer generation. These improvements aim to address the industry's growing demand for more powerful and energy-efficient chips, especially for artificial intelligence and advanced computing applications.
It is a new semiconductor manufacturing process that uses nanostack transistors to achieve sub-nanometer production scales, targeting advanced computing and AI applications.
IBM's 0.7nm process delivers up to a 50% increase in performance and up to 70% higher energy efficiency compared to its previous 2nm node.
Nanostack transistors are the architecture used in the new 0.7nm chips. They require twice as many manufacturing steps in the initial Front-End-Of-Line (FEOL) phase compared to previous methodologies.