5nm nanosheet transistors cut power by 75%
June 05, 2017 // By Nick Flaherty, eeNews
Researchers at IBM have developed a practical way to build transistors on a 5nm process that provides a reduction in power consumption of up to 75% or a performance boost of 40% for the same power.
Instead of using FinFET structures, engineers at the IBM-led Research Alliance at the SUNY Polytechnic Institute Colleges of Nanoscale Science and Engineering’s NanoTech Complex in Albany, NY, used a ‘gate-all-around’ (GAA) built with silicon nanosheets. For the last ten years IBM has been working on nanosheets where each 2D layer is one atom thick and stacked layers build up the structure of the transistor.
This enabled the first practical use of extreme UV (EUV) process technology at research partner GLOBAL FOUNDRIES as the nanosheets can be more easily aligned to build up the devices. This provides a 40% performance boost at fixed power over current 10nm FinFET devices, or a 75% power saving for the same performance.
To read the full article, click here
Related Semiconductor IP
- AP Memory VHM Controller
- High-performance, commercial-grade RISC-V CPU Core IP
- SWI3S Verification IP
- Secure Boot Loader
- Memory Subsystem
Related News
- SureCore announces low power cryogenic memory technology that could help dramatically cut data centre power usage
- PLDA Announces Major PCIe 5.0 Design Win on Cutting Edge 5nm Process Node
- Capacity Cuts and Surging Demand for AI Power ICs Set Stage for Mature-Node Foundry Price Increases
- Microchip develops new flash technology to cut cost of re-programmable MCUs
Latest News
- proteanTecs Unveils Advanced Packaging Solutions to Improve Yield, Performance, and Reliability of Multi-Die Systems
- StarFive Launches Server-Class RISC-V IP to Accelerate Its Strategic Expansion into AI Computing Infrastructure
- Silvaco Announces Partnership with Dassault Systèmes to Advance First-Time-Right Semiconductor Manufacturing
- Socionext Broadens SoC Roadmap with Intel 18A-P Process Technology
- BrainChip and Neuromorphyx Announce an AKD1500 Embedded Developer Board, Bringing Neuromorphic AI to the Embedded Engineering Community