Blog Review: May 25

Coventor’s Michael Hargrove points to the need for a new generation of deep-submicron CMOS circuits that can operate at deep-cryogenic temperatures to achieve a quantum integrated circuit where the array of qubits is integrated on the same chip as the CMOS electronics required to read the state of the qubits.

Ansys’ Marc Swinnen warns about dynamic voltage drop as ultra-low supply voltages, more coupled cells, and sensitive transistors becomes more common, and the inability of modern design flows to completely eliminate that threat.

Cadence’s Paul McLellan checks out the road to 224G SerDes, including the current state of the standard definition process, a new candidate for modulation, and challenges to the analog front end, ADC, DSP, and PLL.

Synopsys’ Ron Lowman explores the different types of edge computing segments and finds that the convergence of edge computing and AI is poised to reshape traditional computing processes and pave the way for new applications and services in the years to come.

Siemens’ Katie Tormala considers how leveraging a parallel PCB thermal design approach for faster design closure and points to tens ways to streamline PCB thermal design.

Arm’s Ashok Bhat considers the environmental impact of training AI models and things can be done to minimize the carbon impact such as careful placement of workloads, consideration of energy sources, and an awareness of embedded emissions.

In a blog for SEMI, Okmetic’s Petri Santala shares highlights from the recent MEMS & Sensors Technical Congress including some of the latest MEMS and sensors advancements in markets ranging from displays to biotech and in key areas of MEMS manufacturing.

Intel’s Gadi Singer explores how knowledge constructs that allow an AI system to organize its view of the world, comprehend meaning, and demonstrate understanding of events and tasks will likely be key to transforming AI from surface correlation to comprehension of the world and digs into the dimensions of knowledge that support higher intelligence.

A Rambus writer explains how over-the-air, or OTA, programming is used to automatically update firmware, software, and encryption keys in automotive systems and some of the security challenges that must be overcome.

Nvidia’s Yi Dong introduces a framework for training conversational AI models using synthetic data created by transformer models that can be used as a valid substitute for real-life data in machine learning algorithms to protect user privacy while making accurate predictions.

Plus, check out the blogs featured in the latest Manufacturing, Packaging & Materials newsletter:

Coventor’s Gerold Schröpfer draws on ideas from the early days of computing to reduce power consumption.

Amkor’s Prasad Dhond looks at why reducing defects becomes more important as the number of chips in cars continues to accelerate.

eBeam Initiative’s Jan Willis presents a discussion with Micron’s Mike Hermes about major photomask technology changes and the challenges EUV poses for the mask shop.

Lam Research’s David Haynes finds that 5G is not the only wireless revolution taking place.

Brewer Science’s Jessica Albright explains the importance of monitoring and reporting different classes of emissions.

SEMI’s Serena Brischetto previews an upcoming keynote from Imec’s Luc Van den hove, who suggests looking beyond PPA to take a device’s environmental footprint into account.

QP Technologies’ Rosie Medina suggests how to get high levels of functionality and ruggedness in a small form factor.