Blog Review: Feb. 22

FPGA complexity mirrors ASIC path; integrating lasers and active gain elements; RISC-V in space; mixing power and data transmission in automotive; lightweight crypto.

popularity

Siemens EDA’s Harry Foster observes that the FPGA market continues to go through a similar complexity curve that the IC/ASIC market experienced in the early and mid-2000 timeframe.

Synopsys’ Mitch Heins explores the benefits of heterogeneous integration of lasers and active gain elements in a silicon-based photonic IC, including reduced system costs, size, weight, and power along with improved reliability and yield.

Cadence’s Paul McLellan checks out NASA’s efforts to build a new high-performance computer for space based on RISC-V and some of the critical requirements for such a processor.

Renesas’ Tobias Belitz examines how Power over Data Line can be used to reduce the number of cables in a vehicle and still accommodate the growing number of sensors by mixing power and data transmission first and separating them later.

Riscure’s Marc Witteman explains how the lightweight crypto algorithm selected by NIST can deliver both encryption and integrity verification at a low cost for IoT and other resource-constrained devices.

In a podcast, Arm’s Geof Wheelwright and Remy Pottier join 311 Institute’s Matt Griffin to chat about the links between the metaverse and digital twins as well as potential applications in both enterprise and consumer markets.

Ansys’ Alex Lam uses observations combined with simulation to determine the circumstances surrounding the crash of the spent Long March 3C rocket into the Moon.

SEMI’s Mousumi Bhat points to five key capability characteristics modern leaders need to possess in order to be able to merge sustainability with business strategy.

Verification blogger Neil Johnson argues that while simulation code coverage and formal code coverage represent fundamentally different information, there is value in merging coverage from different sources.

Memory analyst Jim Handy examines the current state of the memory market as companies report lower revenues, gigabyte shipments, and prices, plus how the cycle is likely to develop with a turnaround on the way.

Intel’s Hechen Wang explains analog approaches to in-memory computing for AI applications and the efforts to balance energy efficiency and accuracy.

Western Digital’s Ronni Shendar checks out a startup looking to build the first data center on the moon  to safeguard data from increasing climate change induced unpredictability on Earth.

NXP’s Meindert van den Beld envisions a transformation of vehicle architecture in which all functions will be designed, tested, validated, and possibly modified on a virtual model that will exist alongside the physical car throughout its lifetime.

Nvidia’s Somshubra Majumdar introduces a method to select models that can balance the recognition accuracy of general speech with the help of adapter modules for transducer-based speech recognition systems.

Marvell’s Amit Sanyal expects that we are on the verge of a major retrofit of the underlying technologies for networking in data centers, with an increasing focus on power efficiency and telemetry data.

And don’t miss the blogs featured in the latest Manufacturing, Packaging & Materials newsletter:

Technology Editor Katherine Derbyshire pinpoints what to watch for in lithography and patterning at SPIE and beyond.

Amkor’s Brendan Wells explains how AiP/AoP modules can replace RF SoC, baseband SoC, surface mount technology matching circuits, and a discrete antenna.

Coventor’s Sandy Wen looks at the benefits of reducing wiring congestion at the most advanced nodes.

MITRE-Engenuity’s Raj Jammy lays out what needs to be done to create a robust semiconductor workforce, and what lessons can be learned from the past.

SEMI’s Ashley Huang highlights key issues in establishing a multi-vendor chiplet ecosystem.

eBeam Initiative’s Harry Levinson summarizes the lithography challenges and advancements at the most advanced process nodes.



Leave a Reply


(Note: This name will be displayed publicly)