Blog Review: Sept. 22

Ansys’ Tyler Ferris describes some of the many ways electronics on a PCB assembly can fail, from component level failures like wirebond breaking and liftoff to board-level failures such as conductive anodic filament failure.

Cadence’s Paul McLellan considers the switch from low-speed parallel interfaces to high-speed serial interfaces as one of the key advancements making modern data centers possible.

Siemens EDA’s Ray Salemi explores refactoring UVM in Python to simplify the code and take advantage of Python features such as multiple inheritance and the differences from SystemVerilog.

Synopsys’ Blake Crowther provides a look at the optical design of the zoom camera lenses for the Mars Perseverance rover and some of the challenges faced.

Xilinx’s John Bieker provides some tips on how to improve floorplans in Vivado to help reduce routing congestion and improve the quality of timing results.

Arm’s Girish Shirasat examines what needs to happen to make software-defined vehicles a reality and the importance of cloud-native technologies.

Infineon’s Olaf Bendix provides some guidelines for developing a gate driver for silicon carbide MOSFETs, including identifying peak current, selecting gate resistor value, and determining expected power dissipation.

Nvidia’s Jean Kossaifi and Christopher Choy explain how tensors are used in machine learning to improve generalization, robustness, and speed up computation, and share some tools to help implement tensor-based deep neural networks.

SEMI’s Serena Brischetto chats with Aurelie Ludemann of EMD Electronics about the role of diversity and inclusion programs in strengthening the advanced materials and manufacturing industry workforce.

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

Executive editor Mark LaPedus sees that silicon wafer vendors are responding to shortages with new capacity.

Brewer Science’s Jessica Albright explains why it’s worth tech companies aiming for environmental and social responsibility.

SEMI’s Pushkar Apte and Tom Salmon conclude that from ecosystem development to talent, much effort is still required for practical implementation of edge AI.

TechInsight’s Stacy Wegner looks inside the Fitbit Luxe and Charge 5.