Blog Review: April 10

Computer vision on mobile; new memories; cell-aware test.


Arm’s Paul Whatmough discusses the growing use of real-time computer vision on mobile devices and proposes transfer learning as a way to enable neural network workloads on resource-constrained hardware.

Cadence’s Anton Klotz highlights a collaboration with Imec and TU Eindhoven on cell-aware test that reduces defect simulation time by filtering out defects with equivalent fault effects.

Mentor’s John McMillan takes a look at why routing automation for PCBs is important and how intelligent routing can assist in creating fanouts, design rules, and sketch routing for multi-trace routing.

Synopsys’ Taylor Armerding listens in on what cybersecurity experts at Dow Jones and KLC Consulting see as the biggest security threats this year and what can make security a mainstream part of software development.

Applied Materials’ Gill Lee checks out what differentiates magnetic RAM, phase change RAM, and resistive RAM, three new types of memory moving toward commercial adoption.

Rambus’ Steven Woo explains two types of neural networks behind recent advancements in machine learning, spiking neural networks and generative adversarial networks.

Memory analyst Jim Handy digs into the details of Intel’s new Optane DIMM and the features that make its architecture similar to an SSD.

ANSYS’ Karthik Srinivasan points out the criticality of ensuring electromagnetic compatibility in automotive chips and the factors that are important for testing.

VLSI Research’s John West notes that both the volume and the value of critical subsystems for the sub fab are rising due to the increased complexity of leading edge fabs and the growing vacuum intensity of semiconductor manufacturing.

Nvidia’s Isha Salian points to how human rights organizations are using AI to help sort through the vast amount of satellite and drone data to identify potential conflicts.

And don’t miss the blogs featured in the latest IoT, Security & Automotive and Test, Measurement & Analytics newsletters:

Editor In Chief Ed Sperling zeroes in on why so many companies are rushing to do 7nm designs.

Achronix’s Alok Sanghavi highlights lessons newer cryptocurrencies have learned from Bitcoin that make them friendlier to FPGAs.

Synopsys’ Gordon Cooper explains how to enable realistic interactions between real and virtual objects.

Arteris IP’s Kurt Shuler compares different machine learning use-cases and the architectures being used to address them.

Mentor’s Puneet Sinha demonstrates an approach that lets powertrain designers use CFD tools from the earliest design phases.

Marvell’s George Hervey contends you can improve efficiency by taking a modern software-defined approach to data center implementations.

Editor In Chief Ed Sperling warns the tech world is planning for an onslaught of data, but it’s far from clear who will own it.

Advantest’s Judy Davies contends that the emerging machine learning techniques are pushing the boundaries of what computers are capable of.

Mentor’s Rahul Singhal explains why AI-specific processors call for design-for-test techniques that boost time-to-market.

Leave a Reply

(Note: This name will be displayed publicly)