Blog Review: Feb. 12

Complexity is growing by process node, by end application, and in each design. The latest crop of blogs points to just how many dependencies and uncertainties exist today, and what the entire supply chain is doing about them.

Mentor’s Shivani Joshi digs into various types of constraints in PCBs.

Cadence’s Neelabh Singh examines the complexities of verifying a lane adapter state machine in a router. Hint: It’s not simple.

Synopsys’ Jonathan Knudsen equates humans with security vulnerabilities, noting that it’s the system that’s causing the problems, not the developers.

Imagination’s Alex Willard examines a new AI marketplace for edge compute toolsets and building blocks that makes it possible to create, test and deploy edge AI in less than a day.

In a blog for ANSYS,  Chad Jackson, Chief Analyst from Lifecycle Insights, explains how to improve designs by compressing simulation analysis using multi-physics approaches, automation, data management and digital twins.

Swift Navigation CTO Fergus Noble talks about his company’s collaboration with Arm on autonomous vehicle location technology.

And in case you missed last week’s Auto, Security & Pervasive Computing newsletter, or yesterday’s Test, Measurement and Analytics newsletter, here are some blogs worth checking out:

Editor In Chief Ed Sperling contends that even though silicon photonics is a promising technology, it may take a while to catch on. He also questions the value proposition for the fastest versions of 5G.

KLA’s Ben Tsai and Cathy Perry Sullivan look at how future-node ICs for the automotive industry will fare.

ProteanTecs’ Alex Burlak digs into how advanced analytics can yield advanced visibility.

Mentor’s Jay Jahangiri explains how end-to-end automation keeps DFT out of the critical path.

YieldHub’s Marie Ryan reveals the telltale signs that prove the old way of doing things doesn’t keep up anymore.

Rambus’ Tim Messegee warns that with the vast number of connections it will enable, 5G networks will have increased attack surface that adversaries can try to exploit.

Mentor’s Doug Burcicki contends that as complexity rises in response to consumer demands, a siloed approach to vehicle development will no longer be sustainable.

Synopsys’ Stelios Diamantidis talks about how ML can enable self-optimizing tools that look for DRC hotspots, EM/IR distribution, and more.

Flex Logix’s Geoff Tate explains why all the TOPS in the world are of little use if software can’t use them efficiently.

Cadence’s Paul McLellan describes how little problems gets progressively worse over new process generations, eventually requiring integrated tools to address.