Author's Latest Posts


Week In Review: Design, Low Power


Tools & IP Rambus debuted 112G XSR/USR PHY IP on TSMC's N7 7nm process. The PHY IP enables die-to-die and die-to-optical engine connectivity for chiplets and co-packaged optics targeting data center, networking, 5G, HPC, and AI/ML applications. It has been demonstrated in silicon to exceed the reach/BER performance of the CEI-112G XSR specification and supports NRZ and PAM-4 signaling at v... » read more

Blog Review: June 17


Mentor's Chris Spear provides an introduction to SystemVerilog Multidimensional Arrays and shares code samples to follow along. Cadence's Paul McLellan listens in on Sophie Wilson's 2020 Wheeler Lecture that traces the history of the microprocessor from the early days of Moore's Law through to increasing power and economic constraints that are causing a transition from general purpose to spe... » read more

Power/Performance Bits: June 16


One-directional optical Researchers from University of Pennsylvania, Peking University and Massachusetts Institute of Technology developed a design for optical devices that radiate light in only one direction, which could reduce energy consumption in optical fiber networks and data centers. Light tends to flow in a single direction optical fibers, but while most of the light passing through... » read more

Week In Review: Design, Low Power


Synopsys acquired Qualtera, a provider of big data analytics for semiconductor test and manufacturing. Based in Montpellier, France and founded in 2010, Qualtera's Silicondash platform provides both off-line and in-line modules for data analytics, visualization, simulation, and modeling to allow for development of control strategies. Combined with Synopsys' TestMAX test automation solution, the... » read more

Blog Review: June 10


Cadence's Paul McLellan considers the issues around benchmarking neural networks running on different hardware and challenges in comparing designs. Mentor's Shivani Joshi points to a few of the different types of jitter and some key factors to review when trying to limit jitter. Synopsys' Fred Bals notes that while the National Vulnerability Database is a good source for information on public... » read more

Startup Funding: May 2020


It was a good month for semiconductor startups, with investment spanning a larger company in later funding rounds to brand new seed funding for two chip manufacturing startups. Two AI hardware startups bridge data center and edge, plus EV companies around the world get funding. In total, the eighteen startups profiled this month raised $446.3 million. Semiconductor & design Shanghai-based ... » read more

Power/Performance Bits: June 8


High temp capacitor Researchers at Pennsylvania State University doped a dielectric capacitor to increase storage capacity while also increasing electric charge efficiency, enabling the capacitor to withstand greater voltage with very little energy loss at temperatures higher than 300 degrees Fahrenheit. “What we have done is to use interface effects in nano-dopants to increase both the stor... » read more

Week In Review: Design, Low Power


Tools & IP Synopsys introduced its DesignWare USB4 IP solution consisting of controllers, routers, PHYs, and verification IP. It supports USB4, DisplayPort with HDCP 2.3 security, PCI Express, and Thunderbolt 3 connectivity protocols through USB Type-C connectors and cables. The USB4 IP operates at up to 40 Gbps, twice the maximum data rate of USB 3.2, and is backwards compatible with USB 3... » read more

Blog Review: June 3


Cadence's Paul McLellan takes a look at how Ethernet came to dominate wired networking and is now taking on automotive to provide the bandwidth necessary for the increasing number of sensors in modern vehicles. Mentor's Colin Walls notes the difficulty of assessing the quality of software, some key areas to pay attention to when assessing quality or trying to write quality code, and the bottom... » read more

Power/Performance Bits: June 2


Neuromorphic memristor Researchers at the University of Massachusetts Amherst used protein nanowires to create neuromorphic memristors capable of running at extremely low voltage. A challenge to neuromorphic computing is mimicking the low voltage at which the brain operates: it sends signals between neurons at around 80 millivolts. Jun Yao, an electrical and computer engineering researcher at ... » read more

← Older posts Newer posts →