Power Integrity Analysis For High-Performance FPGAs


Efinix high-performance Titanium field-programmable gate arrays (FPGAs) are custom-tailored for the computing demands of mainstream applications, targeting markets from intelligent edge devices to industrial automation to vision systems to edge servers and communications (figure 1). Efinix customers use the Titanium line of FPGAs to ensure their complex, high-performance designs minimize power ... » read more

Easing The CFD Engineer’s Life With Automated Meshing


Mesh generation is where the user’s expertise and ingenuity can influence the convergence and accuracy of a computational fluid dynamics (CFD) solution by selecting mesh type, topology, and cell quality. But with the rush to automate mesh generation, will the control be ripped out of the user’s hand, or will a valuable engineering skill be lost? The extent to which meshing can be automated... » read more

Brrraaap: Full Throttle On Dirt Bike Electrification


On a dirt bike almost any terrain is fair game. Once you hit the back roads, course correcting through open fields or wooded areas to get where you’re going (or not) is easy. These powerful-yet-agile gravel-and-dirt surfing, rock-hopping machines have been identified by some as the undisputed off-road champions — perfect for exploration in places larger vehicles simply can’t go. Dirt b... » read more

Efinix Implements Effective EM/IR Analysis For Leading-Edge FPGA Designs With The MPower Platform


Efinix turned to the Siemens mPower power integrity analysis platform to obtain the capabilities they needed for fast, accurate, full-chip EM/IR analysis of their Titanium FPGA designs. With no artificially enforced digital methodology, and a flat transistor analysis without elaborate views or modeling, the mPower platform analyzes custom layout and P&R IP in a single, seamless run. The mPo... » read more

What Designers Need To Know About USB Low-Power States


In addition to performance and interoperability, achieving low power has been one of the requirements for industry standards specifications. Some of the key specifications like Universal Serial Bus (USB), PCI Express (PCIe), and MIPI have defined power saving features for burst traffic. This whitepaper explains how Synopsys USB IP offers low power using various low power states that go beyond t... » read more

Multiphysics Analysis of Traction Motors: Accelerating Innovation in Electric Mobility Solutions


Between 2021 and 2023, electric vehicles (EVs) represent a $7 trillion market opportunity, and it’s only projected to increase. In a rapidly growing EV market, automakers are turning to advanced design, testing, and manufacturing technologies. Optimized system design requires the evaluation of many different concepts, topologies, and electronic interactions across disciplines. Design and s... » read more

On The Reverse Breakdown Behavior Of GaAs PIN Diodes For High Power Applications


In the field of power electronics, the compound semiconductors gallium nitride and silicon carbide are dominating the market. Due to its beneficial properties, gallium arsenide is gaining more and more importance. The aim is to manufacture devices based on gallium arsenide for use in power electronics with comparable or better properties, but at lower costs. In this work, a first GaAs PIN diode... » read more

What’s The Difference Between An NPU And A GPNPU?


To understand the difference between an NPU (neural processing unit) and a GPNPU (general-purpose neural processing unit) let’s start with the NPU, a processing engine that accelerates machine learning (ML) workloads in System on Chip (SoC) designs. Click here to read more. » read more

Achieving Greater Accuracy In Real-Time Vision Processing With Transformers


Transformers, first proposed in a Google research paper in 2017, were initially designed for natural language processing (NLP) tasks. Recently, researchers applied transformers to vision applications and got interesting results. While previously, vision tasks had been dominated by convolutional neural networks (CNNs), transformers have proven surprisingly adaptable to vision tasks like image cl... » read more

Choosing The Right Memory At The Edge


As the amount of data produced by sensors in cars and phones continues to grow, more of that data needs to be processed locally. It takes too much time and power to send it all to the cloud. But choosing the right memory for a particular application requires a series of tradeoffs involving cost, bandwidth, power, which can vary greatly by device, application, and even the data itself. Frank Fer... » read more

← Older posts Newer posts →