21/01/2019
In not much more than a decade, deep learning has moved from a research curiosity to a technology that could underpin a new generation of autonomous vehicles and robots: machines that can respond more intelligently to the world they perceive.
|
|
|
29/10/2018
Artificial intelligence (AI) has kindled the imagination of computer scientists for decades and while the ambition and enthusiasm around AI has tended to clash with the complexity of the task, today’s computational power has risen exponentially and the ambition of general AI has been curbed sufficiently to match that power.
|
|
|
12/09/2018
Twenty years ago, it looked like a concept that was ready for primetime: putting programmable logic inside ASICs and SoCs. At the time, the move seemed inevitable. ASIC mask prices were rising fast, driven by the need to pull more and more from a bag of optical tricks to keep Moore’s Law on track. The cost of respins alone seemed enough to persuade designers to leave some reprogrammable “sewing kits” in their SoCs to let them iron out bugs after tapeout instead of committing tens of thousands of dollars more to the project to get some new masks.
|
|
|
11/05/2018
How are embedded system designers addressing the incredible complexity associated with machine learning?
|
|
|
10/05/2018
Poor battery life is affecting the take-up of too many devices. How can power be analysed in enough detail to ensure products live up to consumer expectations?
|
|
|
25/04/2018
With the growth in more subtle attacks, SoC and system designers now need to address the problems of secure execution up front.
|
|
|
25/04/2018
Demand is growing for faster processor architectures to support embedded vision and artificial intelligence.
|
|
|
18/04/2018
Here, Rich Miron, Applications Engineer at Digi-Key Electronics explores the operation of switch mode power supplies and explores make Vs. buy decision process for power supplies. Miron also investigates the design of a single output supply utilising flyback topology and provides a sample design using readily available parts and components.
|
|
|
11/04/2018
Now more than 30 years old, the field-programmable gate array has evolved from a glue-logic device that made it possible to customise boards easily to a complete configurable system-on-chip (SoC). Hardwired 64bit processors, digital signal processing (DSP) engines and dedicated memory arrays have helped overcome the FPGAs density handicap versus fully custom silicon even for projects that expect to move into high volume.
|
|
|
16/03/2018
While the debate about VME’s future continues, companies are developing new VME products and customers are still specifying the standard.
|
|
|
14/03/2018
The intelligent connected world needs adaptable accelerated computing. As a result, more engineers are turning to FPGA as a Service providers via the cloud.
|
|
|
14/03/2018
The search for a breakthrough display technology that addresses the needs of next generation products could be over.
|
|
|
23/01/2018
Trojans aren’t restricted to the software domain; in some instances, hardware Trojans could even open backdoors in custom silicon.
|
|
|
15/11/2017
As the deployment of Industrial IoT systems continues to proliferate, the streams of data transferred to the cloud skyrockets, drastically increasing the cost for cloud computing.
|
|
|
11/10/2017
Tired of the countless articles talking about the Industrial Internet of Things (IIoT)? Surely all of the pundits, industry-leading companies, and technology providers excited to share their perspective on the growing impact of the IIoT have exhausted the topic. Instead of focusing on what the IIoT is, this article takes the opposite approach and talks about what the IIoT isn’t. Let’s be honest, the IIoT isn’t defined. It isn’t a known target with a clear set of parameters and rules. But there’s one thing we do know—as we build and define the IIoT, it’s critical that providers overdesign their technology offerings for flexibility.
|
|
|
27/09/2017
Artificial Intelligence (AI) and deep learning is the future of computing. Intelligent machines that understand the world as humans do, interpret our languages and learn from data will habitually be used to resolve problems too complex for the human brain.
|
|
|
27/09/2017
Chipmakers are set to break the 10nm barrier as they move from the test-chip stage to full production on 7nm processes. The move marks another extension of a lithography technology that was meant to be phased out 15 years in favour of so-called next-generation lithography.
|
|
|
01/08/2017
The traditional development flow of an all programmable Zynq SoC segments the design between processor system and programmable logic. The Zynq is a complex heterogeneous system which combines advanced ARM dual core Cortex-A9 processing systems with programmable logic. This programmable logic provides not only the traditional Flip Flops and Look Up Tables but also block RAM and distributed RAM, DSP Slices, PCIe endpoints and multi-gigabit transceivers. Users need a development environment which enables them to exploit the capability provided by both the processor and the programmable logic.
|
|
|
27/06/2017
The latest ‘buzz phrases’ to emerge into general use are ‘machine learning’ and artificial intelligence, or AI. Both techniques have wide application, so it’s no surprise to find technology companies all of types pursing the opportunities.
|
|
|
12/04/2017
Device developers have always looked to provide some degree of programmability in their products. But, with the exception of FPGAs and a few other parts, the components which developers have available to them offer fixed functionality.
|
|
|
29/03/2017
FPGA technology has seen a significant change in the last two decades. From relatively simple devices used mainly as glue logic and for last minute board fixes, FPGAs have evolved into highly complex parts.
|
|
|
19/01/2017
The application areas for embedded GPUs have expanded from smartphones and tablets, via TVs and vehicle infotainment systems, to sectors such as the IoT, wearables, VR/AR and servers.
|
|
|
14/12/2016
While there’s nothing new about the use of FPGAs in electronics products, many engineers are only just beginning to explore how the devices could help to improve their designs.
|
|
|
22/11/2016
FPGAs are rarely out of the news, but the acquisition of Altera by Intel in 2015 pushed the technology firmly into the headlines. Intel’s main reason for buying Altera was to provide a way to accelerate the performance of its Xeon processors, used widely in data centres. But Intel hasn’t been the only company turning its attention to the use of FPGAs in such applications; earlier in 2015, Microsoft said that Project Catapult used Altera’s Arria 10 FPGAs to boost data centre power/performance.
|
|
|
22/11/2016
As the landscape of distribution continues to change with the prospect of further mergers and acquisitions, the rise of the Maker movement and megatrends such as the IoT, distributors are re-assessing how they interact with design engineers.
|