Semiconductors
Top stories summarized by our editors
2/19/2018

Executives from Amazon, Facebook, Google and Nvidia presented perspectives on artificial intelligence and machine learning technology at the SysML symposium, held at Stanford University. "Deep learning is transforming how we design computers … [but] custom machine-learning hardware is in its infancy, so it will be an exciting time ahead with a lot of creativity in processor design," said Jeff Dean of the Google Brain team.

Full Story:
EE Times
More Summaries:
Google, Nvidia, AI, Facebook, Amazon
2/19/2018

Register now for the Memory System Tutorial on March 27 in Santa Clara, Calif., taught by seasoned industry veteran Desi Rhoden of Montage Technology. Covering every DRAM generation, this tutorial will also include system and module implications for servers and other applications. Attendees will benefit from this tutorial by gaining an improved understanding of memory technology and standards, their history, current status and future trends.

Full Story:
jedec.org
2/19/2018

Learn about a variety of market-specific memory modules in development in the industry and coordinated through JEDEC at this Tutorial on Memory Modules, Configurations & a New SPD Architecture Using MIPI I3C on March 28 in Santa Clara, Calif. It will be taught by industry experts Bill Gervasi (Nantero), Michael Joehren (NXP) and Sam Patel (IDT). See the full description and register today.

Full Story:
jedec.org
More Summaries:
JEDEC, Bill Gervasi, IDT, NXP
2/16/2018

QuTech, a research group associated with Intel, reports it was able to develop a spin-based quantum computer with two quantum bits on a single microchip. Meanwhile, physicists at Princeton University say they paired photons to electron spins, an advance that may help in creating larger quantum computers.

Full Story:
TechRepublic
More Summaries:
Intel, Princeton University
2/16/2018

Researchers are making progress in advancing machine learning and neuromorphic computing. "The most significant advances in machine learning in recent years have come from hardware improvements," writes Katherine Derbyshire.

2/16/2018

Infineon Technologies says its EconoDUAL3 modules now contain shunt resistors for AC current monitoring. The integrated shunts eliminate the need for external current sensors, the chipmaker notes.

More Summaries:
Infineon Technologies
2/16/2018

MIT researchers have crafted a neural network processor that has an up to 95% overall power decrease. This advance will enable calculations for artificial intelligence and machine learning on a single system-on-a-chip device, rather than using cloud-based servers to perform the calculations.

Full Story:
ExtremeTech
More Summaries:
MIT
2/16/2018

ON Semiconductor brought out the AR0430 complementary metal-oxide-semiconductor digital sensor, which is said to provide 2D video imaging and 3D image recognition at the same time. FRAMOS, a distributor, offers the CMOS image sensor.

Full Story:
New Electronics
2/16/2018

Diodes introduced the PT7M3808 line of microprocessor supervisory circuits, which is said to monitor system voltage from 0.4 volt to 5V. The PT7M3808 has an ultra-low quiescent current, usually around 2.8 amperes, and is available in SOT26 and DFN2020-6 packages.

Full Story:
New Electronics
2/16/2018

Ellie Yieh of Applied Materials talks about a number of topics concerning next-generation semiconductor manufacturing in this interview. "New compute and memory architectures and system-level design schemes are creating different types of challenges, which impact processing and memory," she says.

More Summaries:
Applied Materials