Linley Newsletter: May 9, 2019

 weSRCH's Best of the Internet Award
  9th-May-2019
 129


Linley Newsletter

Please feel free to forward this to your colleagues

Issue #651

May 9, 2019


Independent Analysis of Microprocessors and the Semiconductor Industry

Editor: Tom R. Halfhill

Contributors: Linley Gwennap, Mike Demler, Bob Wheeler


In This Issue:

- AMD Accelerates Ryzen Embedded

- Fabu Aims to Deliver ADAS Processors


Free White Paper:

"Arm Ecosystem Reduces SoC Design Cost and Time to Market"

The Linley Group is offering a new white paper that discusses three critical facets of the Arm ecosystem: design verification, physical design, and software development. Each one directly affects the time required to develop a complete SoC and software stack. Whereas Arm and its partners offer a wide set of tools and support, some other CPU ISAs fall short in these areas. The paper, by principal analyst Linley Gwennap, is available for free download in PDF format:

https://www.linleygroup.com/uploads/arm-ecosystem-reduces-soc-design-cost-wp.pdf


Video On Demand:

Linley Spring Processor Conference Keynotes

Keynote presentations from the Linley Spring Processor Conference are now available on our YouTube channel. Linley Gwennap provides an overview of the AI chip landscape with "A Deep Dive into Deep Learning" while Jon Masters of Red Hat delves into the grim realities of speculative execution in "After Meltdown and Spectre." Available for free on-demand viewing:

https://www.youtube.com/playlist?list=PLp4cIkwR3LXjPk-KHZ7g73uqN1lRovElp


AMD Accelerates Ryzen Embedded

By Tom R. Halfhill

Squeezing a bit more performance from its 14nm Zen CPUs, AMD is sampling two new Ryzen Embedded processors with integrated graphics and networking. These R1000-series chips -- or accelerated processing units (APUs), as the company calls them -- target casino games, digital kiosks, industrial machines, thin clients, and other embedded systems that need relatively fast GPUs and Ethernet connectivity. Both dual-core models are scheduled for production this quarter.

The new R1505G and R1606G differ only in clock speed and price. The former operates its dual CPUs at 2.4GHz (base clock) to 3.3GHz (turbo); the latter nudges those frequencies to 2.6GHz and 3.5GHz, respectively. And the R1606G's GPU is 20% faster: 1.2GHz versus 1.0GHz. AMD rates both processors at 15W thermal design power (TDP).

These R1000 products are variations of the dual-core Ryzen Embedded V1202B introduced last year. Specifically, the R1606G's CPUs are up to 13% faster and its Radeon Vega GPU with three compute units (CUs) is 20% faster. Both new chips also have dual 10 Gigabit Ethernet (10GbE) ports instead of the older model's Gigabit Ethernet (GbE). And their FP5 BGA packages make them pin compatible with the V1000-series. The main differences are that all V1000 models except the V1202B have four CPUs and up to 16 PCI Express (PCIe Gen3) lanes, whereas both R1000 models announced so far have only two CPUs and 8 lanes.

Although the new products are minor additions to the Ryzen Embedded family, they broaden the entry-level options for customers that need faster CPUs, graphics, and networking but don't need more than two dual-threaded cores. The R1000-series allows AMD to offer lower-power, lower-price products that beat Intel's graphics performance and provide faster Ethernet connectivity without an external south-bridge hub.

Microprocessor Report subscribers can access the full article:

https://www.linleygroup.com/mpr/article.php?id=12138

Fabu Aims to Deliver ADAS Processors

By Mike Demler

Autonomous-vehicle-technology startup Fabu is preparing to tape out a Level 3 ADAS processor called Phoenix-100. The chip integrates a 5.5-TOPS deep-learning accelerator (DLA) that supports cameras, lidar, and radar sensors.

The DLA enables object identification in 40fps video streams at 1080p, the most common resolution in ADAS cameras. The inference engine can classify more than 300 objects in 15 categories, such as animals, pedestrians, road signs, traffic lights, and vehicles. Phoenix-100 will integrate the DLA with multiple ISPs and a vector processing accelerator (VPA), but the company withheld architectural details. It says the processor will also support lidar and radar sensors; we therefore expect the VPA is a DSP engine needed to process the signals from those sensors before running neural-network algorithms.

Fabu and its automotive partner plan to begin production of ASIL B-compliant Phoenix-100 chips in 1Q20 for use in a 2Q20 pilot run of Level 3 passenger vehicles. Later in 2020, it expects to tape out the Phoenix-200 sensor-fusion chip and begin developing the Phoenix-300 decision processor. The design target for the entire system is 80 trillion operations per second (TOPS) with 80W of total power consumption.

Microprocessor Report subscribers can access the full article:

https://www.linleygroup.com/mpr/article.php?id=12137

About Linley Newsletter

Linley Newsletter is a free electronic newsletter that reports and analyzes advances in microprocessors, networking chips, and mobile-communications chips. It is published by The Linley Group. To subscribe, please visit:

http://www.linleygroup.com/newsletters/newsletter_subscribe.php

Domain: Electronics
Category: Semiconductors

Recent Newsletters

Linley Newsletter: June 13, 2019

Linley Newsletter Please feel free to forward this to your colleagues Issue #656 June 13, 2019 Independent Analysis of Microprocessors and the Semiconductor Industry Ed

13 June, 2019

Linley Newsletter: June 6, 2019

Linley Newsletter Please feel free to forward this to your colleagues Issue #655 June 6, 2019 Independent Analysis of Microprocessors and the Semiconductor Industry Edi

06 June, 2019

Linley Newsletter: May 30, 2019

Linley Newsletter Please feel free to forward this to your colleagues Issue #654 May 30, 2019 Independent Analysis of Microprocessors and the Semiconductor Industry Edi

31 May, 2019