Eta Compute Named in Forbes top IoT Startups 2020

Eta Compute Named in Forbes top IoT Startups 2020

Eta Compute has been able to adapt quickly to the COVID-19 hard reset by offering a variety of licensing models designed to meet the unique needs of each industry they serve.

Louis Columbus, Forbes Senior Contributor

Use the below link to view the full story. Eta Compute is labeled number 6 in the top 20.

Eta Compute’s ECM3532 cover as cool product by Ganssle Group

This Week’s Cool Product
I’ve been following Eta Compute for a while. Like some other vendors, their interest is putting a little machine learning capability at the “edge” – in a microcontroller right at the sensors. What makes them a bit different is their focus on doing this at extremely low power levels.Eta’s new ECM3532 is an MCU with a Cortex M3 and DSP with dual MACs. The latter, of course, is optimized for multiply-accumulate operations, which are at the heart of inferencing. The company claims (I’ve seen their eval boards running off 1 cm2 solar cells) power consumption of under 5 µA/MHz, which is around an order of magnitude better than most ultra-low-power MCUs. Instead of dynamic voltage and frequency scaling, they tout continuous voltage and frequency scaling, with a near-threshold low-end Vdd of 0.54 V.Near- and sub-threshold operation yields very low power operation at low frequencies. Ambiq is one of the pioneers in this area. FETs can operate in three regions: subthreshold, linear, and saturated. But in the subthreshold region leakage is a problem, and temperature even more so. Transistors are not well characterized at those voltage levels, so vendors resort to clever, often secret, tricks. I’m told digital watches, at least the non-smart versions, operate in the subthreshold region giving them tremendous battery lives. For more on subthreshold voltage operation of FETs see this.One of those tricks Eta uses is a non-synchronous architecture. Though there is a clock, the logic isn’t conventionally clocked. It’s self-timed. Synchronous circuits need plenty of margin to insure all of the timing is properly closed; self-timed versions report back as soon as an operation has completed so the next can begin. Absent this, since FETs operating in or near the subthreshold region can have wildly-varying characteristics depending on temperature, huge

Eta Compute named Cool Vendor by Gartner

Eta Compute Named as a Cool Vendor in Gartner’s May 2020 Cool Vendors in IoT Thingification Report

WESTLAKE VILLAGE, Calif., May 27, 2020 – Eta Compute Inc., a company dedicated to delivering machine learning to IoT and edge devices using its revolutionary Neural Sensor Platform, announced that it was recognized in Gartner’s May 2020 Cool Vendors in IoT Thingification* report.

“We are delighted to be named a Cool Vendor by Gartner,” said Ted Tewksbury, CEO of Eta Compute. “Our innovative multicore processor and self-timed CVFS technology are changing the way designers approach machine intelligence for energy-constrained products. In our opinion this Gartner recognition  validates our vision of enabling intelligence everywhere in the smallest embedded devices and making IoT a reality.”

Eta Compute’s unique approach has been to attack constraints directly by reinventing the way intelligent devices are architected and designed. The company leverages its transformative low power technology and hybrid multicore design to address performance, power and cost constraints. With its patented Continuous Voltage Frequency Scaling (CVFS) technology and production of its ECM3532, the world’s first true AI multicore processor, the company has solved the hardware issues hindering the wide deployment of AI at the extreme edge.

Gartner subscribers can view the full Cool Vendors in IoT Thingification report at this link.

*Gartner, “Cool Vendors in IoT Thingification,” Bill Ray, Alan Priestly, George Brocklehurst, and Martin Reynolds, 14 May 2020

Disclaimer: Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner’s research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of

Eta Compute and Edge Impulse accelerate development and deployment of machine learning at the edge

WESTLAKE VILLAGE, Calif., May 12, 2020Eta Compute and Edge Impulse announce that they are partnering to accelerate the development and deployment of machine learning using Eta Compute’s revolutionary ECM3532, the world’s lowest power Neural Sensor Processor, and Edge Impulse, the leading online TinyML platform. The partnership will speed the time-to-market for machine learning in billions of IoT consumer and industrial products where battery capacity has been a roadblock.

“Collaborating with Edge Impulse ensures our growing ECM3532 developer community is fully equipped to bring innovative designs in digital health, smart city, consumer, and industrial applications to market quickly and efficiently,” said Ted Tewksbury, CEO of Eta Compute. “We believe that our partnership will help companies debut their ground-breaking solutions later in 2020.”

Eta Compute’s ECM3532 ultra-low power Neural Sensor Processor SoC that enables machine learning at the extreme edge, and its ECM3532 EVB evaluation board are now supported by Edge Impulse’s end-to-end ML development and MLOps platform. Developers can register for free to gain access to advanced Eta Compute machine learning algorithms and development workflows through the Edge Impulse portal.

“Machine learning at the very edge has the potential to enable the use of the 99% of sensor data that is lost today because of cost, bandwidth, or power constraints,” said Zach Shelby, CEO and Co-founder of Edge Impulse. “Our online SaaS platform and Eta Compute’s innovative processor are the ideal combination for development teams seeking to accurately collect data, create meaningful data sets, spin models, and generate efficient ML at a rapidly accelerated pace.”

“Trillions of devices are expected to come online by 2035 and many will require some level of machine learning at the edge,” said Dennis Laudick, vice president of marketing, Machine Learning Group, Arm. “The combination of Eta Compute’s TinyML hardware based on Arm®

Semir Haddad to go live on SemiWiki

White Paper Registration

Fill out the below form and we will send you a copy of the white paper.
  • Please use your company email (no Gmail, Yahoo, Hotmail, or AOL addresses will be accepted).
  • This field is for validation purposes and should be left unchanged.

Semiconductor Digest March Edition – exerpt

Deep Learning at the Extreme Edge: A Manifesto
By Gopal Raghavan, Ph.D., Co-founder, Chief Technologist and Semir Haddad, Senior Director Product Marketing, Eta Compute, Westlake Village, CA
With continuous voltage and frequency scaling, the logic is self-timed, adjusting core voltage and clock frequency automatically, ensuring no timing violation. This allows the chip to always run the most efficient way for a given workload.

White Paper Registration

Fill out the below form and we will send you a copy of the white paper.
  • Please use your company email (no Gmail, Yahoo, Hotmail, or AOL addresses will be accepted).
  • This field is for validation purposes and should be left unchanged.