Tech

Startup taps GlobalFoundries to produce 166x more efficient embedded chips

What just happened? Embedded computing is poised for a major efficiency boost thanks to a new partnership between chip manufacturing giant GlobalFoundries and Silicon Valley startup Efficient. The two companies are collaborating to produce a new generation of ultra-low-power processors, which could revolutionize edge devices by eliminating battery life concerns.Startup taps GlobalFoundries to produce 166x more efficient embedded chips

The new chip architecture, unveiled by Efficient earlier this year, promises performance gains of up to 166x over today’s embedded CPUs while consuming only a fraction of the power. If the chip lives up to these claims, it could be a game-changer for industries such as smart gadgets, IoT sensors, space sensors, wearables, and implantable medical devices – key markets Efficient is targeting.

Under the partnership, Efficient will provide the architecture, while GlobalFoundries will handle manufacturing at its US fabs using its 22nm FDX platform. The duo expects early customer sampling to begin in the summer of 2025.

Instead of relying on general-purpose CPU cores, which Efficient describes as “over-designed for generality” and inefficient due to “unnecessary internal data movement and instruction control overheads,” the company has reimagined processor design from the ground up.

Its “Fabric” architecture eliminates power-hungry complexities by providing “reconfigurable hardware at compile time,” which can reduce power consumption by up to 99 percent without compromising performance. Efficient aims to create a new category of ultra-low-energy, high-performance processors that bring advanced edge computing capabilities to use cases previously limited by strict energy constraints.Startup taps GlobalFoundries to produce 166x more efficient embedded chips

You can read more Technology articles

See also  Apple is ready for September launch of iPhone 16, new wearables, and new Airpods

Developers can write code for Fabric processors using familiar languages like C and TensorFlow Lite, avoiding the need for specialized domains or instruction sets. This should ease the adoption curve as companies seek to incorporate AI, ML, signal processing, and data analytics into their edge products.

Brandon Lucia, Efficient’s co-founder and CEO, emphasized that partnering with a seasoned semiconductor foundry like GlobalFoundries was crucial to “empower the world’s most critical industries to embrace the next evolution of energy-efficient computing with ease and reliability.”

GlobalFoundries’ SVP of end-markets, Faisal Saleem, echoed that enthusiasm, expressing excitement about collaborating with Efficient’s “brilliant team of engineers” to address the “growing market need for ML-enabled intelligent edge devices across various domains that are power efficient and deliver strong computing performance.”

Efficient emerged from stealth mode in March after securing $16 million in seed funding. At that time, the company claimed its Fabric processors were 100 times more efficient than “the best embedded von Neumann processor” and 1,000 times more economical than power-hungry GPUs.

Follow HiTrend on X

Rate this post

Trend Admin

Stay with us for all the trend news of the day

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button