In an attempt to boost its chances in the AI landscape, Intel has unveiled its Movidius Myriad X, an upgraded version of Myraid 2 computer vision chipset. The chipset is designed by Movidius, the IoT company, which was acquired by Intel last year. The chip is designed to be a power efficient solution for artificial intelligence apps in devices, including drones, VR/AR headsets, and 360 -degree cameras.
Myriad X is the world’s first system-on-chip (SOC) with a dedicated Neural Compute Engine, claims Intel. Myriad X is an on-chip hardware block designed to run deep neural networks at high speed and low power without compromising accuracy, enabling devices to see, understand and respond to their environments in real time.
“We’re on the cusp of computer vision and deep learning becoming standard requirements for the billions of devices surrounding us every day,” said Remi El-Ouazzane, vice president and general manager of Movidius, Intel New Technology Group With Myriad X, we are redefining what a VPU means when it comes to delivering as much AI and vision compute power possible, all within the unique energy and thermal constraints of modern untethered devices.”
The Myriad X architecture is capable of 1 TOPS (trillion operations per second) of compute performance on deep neural network inferences. The Myraid X has a 2.5MB of on-chip memory. That 2.5MB of on-chip memory can handle up to 450 GB per second of internal bandwidth, minimizing latency and reducing power consumption by minimizing off-chip data transfer. The chip is capable of running multiple imaging apps at once and can support up to 700 million pixels per second of image signal processing.