Intel® Movidius™ Myriad™ X VPUs

The Intel® Movidius™ Myriad™ X VPU is the third generation and most advanced VPU from Intel. It is the first of its class to feature the Neural Compute Engine – a dedicated hardware accelerator for deep neural network inferences. The Neural Compute Engine in conjunction with the 16 powerful SHAVE cores and an ultra-high throughput intelligent memory fabric makes Intel Movidius Myriad X the industry leader for on-device deep neural networks and computer vision applications. Intel’s Myriad™ X VPU has received additional upgrades to imaging and vision engines including additional programmable SHAVE cores, upgraded and expanded vision accelerators, and a new native 4K ISP pipeline with support for up to 8 HD sensors connecting directly to the VPU.

The Intel Movidius Myriad X VPU is programmable via the Myriad Development Kit (MDK) which includes all necessary development tools, frameworks and APIs to implement custom vision, imaging and deep neural network workloads on the chip.

Key Benefits
  • Dedicated Neural Compute Engine
  • 16 High Performance SHAVE Cores
  • Enhanced ISP with 4K support
  • New Vision Accelerators including Stereo Depth

Neural Compute Engine

The Intel® Movidius Myriad™ X is the first VPU to feature the Neural Compute Engine – a dedicated hardware accelerator for running on-device deep neural network applications. Interfacing directly with other key components via the intelligent memory fabric, the Neural Compute Engine is able to deliver industry leading performance per watt without encountering common data flow bottlenecks encountered by other architectures.

16 Programmable SHAVE Cores

Intel® Movidius Myriad™ X features 16 programmable VLIW vector processors, called SHAVE cores, to accelerate neural networks by performing workloads in parallel. This is 4 more programmable SHAVE cores compared to the previous generation Intel Movidius Myriad™ 2. This additional 30% of programmable compute can be used to run traditional computer vision workloads, or can complement the Neural Compute Engine by running custom layer types for CNN applications thanks extensive support for sparse data-structures.

Flexible Image Processing and Encode

The Intel® Movidius Myriad™ X VPU features a fully tune-able ISP pipeline for the most demanding image and video applications. The VPU also features hardware-based encode for up to 4K video resolution, meaning the VPU is a single-chip solution for all imaging, computer vision and CNN workloads.

Enhanced Vision Accelerator Suite

Intel has added a new suite of vision accelerators to the Intel Movidius Myriad X VPU, including a new stereo depth block that is capable of processing dual 720p feeds at up to 180Hz. With the suite of vision accelerators available, it is possible to offload key vision workloads onto fixed-function hardware, resulting in higher performance at lower price.

Neural Compute Engine: Hardware Based Acceleration for Deep Neural Networks

The Intel Movidius Myriad™X features the all-new Neural Compute Engine – a purpose-built hardware accelerator designed to dramatically increase performance of deep neural networks without compromising the low power characteristics of the Myriad VPU product line. Featuring an array of MAC blocks and directly interfacing with the intelligent memory fabric, the Neural Compute Engine is able to rapidly perform the calculations necessary for deep inference without hitting the so-called “data wall” bottleneck encountered by other processor designs. Combining the neural network performance of the 16 proprietary SHAVE cores with the neural compute engine, Intel Movidius Myriad X delivers 10X the performance compared to previous generations*.

Key Benefits
  • Native FP16 and fixed point 8 bit support
  • Rapidly port and deploy neural networks in Caffe and Tensorflow formats
  • End-to-End acceleration for many common deep neural networks
  • Industry-leading Inferences/S/Watt performance

Movidius

Maximum performance based on peak floating-point computational throughput of Neural Compute Engine. Actual results on deep neural networks may achieve less than peak throughput.

Intel® Movidius™ Myriad™ 2 VPU

Intel’s Movidius ™ Myriad™ 2 VPU is an always-on vision processor, and the second generation VPU from Intel. The Intel Movidius Myriad 2 can be found in millions of devices on the market today and is used for some of the most ambitious AI, vision and imaging applications where both performance and low power consumption are required.

Key Benefits
  • 12 Programmable SHAVE Cores
  • Ultra-Low Power Architecture
  • High Performance VPU
  • Small footprint for embedded application

Intel® Movidius™ Myriad™ 2 VPU

An ultra-low power design

For mobile and connected devices where battery life is critical, Myriad 2 provides a way to combine advanced vision applications in a low power profile. The Intel Movidius Myriad 2 VPU delivers vision capabilities to classes of devices previously unable to perform such demanding vision tasks.

Designed for vision and AI workloads

Featuring 12 VLIW programmable SHAVE cores, dedicated vision accelerators and 2 CPUS, all connected by an intelligent memory fabric, Intel’s Movidius™ Myriad™ 2 VPU is a fully functional vision SoC designed for high performance at ultra-low power. The highly parallel design and optimizations for sparse data structures make Myriad VPUs well suited to deep neural network applications and other modern vision workloads.