Les commandes expédiées à l’extérieur des États-Unis seront en attente pendant que l’entreprise évalue les impacts tarifaires.
Veuillez contacter votre vendeur pour obtenir de l’aide.
Microchip PolarFire & PolarFire SoC FPGAs
Lowest Power, Cost-Optimized, Mid-Range FPGAs & AI/Machine Learning Solution
Award-winning PolarFire & PolarFire SoC FPGAs by Microchip deliver the industry’s lowest power at mid-range densities with exceptional security and reliability. The product family spans from 48K logic elements (LEs) to 500K LEs, features 12.7G transceivers and offers up to 50% lower power than competing mid-range FPGAs. The devices are ideal for a wide range of applications within wireline access networks and cellular infrastructure, defense and commercial aviation markets, as well as industrial automation and IoT markets. Get started with your design today by downloading the Libero SoC PolarFire Design Suite and ordering a PolarFire development kit that best matches your needs. Get started with your design today with the VectorBlox™ Accelerator Software Development Kit for PolarFire® FPGAs, or by downloading the Libero SoC PolarFire Design Suite and ordering a PolarFire development kit that best matches your needs.
Highlights
- Industry’s Best FPGA Security - Microchip’s PolarFire & PolarFire SoC FPGAs represent the industry’s most advanced secure programmable FPGAs
- Industry’s Best FPGA Security - Cyber Security is the #1 Concern for Connected Devices on the Network Edge
- Smallest Form Factors, PolarFire & PolarFire SoC FPGAs offer best-in-class form factors at 100K, 200K, and 300K LES
FeaturesCost-Optimized Architecture
| Power Optimization
|
Applications
|
|
PolarFire FPGA Architecture
Avalanche Development Kit
Platform with Microchip MPF300TS 300kLE PolarFire FPGA and VSC8531 Gigabit Ethernet PHY, Panasonic PAN9420 WiFi module, and Arduino/mikroBUS/PMod connectors, ideal for object detection and monitoring in smart embedded vision applications.
Northern Lights Demonstration Kit
Platform combines machine learning with vision and digital signal controller technologies to provide an all-in-one demonstration for object recognition-based motor control applications. It includes a Video Board with a color CMOS image sensor and a low cost, low power FPGA; a Display Module with color OLED panel; and a Motor Control Board with a digital signal controller.
VectorBlox™ Accelerator Software Development Kit for PolarFire® FPGAs
The VectorBlox Accelerator Software Development Kit (SDK) offers the most power-efficient Convolutional Neural Network (CNN)-based Artificial Intelligence/Machine Learning (AI/ML) inference with PolarFire FPGAs. VectorBlox software enables:
- OpenVINO™ toolkit-based front-end tools
- Support for most common frameworks like TensorFlow, Caffe, MxNet, PyTorch and DarkNet
- Quick evaluation without prior FPGA knowledge
- Software-overlay-based implementation; there's no need to reprogram the FPGA while updating CNNs
The VectorBlox Accelerator SDK contains different tools that compile a neural network description from frameworks like TensorFlow and ONNX into a Binary Large Object (BLOB). These BLOBs are stored in Flash and loaded into the DDR memory during execution.
- Model Optimization: Converts a trained network to a common Intermediate Representation (IR) and cleans the network for inference by removing several layers used during training (e.g., batching operators and fuse layers)
- Quantization: Converts optimized networks from FP32 to INT8 and enables networks to be represented using less memory with minimal loss in accuracy
- Calibration: Involves adjusting activations and weights of a model represented in INT8 precision
- Runtime Generation: Creates a BLOB that is written into embedded nonvolatile storage (e.g., SPI Flash)
