Ultra96 Factory Image

Base Technical Reference Design

The Base Targeted Reference Design (TRD) is an embedded video processing application running on a combination of APU (SMP Linux), RPU (bare-metal) and PL.

Tutorial 01 Build a ZU+ MPSoC Hardware Platform

The first step in creating a design for Zynq UltraScale+ MPSoC is to create the Hardware Platform in Vivado.

Tutorial 02 First ZU+ Application - Hello World

After creating the hardware platform, the next step is to import that hardware platform into SDK, create a BSP, create an application, and then run it on the board. This tutorial builds on the exported hardware platform from Tutorial 01.

Tutorial 03 Generate and Run Bare Metal ZU+ Test Applications

After Hello World is working, you can move on to more advanced applications to test the memory and all the peripherals on ZU+.

Tutorial 04 FSBL and Boot from microSD Card

In this tutorial, we will create the FSBL, and then use it to create a boot image. The boot image will then be stored on the microSD Card. Lastly, instructions are given for booting from the microSD Card.

Tutorial 01-04 Solutions

Zipped archive of the Vivado hardware platform project and the SDK Applications workspace.

PYNQ Framework for Ultra96

Accelerate your designs with PYNQ a Python friendly development framework for the ZYNQ SoC family.  Available now for Ultra96.

PYNQ Quick Start Guide for Ultra96

Video: Setting Up PYNQ On Ultra96

More about PYNQ

Xilinx GitHub for PYNQ

Avnet GitHub for Ultra96

SDSoC Baremetal Platform - Xilinx Matrix Multiply Example

SDSoC PetaLinux Platform - Xilinx Matrix Multiply Example

Accelerated Image Classification via Binary Neural Network

This page provides an introduction to the "Accelerated Image Classification via Binary Neural Network" (short AIC) design example.
This design example demonstrates how moving software implemented neural networks can be dramatically accelerated via Programmable Logic. In this design a Binary Neural Network (BNN) is implemented. Depending on silicon platform an acceleration of 6,000 to 8,000 times is demonstrated. Via the graphical user interface the user can see metrics, images and classification results.

Deephi Deep Neural Network

DNNDK™ (Deep Neural Network Development Kit) - DeePhi™ deep learning SDK, is designed as an integrated framework, which aims to simplify & accelerate DL (Deep Learning) applications development and deployment on DeePhi DPU™ (Deep Learning Processing Unit) platform. 

PetaLinux Board Support Packages

Compressed PetaLinux BSPs for Avnet Zynq system platforms.
Please CLICK HERE to access the latest BSPs for 2020 and 2019. Note: you wil be sent to our Sharepoint site.

Development Using Ubuntu Desktop Linux

These tutorials provide a means to integrate several different technologies on a single platform.  Using the Avnet target boards, we have the power of ARM processors, combined with the unrivaled flexibility of Xilinx programmable logic to implement custom hardware systems.    We use a Linux kernel as the foundation operating system running on the processor cores which enables a very large ecosystem of software to be run on our development kits. Virtual machines can provide a very convenient Ubuntu development environment for building the hardware platform and cross-compiling software to target the Processing System.