PYNQ-Z2 Dev Board: Python Productivity for Zynq® - Review

Table of contents

RoadTest: PYNQ-Z2 Dev Board: Python Productivity for Zynq®

Author: ralphjy

Creation date:

Evaluation Type: Development Boards & Tools

Did you receive all parts the manufacturer stated would be included in the package?: True

What other parts do you consider comparable to this product?:

What were the biggest problems encountered?: The Microblaze applications provided did not allow for multiple Grove interfaces simultaneously. But this was per design, so not a fault of the product.

Detailed Review:



The roadtest for the PYNQ-Z2 has been one of my most satisfying roadtest experiences.  The product is well documented and it works robustly.


I documented my roadtest in a series of blogs:

  1. PYNQ-Z2 Dev Kit - Getting Started
  2. PYNQ-Z2 Dev Kit - Xilinx/PYNQ_Workshop Introduction
  3. PYNQ-Z2 Dev Kit - Working with Base Overlays
  4. PYNQ-Z2 Dev Kit - External Peripherals
  5. PYNQ-Z2 Dev Kit - CIFAR-10 Convolutional Neural Network
  6. PYNQ-Z2 Dev Kit - CIFAR-10 Webcam continued....
  7. PYNQ-Z2 Dev Kit - CIFAR-10 Webcam revisited....
  8. PYNQ-Z2 Dev Kit - ImageNet Classification
  9. PYNQ-Z2 Dev Kit - Tiny-YOLO Object Detection




The PYNQ-Z2 Dev Kit came with everything that was required to get started immediately including a pre-flashed 16GB microSD card.  The Xilinx Zynq 7020 SoC has a dual-core ARM Cortex A9 processor and programmable logic equivalent to an Artix-7 FPGA.  With the Arduino Uno, Raspberry Pi, and PMOD compatible connectors it was very easy to interface to a wealth of existing peripherals.  Using the Jupyter Notebook environment provided rapid access to examples and documentation.   Xilinx has done a great job with their GitHub repositories of example notebooks - which I used extensively in my roadtest.  The abstraction provided by the Python wrapper libraries made it easy for a novice like me to start playing with programmable logic.  I think that this is a good entry into using FPGAs and facilitates the move to more bare metal programming.


What I really appreciated about the PYNQ-Z2 Dev Kit is that I got to play around with binarized neural network structures.  I hope this is the start of a productive learning experience.