Open Hardware



Platforms





For full details of eligible hardware see the competition rules.

Ryzen AI, selected AMD GPUS, and the following Academic boards, Alveo and partner platforms are recommended.





Ryzen AI and selected GPUs



The following hardware is recommended for the Accelerated Computing category



AMD Radeon GPUs



The AMD Radeon GPU is well know by gamers and can also be used for general purpose compute, including AI/ML. ROCm is an open software stack that includes a broad set of programming models, tools, compilers, libraries, and runtimes for AI and HPC solution development on AMD GPUs can also be used for general purpose compute, including Machine Learning. Any Radeon supported by ROCm can be used for the competition.



Ryzen AI



Ryzen AI is a Neural Processing Unit or NPU, optimized for machine learning. Projects using Ryzen AI, found inside selected Ryzen 7000 and 8000 processors are eligible for Open Hardware.



AMD Instinct Accelerators



AMD Instinct™ accelerators are designed to deliver high-performance computing in data centers, catering to various scales, from single-server setups to the largest Exascale-class supercomputers. These accelerators excel in handling demanding artificial intelligence (AI) and high-performance computing (HPC) workloads, providing outstanding compute performance, substantial memory density, high bandwidth memory, and compatibility with specialized data formats. Projects using any AMD Instinct accelerators are eligible for Open Hardware.





Adaptive Computing Academic Platforms



The AMD University Program works with partners to make available a selection of boards for university teaching and research projects. See the selected boards below or visit www.amd.com/aup

The following boards are recommended for the Adaptive Computing category:



Pynq-z2 tul



The PYNQ-Z2 from TUL is the recommended platform for getting started with PYNQ. The PYNQ-Z2 includes a Zynq Z7020, 512 MB DDR3 memory, 2x Pmods, Arduino, and Raspberry Pi headers. For more information, please see the TUL PYNQ-Z2 page.



The boolean board



  • A Xilinx Spartan 7 XC7S50 FPGA
  • A USB port for power & programming
  • An on-board Bluetooth Low Energy radio
  • A HDMI source port that can drive up to 1080p
  • An on-board analog-to-digital converter
  • A large QSPI ROM for storing FPGA configurations and user data
  • Four servo connectors (conventional or continuous rotation)
  • An 8-digit seven-segment display
  • Four Pmod®-compatible expansion connectors

For more information see the RealDigital page.



Pynq-zu



TUL PYNQ-ZU board, based on Xilinx Zynq UltraScale+ MPSoC, is designed for the Xilinx University Program to support the Xilinx PYNQ open source framework and embedded systems development. Features:

  • Quad core ARM Cortex-A53.
  • Dual R5 processors DDR4 memory controller with 8 DMA channels and 4 High Performance AXI3 Slave ports.
  • 1G Ethernet, 2x USB 3.0, DisplayPort.
  • HDMI 2.0 sink port (input), HDMI 2.0 source port (output), Mini Display Port.

For more information see the PYNQ-ZU page.






Adaptive Computing acceleration platforms



AMD Alveo are PCIe boards intended for adaptive compute acceleration. Remote access to Alveo and Versal hardware is available to academics through the HACC research program. For more information see the AMD Alveo webpage.



AMD VErsal



Versal is the latest Adaptive SoCs from AMD. Open Hardware teams can use any Versal platform including the VCK190, VCK5000, and the VEK280.



AMD Alveo



AMD® Alveo™ Data Center accelerator cards are designed to meet the constantly changing needs of the modern Data Center.


For more information, see the AMD Alveo™ page.





Selection of partner boards



Red PiTaYa



The Red Pitaya is a credit card-sized, open-source test and measurement board that can be used to replace most measurement instruments used in electronics laboratories. The device, based on Linux, includes an FPGA, digital signal processing (DSP), dual core ARM Cortex processor, signal acquisition and generation circuitry, micro USB socket, microSD card slot, RJ45 socket for Ethernet connection, and USB socket – all powered from an external mains adaptor. Red Pitaya offers free loan boards for use in the Open Hardware design competition. For more information, please see the Red Pitaya page.



Avnet Ultra96



The Ultra96 from Avnet is a Zynq Ultrascale+ board supported by PYNQ. The board features

  • A Zynq UltraScale+ MPSoC ZU3EG
  • 2 GB DDR4 Memory
  • Wi-Fi / Bluetooth, Mini DisplayPort, USB 3.0.

For more information, please see the Avnet Ultra96 page.



Sundance VCS-1



The Sundance VCS-1 is a PC/104 Linux stack composed of the EMC2 , and FM191 expansion card. EMC2 is a PCIe/104 OneBank™ carrier for a Trenz compatible SoC Module. Sundance are also offering a Lynsyn kit giveaway. Lynsyn measures power for sections of code helping you make your design as power efficient as possible.

For more information, please see the Sundance VCS-1 page.



digilent



A range of XUP boards with academic discount are available from Digilent, including the Basys 3, Nexys 4 DDR, Zybo, and Zedboard. Register with Digilent to apply for academic pricing.

For more information, please see the Digilent page.



UltraZohm



UltraZohm is an open-source rapid-control prototyping system for power electronics and drive systems. It uses the Zynq UltraScale+ and adds interfacing hardware for power electronics. UltraZohm let's you move quickly from simulation to testing your control algorithm in hardware.
UltraZohm Documentation