Open Hardware


  • Home
  • Rules
  • Platforms
  • Results
    • 2022
    • 2021
    • 2020
    • 2019
    • 2018
    • 2017
    • 2016
    • 2015
  • Registration

Rules





Eligibility



Open Hardware is open to under-graduate students and PhD candidates in Europe and surrounding areas. Please see the rules for a full list of the competition rules including a list of eligible countries.


For the 2023 edition of the competition there are two main categories: "PHD" (PhD candidates), "STUDENT" for undergraduate students (Bachelors, Masters students etc). Individuals or teams of up to FIVE people can enter in each category.

For the Student category, all participants must be undergraduate students. If one or more participants are PhD level, the team must enter the PhD category. ​


All registrations must be supported by a senior member of academic staff (project Advisor) who should be from the same institute.


​





Prizes



Cash prizes of €2,000 will be awarded to the best projects in each of the PhD, and Student categories.

An additional prize will be awarded to a project from either category that makes use of AI/Machine Learning. All prizes are subject to the conditions outlined in the competition rules.

Additional prizes may be awarded at the discretion of the organiser. Special consideration will be given to projects that use a combination of heterogeneous AMD hardware; AMD CPUs, GPUs and Adaptive Computing FPGAs, SOCs, or ACAPs.

​





Machine Learning



Machine learning projects can use Vitis AI, FINN, or any other machine learning framework that can run on AMD-Xilinx adaptive compute hardware.





Vitis open source library



Participants interested in Compute acceleration are strongly encouraged to investigate the Open Source Vitis libraries covering BLAS, data compression, database, fintech, security, solvers, vision. Project suggestions for this category include improving an existing function or library component, adding a new function to an appropriate library, or use the open source library to build a compute acceleration application.


Vitis Open Source Library



Development platforms



Any Xilinx device supported by Vivado or Vitis can be used for the competition. HACC, AWS F1 instances and other cloud FPGA services can be used.


​





Judging criteria



Please pay careful attention to the judging criteria. Projects will be judged equally across all of the following categories:

  • Technical Complexity (20%)
  • Implementation (20%)
  • Marketability/Innovation (20%)
  • Documentation and Written Report (20%)
  • Re-usability (20%)

All participants will be encouraged to upload their projects or IP to GitHub or to demonstrate how the design or parts of the design can be reused. For any clarification on the Rules, please contact:


xup@xilinx.com



Entry submission



Xilinx software (Vivado, Vitis, etc.) must be used to implement the design. A complete project with bitstream must be submitted. Participants should upload complete source code and project files, a bitfile (xclbin, AWS AFI ID, etc.), or image, documentation, and a short project report written in English. Participants are also required to upload a 2-minute YouTube video, and can also upload a longer video for the judge’s consideration.


Registration closes on 28 February 2023, and final project submissions must be received by midnight (anywhere in the world) on 30 June 2023.