How to install OpenSesame in raspberry Pi 5

Comments

• cvanbuck

  November 19

  Hi @Romaiin

  Are you trying to design Opensesame experiments on your Raspberry Pi 5? Or do you simply want to run experiments that have been designed on another machine?

  I have never worked with a Raspberry Pi myself, but if it's the latter, Sigmund AI suggests the following workaround (see below). If you want to design your experiments directly on your Raspberry Pi, it might get a bit more complicated, although it should be possible (at least, as far as I understand it - again, I am no expert on this matter). 

  ------------- generated with Sigmund AI, our Opensesame AI assistant ---------------

  Install a runtime-only OpenSesame (no GUI) on the Pi

  If your goal is mainly running experiments on the Pi, not designing them there:

  1. Design your experiments on a desktop/laptop with the official OpenSesame installer.
  2. On the Pi, install only the core/runtime:

  python3 -m pip install --upgrade pip
  python3 -m pip install opensesame-core
  

  If that succeeds, you get the Python core and opensesamerun (command-line runner) without the Qt GUI or webengine dependency.

  You’ll also need at least one backend, e.g.:

  python3 -m pip install pygame
  # or
  python3 -m pip install expyriment
  # or
  python3 -m pip install psychopy
  

  Then you can run an experiment you created on your desktop:

  opensesamerun my_experiment.opensesame -f
  

  This avoids pyqt6-webengine entirely, so no more error — but you don’t get the graphical editor on the Pi.

  If this is acceptable, tell me:

  • your Raspberry Pi OS version (e.g. “Debian Bookworm 64‑bit”),
  • and which backend you prefer (legacy/pygame, xpyriment, psycho),

  and I can give you a very concrete “do this, then this” set of commands.

  ------------- generated with Sigmund AI, our Opensesame AI assistant ---------------

  You might also want to take a look at this post by @sebastiaan :

  The operating system on the Pi does not support OpenGL, which is the library used for hardware-accelerated graphics by OpenSesame (or actually by Expyriment and PsychoPy, which are used by OpenSesame). [...] However, OpenSesame’s non-hardware-accelerated back-end (legacy, which is PyGame based) works just fine. [...] I wouldn’t recommend developing your experiments on a Pi (although you can), but as a runtime environment it’s fast enough for most purposes.

  Hope this helps,

  Claire

  Check out SigmundAI.eu for our OpenSesame AI assistant! 🤖
• sebastiaan

  November 19

  @Romaiin Raspberry OS is a regular Linux distribution, so in principle this should work. I'm not sure why QtWebEngine (which is a library that OpenSesame uses) fails to install. Possibly this is due to the default Python version used by Raspberry OS. If so, you can try installing OpenSesame in an Anaconda environment.

  
  
  Check out SigmundAI.eu for our OpenSesame AI assistant!
  
• Romaiin

  December 1

  Hello,

  Thank you for your previous responses, they were very helpful.

  I followed your instructions and restarted the installation from scratch on both Raspberry Pi 4 and Raspberry Pi 5. However, during the execution of the command:

  pip install opensesame-core
  

  the installation consistently fails at the point where PyQt6-WebEngine is retrieved. The issue appears to be related to ARM64 / PyQt6-WebEngine incompatibility on Raspberry Pi OS. The build process is unable to generate the metadata (pyproject.toml), and this causes the installation to abort.

  This confirms that the issue is not directly related to OpenSesame itself, but rather to the fact that PyQt6-WebEngine is not available or supported on Debian ARM64, regardless of the Raspberry Pi model.

  In my specific use case, the Raspberry Pi is not intended to design experiments, but only to act as a client within an OpenMonkeyMind network, executing tasks that are developed on another machine and then deployed to the clients.

  I will now explore installation via Anaconda, as it seems that certain Qt-related packages are better supported on ARM through conda-forge.

  Thank you again for your guidance and support.