This is a general question. We are developing tasks for a multi-site project (Psychopy v2022.5). This project involves more than 100 sites from different countries with a variety of computer settings and tech specs. From pilot studies, we have already encountered multiple issues:
Some computers don’t run if keyboard backend is specified as ioHub instead of PsycToolbox.
Some tasks crash when they are in full-screen mode but will work if they are not.
Images are presented in different sizes.
Unfortunately, as some experimenters will run the tasks in public spaces (e.g., museums), running the tasks on Pavlovia may not be feasible.
Thus, my question is whether there are good practices or suggestions to make tasks as robust as possible so they can run relatively stable across multiple computer settings.
The more generic the question the harder it is to come up with suggestions that target issues.
But if I were in your position I’d try to specify as much as I can in code components, for example, for (1) you can always make PsycToolbox the default keyboard. For (2) I would provide a sample code (and a sort of instructions of how to do so) that people can edit and input the details corresponding to their displays they have on site, and for (3) it depends how you’re calling in the images, but if the proportions are specified, via the “Size [w,h]” field, it can be quite consistent across different screens (in conjunction with the suggestion for 2), un less the monitors have very different aspect ratios (most monitors these days will be wide screen 16:9, but it they are not, again, what I’d do is provide a list of [w,h] values for the aspect ratios that are being used and instruct people how to change this).