I’m using two tasks I’ve built and run in Psychopy3 2020.2.10 on a Windows 10 hp laptop. I have been running both tasks without any issue in a study for over a year now. In the last 4 weeks, we’ve started having issues with Psychopy crashing suddenly.
This doesn’t happen every time, but about half of the time Psychopy will completely shut down after being open for about minute. No error messages or warning signs, it just shuts down. I have tried restarting the laptop, and that doesn’t fix it. It’s a toss-up on any particular day if we will have this issue or not. Again, we have been running this experiment for a while with no issues, we have recently moved location but none of the equipment has changed and it was working correctly for a couple of weeks after moving.
Has anyone else had this issue or have any suggestions for fixing it? I want to avoid updating Psychopy if I can, in case that brings up new problems in the coding of the tasks.
I do not know if my answer is really helpful. However, I noticed that, for example, the programs Opensesame and Psychopy do not quite get along. When installing both programs, Psychopy crashes right after starting unless you open an already existing experiment. This seems to be due to the access to the Python database. Maybe you have installed a new program which is causing problems now. Regarding the new version of Psychopy… try to start your experiment on another computer with the new version. So you could narrow down the error.
Hi Okrim, that’s a good point! I’m not aware of any new software, but I’ll double check. And that’s a good suggestion, I’ll update psychopy on a different laptop and troubleshoot there before I update the presentation laptop. Thanks for the help!
I would strongly recommend updating PsychoPy and keeping it up to date. You can select the version that you use from the use version drop down in experiment settings. If you wanted to test this out before totally removing your old version you could install the latest version of PsychoPy on another machine and test it out with your experiments as @Okrim suggested.