psychopy.org | Reference | Downloads | Github

Long experiment with videos freezes

I’ve programmed an experiment which shows 90 videos (.mp4), each for about 20 seconds. The experiment will freeze after a variable amount of videos, closer to half of the videos most of the times that it happens. It also sends data to the csv each frame of said videos, a good amount of data but still in the tens of MBs total.

Is there a data storage, or video playing limit in Pavlovia that is freezing my experiment? I have checked running the same experiment but with fewer videos, and it seems to work with no problem. There’s no other code change, so I suspect the reason is the amount of videos.

Can anyone confirm this? If so, is there a solution? If not then I guess dividing my experiment in 2 would work, but I want to check if there is another option, and if it makes sense that my experiment is freezing because of many videos.

Thanks a lot.

Hi there,

This is something I have also experienced and is something that we are looking into (see similar thread here: How to trouble shoot when online experiment gets stuck only for few participants - #2 by Becca for a couple of points that might help with debugging).

As far as we are aware there isn’t a data limit from our side of things, so I suspect this is a browser cache/memory issue. Some things I would be curious to know:

  1. Does it happen everytime?
  2. Does it happen if you clear your browser cache before running the study?
  3. Does it happen on several devices? (I found the issue in my experiment on phones but not laptops).
  4. Does it happen across several browsers.

You’ve correctly identified that splitting in 2 might be the solution for now - but it would be good to avoid this if we can!

Becca

Hi again.

  1. It happened only some times
    3 and 4) It did.
    Regarding (2) we haven’t tested it yet, since we decide that it was a good idea to separate the experiment in 2 parts regardless. Now I have another problem involving this same experiment, the data is only saved some times. Each csv is about 30 MB, but some data has been lost! If not related enough to this problem I’ll make another thread. Thanks a lot!

Hi There,

OK pleased that splitting into 2 exps provided a solution to your initial issue.

Yes if you would please be able to make a separate thread for the new issue that would be good,

Thanks,
Becca