Video experiment crashes

Judith

Hi all,

I have a problem in an experiment with multiple 10-minute video files. When running the experiment, it sometimes crashes during video presentation. The screen 'freezes' and I have to end the experiment manually using task manager. This happens on multiple computers. However, it does not happen everytime that I run the experiment.

I am using media_player_vlc (because media_player_mpy seems to play the video and audio not completely synchronous). And I use backend legacy/PyGame (because in Expyriment only the audio is played in fullscreen mode, no video.)

I have read here: http://forum.cogsci.nl/index.php?p=/discussion/2919/python-crashes that this might be caused because the system runs out of memory. Is there a way to fix this?

Thank you!

sebastiaan

I have read here: http://forum.cogsci.nl/index.php?p=/discussion/2919/python-crashes that this might be caused because the system runs out of memory. Is there a way to fix this?

If you open the task manager, you should be able to see how much memory OpenSesame is using. If this indeed accumulates over time until it exhaust almost all memory, then that might indeed be the issue. If so, there's not much you can do about it, because the media_player_vlc plugin is no longer maintained.

because media_player_mpy seems to play the video and audio not completely synchronous

Is this the case for all backends? For me the media_player_mpy works pretty smoothly, and is definitely preferable over the VLC-based one.

Cheers!
Sebastiaan

Judith

Hi Sebastiaan,

Thank you for the quick reply!

The media_player_mpy gives problems both in the legacy and expyriment backend for me. The difference between the video and audio is minimal, but noticeable because the video involves people speaking. The video seems to go a bit faster, which causes the impression that it is not the person in the video that is speaking. So I guess I will have to use the VLC player. Or is there a solution possible?

I checked the memory in task manager, but it does not seem to be a problem. The experiment uses ~10% CPU and ~ 130MB of memory. When I look in the tab Performance, I see that there is ~ 5 GB used, 10 GB free, and this seems to be constant over the experiment...

Should I try to reduce the size of all videos? Now they are ~ 320 MB.

Daniel

Hi Judith, It could well be that the audio and video are out of sync. It's really tough to get this to work correctly and the media_player_mpy misses some rudimentary mechanisms that guarantee the synchronisation.

I would try to convert the videos to a different format. Some formats just work better than others for playing the video and audio in sync. You can use VLC to reencode the videos, as described here. Lowering the size or resolutions of the videos might also do the trick.