current best-practice for playing long videos?

[Brian Murphy]

Hi,

we're preparing some EEG and MEG based movie viewing paradigms, and want to 
make sure we are following best practise in how we prepare and present the 
videos for accurate timing. These videos can be very long (5mins to 10's of 
minutes), and *don't* have particular timing requirements relative to other 
stimulus elements. The only thing we want to be confident of is precise 
synchronisation with the EEG/MEG signals. That is, we want to know 
precisely when the video presentation starts (accurate reporting of actual 
onset), timely communication of this onset to our M/EEG recording machines 
via a serial port trigger, and that once the video is running, it streams 
smoothly and runs at the correct rate (and so duration).

I've been digging around in the group archives, and the PST knowledge base, 
and have found plenty of advice, so first of all thanks to David Vinson, 
Dave McFarlane, "anonymous" Dave, and a load of other people not called 
Dave!

Here is a summary of the recommendations I found, and we would really 
appreciate any comments/corrections. E.g. maybe some of them are out-dated?
 - upgrade eprime 2 to upgrade to version 2.0.8.90 or 2.0.10.242
 - check if the sample MovieRT experiment works (probably unnecessary if 
you already have our video working)
 - consider re-encoding ("transcoding") to match the format/coding used by 
PST-provided examples, which is the MPEG Video (Version 1) codec, packaged 
as an .mpg file (though elsewhere .avi packaging is recommended also)
 - check the codec configuration of the system using the eprime tool: http://pstnet.com/support/kb.asp?TopicID=3162 

- display your video at its natural size (no stretching)
 - log these movie-specific timing properties: FirstFrameTime, FrameRate, 
FramesDisplayed, and FramesDropped
- experiment with choice of renderer, which can reduce prerelease 
(=loading/buffering?) time by 50%: 
http://pstnet.com/support/kb.asp?TopicID=4016
http://pstnet.com/support/kb.asp?TopicID=3836
- pre-load the video (e.g. MovieDisplay1.Load in an inline object) well 
before the movie is due to start playing
- run the simplest possible paradigm (you prob. are already), and get 
eprime to repeat your video many many times, to stress test it

Things that are NOT recommended are: (as of versions 2.0.8.90 and 
2.0.10.*): 
- Matching movie frame rate and refresh rate, which results in skipped 
frames
- using the DivX (TM) codec 

One thing I'm still puzzled about is streaming/buffering settings. When 
doing a similar experiments with long audio book stimuli, you can set the 
size of the buffer. Are there similar movie settings? 

best,

Brian

-- 
You received this message because you are subscribed to the Google Groups "E-Prime" group.
To post to this group, send email to e-prime at googlegroups.com.
To unsubscribe from this group, send email to e-prime+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msg/e-prime/-/RyME4_tKof4J.
For more options, visit https://groups.google.com/groups/opt_out.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/eprime/attachments/20130125/2f918ce4/attachment.htm>