Announcement

Collapse
No announcement yet.

Experiment runs very slow

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Experiment runs very slow

    Hello. I've created an experiment (SL 4.07b) modeled after the "Choosing Pictures From Different Lists, With No Repetition" Demo experiment. The experiment will show pictures in an RSVP format and then ask participants what they saw after each trial (each stream of pictures has 2 targets we are interested in).

    Each picture is set to be shown for 120 ms, but because the program is so slow, the pictures are being shown for varying times up to 1500 ms. If we can't control the duration each image is shown, we can't run this experiment.

    I don't think it's the pictures because in the recent past, I created an experiment that did the same thing but did not use the macros in the Demo experiment. Using these macros is much better in that we can have unique trials for our questions, allowing data analysis to be MUCH easier due to the summary stats feature in your DATA Viewer program. The "old" experiment required days working with long Excel formulas to achieve what your data viewer does easily, so long as we have unique trials for each question.

    Under OPTIONS the memory management is currently on "load one trial at a time", but I've tried it both ways with the same slow result.

    I've attached a compressed folder with the main part of the experiment and associated files.

    Any help in getting this to run at the correct speed is greatly appreciated.
    Attached Files

  • #2
    Experiment re-arranged (fewer macros)

    Now that I've re-arranged the experiment in a way that utilizes slightly fewer macros and more stimulus lists without the "choose one" macro, it seems to work at the right speed.

    Comment


    • #3
      Experiment is still running slow now that I've included all the blocks and trials.

      So, I re-did the experiment to try to use fewer macros but I still need quite a few (about 15). Now that I've added all the blocks and trials it is again running slow. I disconnected the machine from the network to see if that was causing the slowing, but it still runs very slow. Each picture should be shown for 120 ms but instead there is a lot of variation as some pictures stay on the screen for over one second. And it is not that the same picture is always slow. Rather, it seems to vary so that approximately every 2-4 pictures, one remains for about one second or more rather than 120 ms.

      I've attached the experiment.

      Any help is appreciated.
      Attached Files

      Comment


      • #4
        I think it's the number of blocks that is slowing me down

        I tested it to see if merely duplicating blocks in a simple experiment (using stimulus lists) led to the problems I'm having with extreme slowing. Adding about 450 blocks immediately slowed things down to a crawl. In fact the first picture seemed to stay on indefinitely instead of 120 ms. Is there something I'm doing wrong?

        Comment


        • #5
          SuperLab does not automatically erase anything from the screen unless explicitly done. Try going into the Event Editor, then Settings, here select the Presentation Options tab. Under At Event End, select the option Erase Screen. Let me know if this helps with the timing issue.

          Comment

          Working...
          X