Hi Hisham
A colleague created an experiment with 3 blocks, 7 trials and 640 events. She linked the first 40 events to the first trial in the first block, then linked the next 550 events to second trial in the second block, with the remainder linked to various trials all in the third block.
When run, the experiment gets through the First Block OK, but hangs when trying to load the next block and trial.
Do you have any idea what might be happening?
It’s hard to tell without more details. Are those 550 events all picture events? SuperLab might be trying to load them from disk to RAM (it would look like it has frozen during that time).
Hi Hisham
Yes, at least 95% are images and yes it seems to want to load them into RAM. This slows the experience down for the participant, as they see one trial then might have to wait ten minutes or more before the next trial is presented. i have tried this on an Intel Macbook and a new PC. The Macbook appears to freeze for at least 20 mins and the PC shows a Windows Explorer window over the top of the stimuli then can be started after about ten minutes. I did try to attach the experiment (in ZIP format), but obviously it didn’t come through.
550 trials is a lot but I’ve seen SuperLab handle experiments with at least twice that many. I’ve also seen SuperLab take a very long time on experiments with fewer trials for one or both reasons: not enough RAM, or the images are needlessly too big. By that, I mean that the experimenter sometimes want to display an image that is 100 x 120 pixels but the image file is saved as 1024 x 1200, or something like that.
A small tip: BMP files are quite a bit bigger than JPG but they usually load faster because no decompression is necessary.
the files are GIF’s and only about 56Kb each. i tried converting to BMP’s, but it made it worse. The laptop has 1Gb RAM.
However, i then created a Stimulus List of the files and it now seems to work as expected.
Ah. If I had to take a guess, I’d say that in the earlier design, you probably had the 550 pictures as part of one trial. With the new design using stimulus lists, you have one or two per trial.
Hisham.