Millisecond Forums

Dear all. 
I have a study that uses a pool of 600 photographs from which only 12 are seen by each subject and these are selected randomly (with certain restrictions).
My problem is that uploading so many photographs takes too long and for some subjects it does not even load. Is there a way to keep the photographs on a server so that each subject actually loads only the 12 photos selected for him or her? Obviously I can dissect the set and run it using sub groups but this will hurt randomization.
Thanks much
Shlomo     

Given how your task is set up (as far as I recall from previous posts): No. The selection of the 12 out of 600 stimuli only happens once the task starts. For performance reasons (stimulus and response timing), Inquisit does not fetch files from online sources at runtime; all files are downloaded to the client machine before the experiment starts to execute on the client, and that means 600 pictures in your case. As far as I can see, the only thing you can do is try to reduce the file sizes of your 600 images as much as possible.

For everything else, you'd have to (a) re-consider your design (e.g. compile certain fixed sets of 12 images in advance) and (b) switch to Inquisit 4 (different sets could be loaded via conditional <include> elements, which are not available in Inquisit 3).

Dear Dave Thanks
I reduced the size of the files but this also did not change much. However, I do use Web Inquisit 4. I am using Inquisit 3 to program the study and examine it but on the web I use 4. Need I upgrade my lab version to 4 or I can download 4 for the sake of programming the study ?
Thanks much
Shlomo

You to make use of new features and test them on your machine prior to deploying the script online, yes you should download and install Inquisit 4 Lab on your machine. Whether you have a license for it isn't that important; the trial version will do.

But again, note that the <include> approach would require other changes, i.e., organizing your 600 items into a number of distinct, fixed 12-item sets. Each set would essentially be a between-subjects condition and only the 12 items pertaining to the applicable set would be downloaded.

See e.g. https://www.millisecond.com/forums/Topic15639.aspx for something similar.