Our lab has been using Superlab for about 6 months. Once we got it up and running, the timing has been pretty perfect every time we’ve used it–our scripts last the expected time and end at the same time as the scanner.
During a scan yesterday, all of the scripts ended after the scanner (~4 seconds delayed). From looking at the log file, I can tell that the scripts lasted longer than they were supposed to last, so it wasn’t just that they were triggered late. There is no discrete event that took longer than usual; it seems a bit of time is added to each trial, so that halfway through the paradigm, we were at a 2 second delay. I ran them today outside the scanner and timed them and again they lasted longer than they should have.
What could be causing this? It seems to me that the computer has just started running slower, but I don’t understand why there would suddenly be such a dramatic shift. It’s a fairly new computer, and the day before, the scripts ran fine. There were no big downloads or updates or anything between the two days.
Any suggestions or ideas as to what may have happened?
Computers could indeed slow down for no apparent reason, e.g. they could be indexing your file system, responding to network, hit by a virus, or any number of reasons.
Counting on when a trial ends, e.g. by changing time limit values, is not a recommended approach, e.g. what if you had to use another computer entirely? Please see our tech support note Synchronizing Trial Presentation With fMRI Scanner. It offers two different methods for synchronizing with the scanner.
We initially tried triggering every trial with scanner but were unsuccessful. The other method would require us to change our paradigm (make it longer–we would have to specify a longer interval than the length of time our trials currently take, to make sure we fit everything in before the interval was reached). We can’t do that at this point as we are mid experiment.
I think that this sudden slowdown of 4 seconds seems pretty unusual. This computer is just a couple of months old. The script is now 4 seconds slower every time I run it, so it doesn’t seem like a temporary process was causing the problem (indexing the file system or responding to the network). I tried disconnecting from the wireless signal and it made no difference. The computer is not showing any other signs of problems.
I borrowed another lab’s computer that has SuperLab and ran our script on it and it ran just fine, just like it ran before on our other computer. So the effect of changing computers was far less extreme than the difference we saw overnight in the one computer.
Do you have any other thoughts on what could be causing the problem? Also, if we want to try switching to another computer permanently, is there a way to transfer our SuperLab license?
I have no idea what might be causing the slowdown. One suggestion I can offer is to look into what the Task Manager tells you about CPU usage and available RAM, especially when compared with the other lab computer.
Regarding moving the license, we can help you with that, please call or email.