DirectRT sometimes leaves a 1k file, when executed from MediaLab

[jarvis24]

I recall that issue coming up but it quite a while ago. I was surprised to see that you v2008--is it an early release of it? I *think* the problem used to be that if the subjects hit the ESC key accidentally then no data would be written and you would only get data files with headers because they are created first and the data were written AFTER the session. So we changed the code so that if the session ended prematurely, then the data gathered so far would still be written at the point where the ESC key was hit. Are any of your response keys near the ESC key? Also, what release of v2008 do you have? i.e., 2008.x.x?

[baranan]

2008.1.33 is the version.

However, after some investigation, my student found that the problem this time was not probabilistic. It always a "Row # has a missing value for one of its LOC columns". So it is a DirectRT problem, and probably should be moved to the DirectRT thread about this problem.

Anyway, I deleted most rows from the file, but I still get the same error message. We also have 7 other versions of this file (different IAT condition), but according to my student, this is the only one with that problem. She already solved it by creating that file again, but I wonder what the problem with this file is. It is attached.