[open] edf file transfer taking too long

ChiaraDL

Dear all,
I am running an eye tracking experiment with Opensesame 3 using the portable version, and Eyelink.
Everything works fine, except for the fact that at the end of the experiment, the edf file transfer process takes a VERY long time. I have just above 1000 trials in my experiments, and after 40 minutes from the end it was still running... I had to shut down (ctrl-alt-del) Opensesame to get out of it.
Does anyone else have this problem?
Do you have suggestions on how I could avoid it?

Thank you in advance,
Chiara

sebastiaan

Hi Chiara,

Data transfer can take some time. But this is in order of minutes. 40 minutes is clearly excessive, so something probably went wrong.

A good thing to know is that your data is not lost: Even if it's not transferred to the experimental PC, a copy of the edf is still on the EyeLink PC.

To track down the problem, could you give more details?

• Does the EyeLink PC still indicate that a transfer is in progress?
• What happens in OpenSesame? I.e. has the experiment window closed? And can you see progress in the debug window?
• Does this always happen, or just occasionally?

Cheers,
Sebastiaan

ChiaraDL

Hello Sebastiaan,
it did happen every time, I will now try again and will look at what the eyelink computer says in the meanwhile. Another curious fact is that an edf file is already in my folder on the opensesame pc, and its size does not change over time (I'd expect that it should increase as the transfer progresses...). This happens after the experiment window has been closed. Opensesame during this time is showing the number sequence in the debug window and is otherwise "shadowed", non responsive. The numbers are printed in the debug window very slowly with respect to what I have seen with other versions of opensesame. They are also printed "horizontally", not in a column...
Until soon,
Chiara

ChiaraDL

Ok, during this process Eyelink is closed/offline.

ChiaraDL

Also, it appears that the edf file that I see in my folder from the very end of the experiment is already complete. Could the printing process (in the opensesame debug window) be (much) slower than the actual transfer?

sebastiaan

Hi Chiara,

Could the printing process (in the opensesame debug window) be (much) slower than the actual transfer?

Yes, that's possible.

I haven't really managed to replicate the issue. Transfer takes a bit longer than you might think is necessary, but nothing as excessive as what you're seeing. (Perhaps you need fairly big data files to see it; I tried with short sessions.)

In any case, I packaged a prerelease of 3.0.3 in which output during EDF transfer is suppressed. Could you try the latest prerelease (3.0.3a3 now) and see if this resolves the issue?

• http://files.cogsci.nl/software/opensesame/pre-releases/

Cheers,
Sebastiaan

ChiaraDL

Thank you, I will try this as soon as I have a chance to!

Chiara