Calibration screen unresponsive using PyGaze portable environment with EyeLink1000

EthanDC

Hi,

I am currently trying to run an experiment in the PyGaze portable environment (WinPython-PyGaze-0.6.0). The experiment works fine in dummy mode. However, when connecting to the EyeLink tracker, calling eyetracker.calibrate() brings up the calibration screen but it is completely unresponsive and pressing any key causes the PsychoPy display window to become 'not responding'. This is also the case when running any of the PyGaze examples, such as 'shooting_game.py'.

The EyeLink calibration works fine when running experiments in PsychoPy or with E-Prime.

Any advice would be much appreciated.

Thanks,
Ethan

Edwin

Hi Ethan,

A bit more info would be greatly appreciated! For example, do you have any error messages at all? Does the freezing also occur with the PyGame back-end? Is the system responsive if you direct the calibration from the EyeLink PC?

Cheers,
Edwin

EthanDC

Hi Edwin,

Thanks for your response. There were no error messages, but we have managed to get it to work when directing the calibration from the EyeLink PC as suggested. Also, drift corrections can only proceed by using the EyeLink PC too.

We would be able to run the experiments this way, but it would still be good to know why the keyboard becomes unresponsive in the calibration and drift correction screens.

Thanks,
Ethan

Edwin

Great to hear the experimenter PC at least still works! I honestly don’t know what could be going wrong, and won’t be able to debug the issue without extensive information on your current setup and the runs. For example, my earlier question about the back-end is quite crucial, and further info on the versions of PyGaze, pylink, PsychoPy, PyGame, etc. that you’re using; the OS that your stimulus PC uses, and the version of the EyeLink OS.

Once we have that, we can try to add some debug messages to the code, and see where stuff is hampering.

Cheers,
Edwin

EthanDC

Hi Edwin,

Switching to pygame resolved the issue when using the shooting game example and the calibration screen works, so thank you! I will have to work through changing our experiment code to work with the pygame display instead of psychopy, but I expect it should work.

To answer your other questions:

Versions of:

• PyGaze: WinPython-PyGaze-0.6.0
• Pylink: v1.0.0.27
• PsychoPy: 1.82.01
• PyGame: v1.9.2
• EyeLink OS: Eyelink 1000 v4.56 on Datalight ROM-DOS version 7.1SU

OS of stimulus PC: Windows 7 (64-bit)

Many thanks,
Ethan

tknapen

Hi Ethan/Edwin,

just wanted to second this issue. We're having this problem on a custom install of psychopy/pygaze on windows, but also on the single-file environment download from the pygaze website. This problem occurs only when using psychopy windows in pygaze, as pygame windows (and everything) works. It definitely seems to be something to do with the interaction between psychopy windows and pygaze...

Best, Tomas

Some versions of the custom install:
pygaze 0.6.0a10
pylink 1.1.0.5
psychopy 1.82.01

PS: just to add to this; the example simple_tracker_experiment.py also suffers from this problem, even in the WinPython-PyGaze environment. It seems to go into an infinite loop in the tracker.calibrate() statement on line #52.

Edwin

Hi Tomas,

Hope you're well, and thanks for the feedback!

It sounds like the conflict is between pylink and PsychoPy. I would recommend trying the no-SDL version of pylink, which you can download from the SR Research support forum. The regular pylink library has some SDL functionality (unnecessary if you ask me, but oh well), and this has conflicted with other libraries in the past.

Please do let me know how you get along!

Cheers,
Edwin