[solved] canvas class and psychopy back end

embodiment

Hi Sebastian.
Some time ago you helped me to implement a python code in OpenSesame in order to creating a clock face with a rotating hand (http://forum.cogsci.nl/index.php?p=/discussion/38/solved-intentional-binding-paradigm-creating-animations#Item_21).

Since now, I have used the default legacy back end but I would try to switch to psychopy, in order to enhancing (I hope) the timing precision. I noticed that if I simply select the psychopy option from the pull down menu, the rotation of the clock hand become jerky, not so fluid as it was with the legacy. I'm wondering if it is due to a problem between the openexp modules (canvas, time...) and the psychopy. In your opinion, should I use psychopy modules, such as visual or core, in this case?

sebastiaan

Hi,

Yes, I remember. I hope that worked out for you!

The openexp classes are convenient, because they work the same for all back-ends. But they aren't always optimal in terms of efficiency, particularly when you want to make animations.

So yes, I would recommend directly using the psychopy classes. Basically, you can just use psychopy in the normal way, the only thing that you need to know is that the window is accessible in an inline_script item as self.experiment.window.

Also see

• http://osdoc.cogsci.nl/back-ends/psycho
• http://www.psychopy.org/api/visual.html

Good luck!

Sebastiaan

embodiment

Yes, your code works very well. There is only a small timing problem that maybe I can solve by using psycho back end.

Thanks for the reply.

Andrea