I am trying to use a VIEWPIXX 3D lite connected to a DataPixx3 to present 3-D stimuli using MatLab and the Psychophysics toolbox. I am using Volfoni glasses and a Volfoni emitter. The operating system is Windows 10 and the graphics card is an AMD card. I am using the last free verson of PTB.
Both my own code and the classic stereo demo including the version written by vpixx shows the same issue on our set up.
The polarity of the stereo randomly and unpredictably flips e.g. front becomes back and back becomes front. The stereo can be stable for a while and then it will flip.
I’m not sure what is causing this issue and wondered if it is something that anybody else had experienced with a similar set up?
Thanks
Kevin
Hi Kevin, this is Lindsey, I’m a member of VPixx Technical support.
I think my colleague Jonathan has already answered you by email, but in case anyone sees this and is interested, this sounds like frame dropping.
Frame dropping occurs when the graphics card fails to hand new image data to the display before the next refresh. In this case, the previous video frame is repeated. For alternating-frame 3D, this can cause the glasses and stimuli to become temporarily desynchronized.
Repeated frame dropping is a sign of issues with GPU performance. You can try running VBLSyncTest in Matlab to see if there might be a high number of dropped frames on your display. There are several things you can do to try to reduce dropped frames, such as:
- Use fewer monitors
- Turn off unnecessary background programs
- Try a different machine/graphics card
We also have a top-bottom 3D mode that is robust against frame dropping, as both left- and right-eye image data are passed to the display simultaneously.
We keep an eye on the PTB forums, but if anyone has questions about VPixx hardware or software, don’t hesitate to send us an email (support@vpixx.com). We typically respond very quickly. 
Psychtoolbox ImagingStereoDemo(2,1) would demo that top-bottom mode on VPixx devices for optimal robustness iirc.