I've successfully installed the latest version of psychtoolbox3 through the neurodebian repositories (package name: octave-psychtoolbox-3
).
Calling AssertOpenGL seems to work fine:
>> AssertOpenGL
PTB-INFO: Display ':0' : X-Screen 0 : Output 0 [default]: Primary output : Connected : CRTC 0 [XID 642]
PTB-INFO: Display ':0' : X-Screen 0 : Assigning primary output as 0 with RandR-CRTC 0 and GPU-CRTC 0.
But when I issue a call to Screen, I get a rather cryptic error.
>> Screen('Resolution', iscreen, 1024, 768, refreshrate);
INTERNAL PSYCHTOOLBOX ERROR
error: PsychError_internal
general description: Unspecified error, probably a Psychtoolbox bug
specific description: Attempt to set invalid video settings
module name: Screen
subfunction call: Resolution
file name: Linux/Screen/PsychScreenGlue.c
function name: PsychSetScreenSettings
line number: 1776
INTERNAL PSYCHTOOLBOX ERROR
error: PsychError_internal
general description: Unspecified error, probably a Psychtoolbox bug
specific description: Attempt to restore now invalid video settings
module name: Screen
subfunction call: Resolution
file name: Linux/Screen/PsychScreenGlue.c
function name: PsychRestoreScreenSettings
line number: 1846
PTB-ERROR: Error during error handling! ScreenCloseAllWindows() called recursively! Trying to break out of this vicious cycle...
PTB-ERROR: Maybe it is a good idea to exit and restart Matlab/Octave.
Error using Screen
See error message printed above.
The variable iscreen
in the call to Screen
is set by the line iscreen = max(Screen('Screens'));
and is equal to 0
.
Any idea what could be going wrong, here? I know for a fact that this script works under MacOSX, but for some reason it crashes on this machine (Ubuntu 12.04 / Matlab R2012a).
Thanks in advance!
EDIT: Some have suggested that this could be an issue with my display not supporting the refreshrate I'm passing to the Screen
call. Unfortunately, when I substitute the refreshrate
variable with my display's current refresh rate, I still get the same error. Any other ideas?