Hello Mario and everyone,
I FINALLY got around to updating to the new version of PTB (3.0.8) and
checking out the multiple monitor support on Windows.
I'm having slightly patchy results.
My set-up has one CRT monitor off one graphics card, and two projectors
off a dual-head NVidia graphics card.
In the old PTB, I had the two projectors in NVidia's horizontal span
mode, i.e. Windows believed them to be a single display device. I set
this to be the primary display, so PTB could use it, and used stereomode
4 to put the left image on one projector and the right image on the other.
In the new PTB, this doesn't work. It finds both screens (display1 =
both projectors, display2 = the CRT), and says that these correspond to
Screen 1 and Screen 2 respectively. So you'd think that opening a window
with
Screen('OpenWindow',1)
would put up a window across both projectors (2560x1024). But no, the
window only appears on one of the projectors (1280x1024, ie half the
extent of the Windows display1).
Screen('OpenWindow',2) hung matlab, as did Screen('OpenWindow',0).
I had more joy using the projectors in NVidia's dualview mode. So now
Windows knows it has three screens: the CRT (display2) and two
projectors (display1 and display3). Again, PTB finds all 3 screens.
Screen('OpenWindow',0) does not crash matlab this time, but it only
displays on one of the projectors (not the full Windows desktop as stated).
Screen('OpenWindow',1) and Screen('OpenWindow',3) work, displaying
windows on each of the projectors in turn.
Screen('OpenWindow',2), which should be the CRT, again crashes matlab.
So, it seems that with the new PTB, I have to use dualview and open
separate windows for each projector. I can't decide if I want to do this
or not -- in principle it has some advantages, e.g. being able to apply
separate gamma correction for each projector (very nice). However, it
also means I have to flip the two projectors separately, and it seems to
me that could potentially lead to interocular delays. Also, it means
rewriting all my existing code, which is something of a pain.
Finally, I am only just beginning to play around with this new version,
but I am finding I sometimes get errors where PTB says it cannot
symchronise to the vertical retrace. As far as I can see, these errors
are not deterministic, i.e. code which worked fine yesterday is now
giving me these synch errors, even though as far as I'm aware, I've not
changed anything. So it looks as if it might be a bit buggy to me.
Anyway, that's my experience so far. Thanks very much for all your work
on PTB!
Best wishes,
Jenny
--
Jenny Read Office: +44 191 222 7559
Royal Society University Research Fellow Mobile: +44 794 401 5796
Henry Wellcome Building for Neuroecology, Fax: +44 191 222 5622
University of Newcastle, Framlington Place mailto:J.C.A.Read@...
Newcastle upon Tyne, NE2 4HH, UK. http://www.staff.ncl.ac.uk/j.c.a.read
I FINALLY got around to updating to the new version of PTB (3.0.8) and
checking out the multiple monitor support on Windows.
I'm having slightly patchy results.
My set-up has one CRT monitor off one graphics card, and two projectors
off a dual-head NVidia graphics card.
In the old PTB, I had the two projectors in NVidia's horizontal span
mode, i.e. Windows believed them to be a single display device. I set
this to be the primary display, so PTB could use it, and used stereomode
4 to put the left image on one projector and the right image on the other.
In the new PTB, this doesn't work. It finds both screens (display1 =
both projectors, display2 = the CRT), and says that these correspond to
Screen 1 and Screen 2 respectively. So you'd think that opening a window
with
Screen('OpenWindow',1)
would put up a window across both projectors (2560x1024). But no, the
window only appears on one of the projectors (1280x1024, ie half the
extent of the Windows display1).
Screen('OpenWindow',2) hung matlab, as did Screen('OpenWindow',0).
I had more joy using the projectors in NVidia's dualview mode. So now
Windows knows it has three screens: the CRT (display2) and two
projectors (display1 and display3). Again, PTB finds all 3 screens.
Screen('OpenWindow',0) does not crash matlab this time, but it only
displays on one of the projectors (not the full Windows desktop as stated).
Screen('OpenWindow',1) and Screen('OpenWindow',3) work, displaying
windows on each of the projectors in turn.
Screen('OpenWindow',2), which should be the CRT, again crashes matlab.
So, it seems that with the new PTB, I have to use dualview and open
separate windows for each projector. I can't decide if I want to do this
or not -- in principle it has some advantages, e.g. being able to apply
separate gamma correction for each projector (very nice). However, it
also means I have to flip the two projectors separately, and it seems to
me that could potentially lead to interocular delays. Also, it means
rewriting all my existing code, which is something of a pain.
Finally, I am only just beginning to play around with this new version,
but I am finding I sometimes get errors where PTB says it cannot
symchronise to the vertical retrace. As far as I can see, these errors
are not deterministic, i.e. code which worked fine yesterday is now
giving me these synch errors, even though as far as I'm aware, I've not
changed anything. So it looks as if it might be a bit buggy to me.
Anyway, that's my experience so far. Thanks very much for all your work
on PTB!
Best wishes,
Jenny
--
Jenny Read Office: +44 191 222 7559
Royal Society University Research Fellow Mobile: +44 794 401 5796
Henry Wellcome Building for Neuroecology, Fax: +44 191 222 5622
University of Newcastle, Framlington Place mailto:J.C.A.Read@...
Newcastle upon Tyne, NE2 4HH, UK. http://www.staff.ncl.ac.uk/j.c.a.read