Hi Aashish,
yes, vrpn and vrui are configured and set apparently in the right way
(for another application i'm using vrpn to track for paraview, and also
for a meshviewer based on vrui, through a vrui devicedaemon, so
apparently they are working both in the proper way).
After I compiled the new version of paraview with the vrplugin fix, when
i load the vrplugin, in the GUI, the menu to select the type of
connection (the upper one after you choose to add a connection VRPN or
VRUI) is empty, and therefore i can't select any of them. And if I try
to load a state file with the proper connection (i'm using a vrpn server
at the moment), it doesn't load.
About the inversion of coordinates, i solved, it's the tracker, and i
fixed that externally. Today i achieved to use paraview on a
demonstration on two screens with passive stereo and the result was
definitely perfect in the way is was moving (but i'm still loading the
VRplugin from an older paraview, not from this last one).
Best,
Emiliano
Il 30/01/13 20:50, Aashish Chaudhary ha scritto:
Hi Emiliano,
On Tue, Jan 29, 2013 at 1:07 PM, Emiliano Pastorelli
<[email protected]> wrote:
Hi Aashish, today i installed the master ParaView as you suggested.
There is something wrong with the GUI for the VRPlugin, and now it's not
possible to select anymore the type of the VR connection, not even importing
a state file.
Did you compile with VRPN / VRUI ON? I am not sure if understood the
problem here.
I solved that importing an older version of the plugin, with
that it works, but so i don't know it the problems are fixed. (it took me a
while to figure how to use the new FullScreen option but then i managed, i
was trying to put it in the Server process)
We can put that in the sample VR file.
Now somehow once in a while, even after applying the right transform matrix
to the tracker data, i have the weird effect that once in a while, the axis
seem to randomly invert. So if until a second before it was working fine
with the headtracking around an object, suddenly it gets wrong on all the
four axis. I don't know if it might be a problem of my tracker or is
something in paraview.
I would believe that could be the problem on your tracker (depending
on how stable your tracker is).
As a test, I saved a working system state, went out of paraview and when i
restarted and restored the state, the axis were wrong again.
As a side effect, there is a XGrabKeyboard command to remove in the vtk file
vtkXOpenGLRenderWindow.cxx and if you don't remove it, when paraview is
running fullscreen you can't use the keyboard anymore, nor in paraview nor
in other applications.
This is something that came before. We will push a fix.
Thanks
Best,
Emiliano
Il 28/01/13 23:43, Aashish Chaudhary ha scritto:
On Mon, Jan 28, 2013 at 4:19 PM, Emiliano Pastorelli <[email protected]>
wrote:
Hi Aashish, perfect, thank you very much! I'll give it a try tomorrow, and
if it all works well i'll use it for a demonstration on wednesday, just in
time!
Great. Let us know if we can help.
Thanks,
Regards,
Emiliano
Il 28/01/13 22:15, Aashish Chaudhary ha scritto:
Hi Emiliano,
I just wanted to let you know that recently we fixed couple of minor bugs
and now they are in ParaView master if you want to give it a try.
Thanks
On Fri, Jan 11, 2013 at 12:47 PM, Aashish Chaudhary
<[email protected]> wrote:
On Thu, Jan 10, 2013 at 9:00 AM, Emiliano Pastorelli
<[email protected]> wrote:
Hi,
I'm having some issues with headtracking in ParaView in a 3 screens
environment (2 walls + 1 table). From the GUI (and also from the config pvsm
file, but equal results) I tried to configure the following propertier
connected to the tracker.
a) Camera Position ( I mentioned it in a previous message), the vrpn
server doesn't seem to feed the proper argument to the SetCamera function
and so it doesn't work.
b) View Position : when i start the vrplugin it gives me the following
message :
Generic Warning: In
/home/hogan/projects/ParaView/ParaViewCore/ServerManager/Core/vtkSMPropertyHelper.cxx,
line 429
Call not supported for the current property type.
And nothing happens.
c) EyeTransformMatrix : close to be working, but when i move my head
around the screens, the object doesn't move properly around the screens
(while if i just move the object with the mouse, it goes properly across the
three display, moving from one to another according to the position)
d) ModelTransformMatrix : works properly
Which is the "official" procedure to do the headtracking? Am I doing
something wrong? The fact that it works properly with the
ModelTransformMatrix makes me think that the vrpn server that i wrote for
this tracker works properly, so i don't get what else i shall modify to
properly track the head movement.
Connecting your sensor to the EyeTransformMatix is the right thing to do
for the head tracking. Is it possible that your numbers are too small? When
you move your head the position of the object may not move that much
depending on your object is front of the projected plane or behind. If you
can send me your config file, state file and what values you are getting
from your sensor, I can provide you exact explanation.
Thanks,
Any hint is welcome before i go mad!
Best,
Emiliano
_______________________________________________
Powered by www.kitware.com
Visit other Kitware open-source projects at
http://www.kitware.com/opensource/opensource.html
Please keep messages on-topic and check the ParaView Wiki at:
http://paraview.org/Wiki/ParaView
Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview
--
| Aashish Chaudhary
| R&D Engineer
| Kitware Inc.
| www.kitware.com
--
| Aashish Chaudhary
| R&D Engineer
| Kitware Inc.
| www.kitware.com
--
| Aashish Chaudhary
| R&D Engineer
| Kitware Inc.
| www.kitware.com
_______________________________________________
Powered by www.kitware.com
Visit other Kitware open-source projects at
http://www.kitware.com/opensource/opensource.html
Please keep messages on-topic and check the ParaView Wiki at:
http://paraview.org/Wiki/ParaView
Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview
