I am not sure why it doesn't work but what you are doing is not safe
and is bad practice. The inputVector belongs to the input of this
filter and you should not modify it (other than passing a request
during RequestUpdateExtent). Isn't there another way you can do this?
Maybe having a data member d
Hi Steve,
Which version of ParaView is this?
-berk
On Tue, Oct 13, 2009 at 4:50 PM, Steve Giessler wrote:
> I followed all the instructions for compiling, installing ParaView. It
> seemed to compile without error (64-bit Linux), but when I typed "make
> install," I got this error near the end:
You may also have to call servermanager.Disconnect() before calling
servermanager.ReverseConnect(). This is because paraview.simple
creates a builtin connection by default.
Pat
On Tue, Oct 13, 2009 at 1:23 PM, Berk Geveci wrote:
> It looks like we forgot to add support for reverse connection to
Yes, for me the artifacts are gone when using pure surfaces only. Still,
I need wireframes and surface+edges representations...
Berk Geveci wrote:
Hmmm. Can you try turning off all wireframes?
-berk
On Tue, Oct 13, 2009 at 3:47 PM, Dominik Szczerba wrote:
In my case no volume rendering is i
I followed all the instructions for compiling, installing ParaView. It
seemed to compile without error (64-bit Linux), but when I typed "make
install," I got this error near the end:
[ 88%] Built target QtTesting
[ 89%] Built target pqWidgets
[ 89%] Built target pqWidgetsTest
[ 89%] Built targe
Hmmm. Can you try turning off all wireframes?
-berk
On Tue, Oct 13, 2009 at 3:47 PM, Dominik Szczerba wrote:
> In my case no volume rendering is involved, just a blend of surface meshes
> and some wireframes with varying opacities.
>
> Berk Geveci wrote:
>>
>> Does this happen only with Mesa? Th
In my case no volume rendering is involved, just a blend of surface
meshes and some wireframes with varying opacities.
Berk Geveci wrote:
Does this happen only with Mesa? That looks like volume rendering,
does it happen with geometry rendering?
On Tue, Oct 13, 2009 at 12:02 PM, burlen wrote:
Berk Geveci wrote:
We try to use several offscreen rendering methods on X11. If we still
see an empty screen as a result, we disable offscreen rendering. This
is a common problem with the intel cards. To make sure I understand,
you are not seeing the stitching problem only with the intel cards?
E
I only tried mesa because no graphics hardware are present on that
system. The system is off line today from maintenance, I can see if it
happens with geometry too when it's back on line.
Berk Geveci wrote:
Does this happen only with Mesa? That looks like volume rendering,
does it happen with
Hi,
I want to modify a vtkInformation object (specifically one within a
vtkInformation vector) to point to a new data set.
The use case is I have a derived class for which, in its RequestData
method, I want to do some preprocessing on the input, then send the
preprocessed input to the superclass
What happens if you call MarkModified(None) on your source proxy?
On Tue, Oct 13, 2009 at 11:59 AM, Fabian Wein
wrote:
> I'm still struggling with my live data reader.
>
> A thread calls Update() on my algorithm. This leada to a RequestData()
> and I give back the current data - independend of th
Does this happen only with Mesa? That looks like volume rendering,
does it happen with geometry rendering?
On Tue, Oct 13, 2009 at 12:02 PM, burlen wrote:
> I get the stitching artifacts cvs of a few weeks ago,
> --use-offscreen-rendering & Mesa 7.5,. no error about the empty screen.
>
> http://n
We try to use several offscreen rendering methods on X11. If we still
see an empty screen as a result, we disable offscreen rendering. This
is a common problem with the intel cards. To make sure I understand,
you are not seeing the stitching problem only with the intel cards?
Exact same geometry an
That would be the easiest explanation, but then it is not at all
off-screen rendering or its my wrong understanding about it...
Scott, W Alan wrote:
I believe that ParaView uses the graphics card, and the graphics memory for
it's off screen rendering. I also assume this is done in the back bu
It looks like we forgot to add support for reverse connection to the
simple module. The following should work:
from paraview import servermanager
servermanager.ReverseConnect() # this takes an optional port argument
You will also need to do the following if you want animation to work:
servermana
Yes, that is exactly the same problem I am experiencing on one computer
(integrated intel gfx card) but not the other (NVIDIA).
I do not understand why the gfx card would influence offscreen rendering...
burlen wrote:
I get the stitching artifacts cvs of a few weeks ago,
--use-offscreen-render
You could try using lookmarks, which are design to save a portion of the state
and then apply that to different input data. Lookmarks are created by clicking
on the lookmark icon in that little bar above the view.
That said, although lookmarks are nice when they work, they tend to be quite
inf
The stream tracer filter follows vector fields in 3D space. Tracing flow on
surfaces is problematic because if a position is off the surface, then the
stream tracer will not follow it.
The default stream tracer seed points is a cloud of random points centered
around some location. Because the
Is there a way to accept a (reverse) connection from a remote data server using
the paraview modules in a python shell?
___
Powered by www.kitware.com
Visit other Kitware open-source projects at
http://www.kitware.com/opensource/opensource.html
Please
I believe that ParaView uses the graphics card, and the graphics memory for
it's off screen rendering. I also assume this is done in the back buffer of
the graphics card's double buffer system, but I am not sure.
Alan
> -Original Message-
> From: Dominik Szczerba [mailto:domi...@itis.
Dear last resource :)
My last idea to make my live data reader work is to use animation.
Independed on what time value is requested I give back the current data.
I'll need a repeated loop for that - but once the time values are read,
they are cached and not requested again from my reader.
How c
I get the stitching artifacts cvs of a few weeks ago,
--use-offscreen-rendering & Mesa 7.5,. no error about the empty screen.
http://nashi-submaster.ucsd.edu/movies/SC09/vrl-pv-prob.png
Dominik Szczerba wrote:
Hi, I ma getting stitching artifacts in screenshots larger than my
display (both 3
I'm still struggling with my live data reader.
A thread calls Update() on my algorithm. This leada to a RequestData()
and I give back the current data - independend of the requested time step.
It's just not displayed. I tried with python
Render()
GetRenderView().StillRender()
Show()
all has no
I also tested on another PC, with an NVIDIA gfx card and it is OK there.
So the culprit might be my integrated intel gfx chip in my thinkpad x60.
Still, I expected off-screen rendering to be independent from hardware,
not so?
Dominik
Scott, W Alan wrote:
Looks OK to me. I tested at 1
Is this a surface or a volume that you are creating streamlines for?
If it is a surface you probably want to use "streamtracer with a
custom source" and use the surface as the source (either use subdivide
or quadric clustering to increase/reduce the number of streamlines).
I have written some plug
Hi Fabian,
I am not sure what you mean by a thread (a client side code?),
but perhaps you can simply write
pqApplicationCore::instance()->render();
Regards,
Takuya
Takuya OSHIMA, Ph.D.
Faculty of Engineering, Niigata University
8050 Ikarashi-Ninocho, Nishi-ku, Niigata, 950-2181, JAPAN
From:
Looks OK to me. I tested at 1920x1024, on screen and off screen, and
7680x4096, off screen. I was writing png's, using 3.6.1. No stitching, no
error message, no disabled off screen rendering.
Alan
> -Original Message-
> From: paraview-boun...@paraview.org
> [mailto:paraview-boun..
Check out the vtkAbstractDataArray class and it's derived types for storing
the information you want. You can associate any of those arrays with your
data set through vtkFieldData, vtkPointData, or vtkCellData (probably the
one you want).
ParaView will allow you to color your surface mesh using e
If you really have a simple cube, ie just 8 points conected together, it is
possible that stream tracing be impossible because data are too sparse and the
streamline are not long enough to be considered as valid. I would suggest
trying with a little bit more complex dataset.
From: glada...@hot
Dear All,
Quite simply, I'm using gmsh to generate surface meshes for a CFD
simulation. Gmsh has the wonderful feature of being able to add a
boundary ID flag to a surface or collection of surfaces, this is
essential of setting up a computational problem as I'm sure you all
know. The idea being
Dear Kitware experts and users, :)
I'm still fighting with my live data reader and my deadline is
approaching :(
In my vtkMultiBlockDataSetAlgorithm derived reader
I implement
RequestInformation()
RequestData()
and
RequestUpdateExtent()
and call Modfied() and Update() from a thread.
Sofar all w
Hi!
I have opened a figure in Paraview, just defined by surfaces connected
together. Let’s say a cube with six surfaces. In the xml file the velocity is
defined as PointDataScalars and now I want to create stream traces in the
figure. But when I did that with Filters – Alfabetical – Stream Tr
Hi, I ma getting stitching artifacts in screenshots larger than my
display (both 3.6.1 and recent CVS on linux). I have 'use
offscreen' enabled in the settings, but get an error message it is being
disabled because of a detected empty screen.
Does anyone else have similar problems?
Domini
Berk Geveci wrote:
> Yeah, unfortunately it is hard for us to provide binaries with MPI
> support because MPI is implemented with different (internal) APIs by
> different vendors. We would have to create a different binary for each
> MPI distribution.
>
No problem, building paraview isn't really
Hello,
I try to compil paraview with the CVS sources.
I made changes to successfully compile. These changes are :
- comment the line paraview_build_optional_plugin(H5PartReader "Reader for
*.h5part files" H5PartReader ON) in Plugins/CMakeList.txt because there is
no directory Plugins/H5PartReader
35 matches
Mail list logo
