Joe Borġ | 16 Apr 16:24 2014
Utkarsh Ayachit | 16 Apr 14:45 2014

Changes to Settings in ParaView

Folks,

We've been working on changing how application settings work in
ParaView for the next release.

Here's a blog post that talks about the UI changes:
http://www.kitware.com/blog/home/post/652

Let us know if you have any feedback. The code hasn't landed in
git/master yet and won't for a couple more weeks.

Utkarsh
Richard Grenon | 15 Apr 12:58 2014
Picon

LIC plugin version 2 of PV 410 Windows 32 bits fails : unsupported NVIDIA card driver ?

Hello,

On Windows Vista 32 bits with NVIDIA graphic card GeForce 8700M GT, the 
PV 401 LIC plugin works fine, but the PV 410 LIC plugin gives always 
this error:

> ERROR: In 
> C:\DBD\pvs-x32\paraview\src\paraview\VTK\Rendering\OpenGL\vtkTextureObject.cxx, 
> line 1231
> vtkTextureObject (12FAC3F0): failed at glTexImage2D 1 OpenGL errors 
> detected
>   0 : (1280) Invalid enum
>
>
>
> Generic Warning: In 
> C:\DBD\pvs-x32\paraview\src\paraview\VTK\Rendering\LIC\vtkSurfaceLICPainter.cxx, 
> line 981
> SurfaceLIC is not supported
> vtkWin32OpenGLRenderWindow
> NVIDIA Corporation
> 2.1.2
> GeForce 8700M GT/PCI/SSE2
> LIC support = 1
> floating point texture formats = 0
> render buffers = 1
>
>
> ERROR: In 
> C:\DBD\pvs-x32\paraview\src\paraview\VTK\Rendering\LIC\vtkSurfaceLICPainter.cxx, 
(Continue reading)

Richard Beare | 15 Apr 01:45 2014
Picon

Fixing the nifti plugin

Hi Everyone,
I'd like to use paraview to combine volume rendering with various forms of geometric information - meshes and streamlines. The volume information is typically in the form of nifti files, while the geometries are VTK files. My problem is that the nifti plugin doesn't fully utilize the nifti header information. I think that what needs to happen is an appropriate transformation should be set up in the loading plugin, but I have no idea what form this should take. Is anyone aware of other plugins that load data that includes transform information that might be appropriate as a model?

Thanks
<div><div dir="ltr">
<div>Hi Everyone,<br>
</div>I'd like to use paraview to combine volume rendering with various forms of geometric information - meshes and streamlines. The volume information is typically in the form of nifti files, while the geometries are VTK files. My problem is that the nifti plugin doesn't fully utilize the nifti header information. I think that what needs to happen is an appropriate transformation should be set up in the loading plugin, but I have no idea what form this should take. Is anyone aware of other plugins that load data that includes transform information that might be appropriate as a model?<br><br>Thanks<br>
</div></div>
Max Staufer | 12 Apr 22:10 2014
Picon
Picon

Multiple pieces unstructured grid

Hi all,

   I found that it is easy to write a VTU with multiple pieces, say
diffrent parts of a solution,
however is there a way to make the pieces "selectable" and named ?

Thanks for the help

Max
Léo Pessanha | 11 Apr 23:13 2014
Picon

Seismic Data - SEG-Y

Hi,

I did some research, and i saw some e-mails with this subject in the past, but they didn`t help

Nowadays, where is Paraview when it comes to Seismic Data like .sgy? 

Is there any plans about seismic data readers , or some kind of help about tranforming this type of data to another one that is known by Paraview?

Att,
Leonardo Pessanha
<div><div dir="ltr">Hi,<div><br></div>
<div>I did some research, and i saw some e-mails with this subject in the past, but they didn`t help</div>
<div><br></div>
<div>Nowadays, where is Paraview when it comes to Seismic Data like .sgy?&nbsp;</div>
<div><br></div>
<div>Is there any plans about seismic data readers , or some kind of help about tranforming this type of data to another one that is known by Paraview?<br><br>Att,<br>Leonardo Pessanha</div>
</div></div>
Noah Reddell | 11 Apr 08:40 2014

Seeking advice for high-order and high-dimensional discontinuous Galerkin data

 I am trying to setup a visualization solution for my research code which produces structured, nodal data, based on a discontinuous Galerkin solution method for fluid equations.

  • The domain is either 2D, 3D, 4D, 5D, or 6D.  Yes, greater than 3D, because my primary model produces kinetic data sampled in phase space (3D position and 3D velocity).
  • The domain is discretized by elements that are generalized N-dimensional quadrilaterals. (quadrilaterals, hexahedrals, etc.) organized in a block-structured mesh.
  • The solution on any given element is smooth, represented by arbitrary order polynomials and sampled at surface and interior node locations.  I tend to use high order elements compared to typical practice, i.e. 7th, 8th, 9th-order. I’d like the visualization to capture some of the element’s internal solution structure rather than just a linear interpolation between element vertices.
  • The solution is multi-valued at faces where two elements are adjacent and I want the visualization to well-capture the surface discontinuity.
  • My mesh and nodal data are both stored in HDF5 files, with some flexibility to modify formatting.


I spent some time reviewing the ParaView user guide and document for plugin reader development and now have a few questions before I can proceed.

Obviously ParaView is focused on 2D and 3D data. The underlying VTK looks to only handle up to 3D vertex position vectors, for example.  My takeaway from this, is it is not possible to describe a 4D or higher mesh to ParaView directly.  That’s not surprising but seems to suggest I will have to write a custom reader plugin to access my 4D+data.  My survey of the documentation makes me think that I’ll be able to provide parameters to the reader plugin that control the dataset.  For example, specify what dimensions to slice and where to slice.  I suppose I need to similarly parameterize the load of the mesh.  Does this sound like a workable approach?

For the data discontinuities at element faces, I’ve seen discussions in this forum that suggest unstructured grids with duplicated/co-located vertices allows for assigning two different data values at the same physical location.  Since I am working with structured grids and data, I’m wondering if a different technique might work.  How about inserting zero-volume collapsed elements between each real element?  In this way, the real elements are still adjacent, but their vertices / nodes can be assigned different vales.  One concern I have is the hidden element would have zero-volume.  I’d hate to develop that design and then have some ParaView internal function break due to a divide by zero error or something similar.  Another problem area might be plotting stream tubes from the data set.

Lastly, for the high order visualization, I thinking the most straight forward approach is to interpolate my structured high order elements  into a new fine mesh of structured linear elements.  That should be good enough.  On this forum, vtkBridgeDataSet has been mentioned as a possible approach but requiring much development effort.  That was several years ago.  Has there been any progress by others with similar high-order or spectral data sets?

Here’s an old thread that brought up the same basic goals as mine: http://www.paraview.org/pipermail/paraview/2006-October/003899.html
(The numerical scheme and element types are not exactly what I’m working on, but the ideas are similar.)

I really appreciate any feedback.

Best Regards,

Noah Reddell
University of Washington

<div>
<div><div>
<span>&nbsp;I am trying to setup a visualization solution for my research code which produces structured, nodal data, based on a discontinuous Galerkin solution method for fluid equations.</span><br><ul>
<br><li>The domain is either 2D, 3D, 4D, 5D, or 6D.&nbsp; Yes, greater than 3D, because my primary model produces kinetic data sampled in phase space (3D position and 3D velocity).<br>
</li>
<li>The domain is discretized by elements that are generalized N-dimensional quadrilaterals. (quadrilaterals, hexahedrals, etc.) organized in a block-structured mesh.<br>
</li>
<li>The solution on any given element is smooth, represented by arbitrary order polynomials and sampled at surface and interior node locations. &nbsp;I tend to use high order elements compared to typical practice, i.e. 7th, 8th, 9th-order. I&rsquo;d like the visualization to capture some of the element&rsquo;s internal solution structure rather than just a linear interpolation between element vertices.<br>
</li>
<li>The solution is multi-valued at faces where two elements are adjacent and I want the visualization to well-capture the surface discontinuity.<br>
</li>
<li>My mesh and nodal data are both stored in HDF5 files, with some flexibility to modify formatting.<br>
</li>
</ul>
<br><br><span>I spent some time reviewing the ParaView user guide and</span><span>&nbsp;document for plugin reader development and now have a few questions before I can proceed.</span><br><br>Obviously ParaView is focused on 2D and 3D data. The underlying VTK looks to only handle up to 3D vertex position vectors, for example.&nbsp; My takeaway from this, is it is not possible to describe a 4D or higher mesh to ParaView directly. &nbsp;That&rsquo;s not surprising but seems to suggest I will have to write a custom reader plugin to access my 4D+data. &nbsp;My survey of the documentation makes me think that I&rsquo;ll be able to provide parameters to the reader plugin that control the dataset. &nbsp;For example, specify what dimensions to slice and where to slice. &nbsp;I suppose I need to similarly parameterize the load of the mesh. &nbsp;Does this sound like a workable approach?<br><br>For the data discontinuities at element faces, I&rsquo;ve seen discussions in this forum that suggest unstructured grids with duplicated/co-located vertices allows for assigning two different data values at the same physical location. &nbsp;Since I am working with structured grids and data, I&rsquo;m wondering if a different technique might work. &nbsp;How about inserting zero-volume collapsed elements between each real element? &nbsp;In this way, the real elements are still adjacent, but their vertices / nodes can be assigned different vales. &nbsp;One concern I have is the hidden element would have zero-volume. &nbsp;I&rsquo;d hate to develop that design and then have some ParaView internal function break due to a divide by zero error or something similar. &nbsp;Another problem area might be plotting stream tubes from the data set.<div><br></div>
<div>Lastly, for the high order visualization, I thinking the most straight forward approach is to interpolate my structured high order elements &nbsp;into a new fine mesh of structured linear elements. &nbsp;That should be good enough. &nbsp;On this forum,&nbsp;vtkBridgeDataSet has been mentioned as a possible approach but requiring much development effort. &nbsp;That was several years ago. &nbsp;Has there been any progress by others with similar high-order or spectral data sets?</div>
<div><br></div>
<div>Here&rsquo;s an old thread that brought up the same basic goals as mine:&nbsp;<a href="http://www.paraview.org/pipermail/paraview/2006-October/003899.html">http://www.paraview.org/pipermail/paraview/2006-October/003899.html</a><br>(The numerical scheme and element types are not exactly what I&rsquo;m working on, but the ideas are similar.)<br><br><span>I really appreciate any feedback.</span><div><br></div>
<div>Best Regards,</div>
</div>
<div><br></div>
<div>Noah Reddell</div>
<div>University of Washington</div>
</div></div>
<br>
</div>
Mohammad Mirzadeh | 11 Apr 03:10 2014
Picon

loading vtu files crashes paraview in parallel

Hi,

I have a modest size problem, 100M unstructured, and I can easily load into paraview with 64 processors / 128G ram when I load the pvtu file.

I can also work on the file, meaning creating contours, etc. However, when I load all vtu files separately and create contours on each file separately it seems that PV runs out of memory and my process gets terminated.

Is this normal?

<div><div dir="ltr">Hi,<div><br></div>
<div>I have a modest size problem, 100M unstructured, and I can easily load into paraview with 64 processors / 128G ram when I load the pvtu file.</div>
<div><br></div>
<div>I can also work on the file, meaning creating contours, etc. However, when I load all vtu files separately and create contours on each file separately it seems that PV runs out of memory and my process gets terminated.</div>

<div><br></div>
<div>Is this normal?</div>
<div><br></div>
</div></div>
Priyanka Perera | 10 Apr 09:39 2014
Picon

Plotting from SQL

Hi,

Does ParaView support plotting from SQL DB ? Any pointers\insights for  
this be would be very helpful.

Sorry if this is already asked somewhere.

Thanks
Priyanka
Xujun Zhao | 9 Apr 22:10 2014
Picon

plot 2D contour with dashed lines

Hi all,

I have a 2D function f(x,y), and want to make a 2D contour plot with dashed lines. But I didn't find any options that I can change the line types, although I can change colors, line width and others. Can anyone help me figure it out? Thank you.

XZ
<div><div dir="ltr">
<span>Hi all,</span><div><br></div>
<div>I have a 2D function f(x,y), and want to make a 2D contour plot with dashed lines. But I didn't find any options that I can change the line types, although I can change colors, line width and others. Can anyone help me figure it out? Thank you.</div>
<div><br></div>
<div>XZ</div>
</div></div>
Jack Vant | 9 Apr 21:59 2014

trying to install Paraview - last pass errors but no logs

We're trying to install Paraview on a cluster here.  We've used mvapich, mvapich2 and openmpi and turned off qt to do a server installation.  After going through the iterative process with ccmake numerous times, configure shows 99% then a screen pops up that says it found the blast libs and at the bottom of the the screen it says there were errors on the last pass.  It doesn't log any errors in any of the logs in the build directory.  Has anybody else seen this behavior?  Can I buy an error message?

--
Jack Vant
System Engineer - HPC
Office of Information Technology
Boise State University
208-426-4446
208-863-0031
<div><div dir="ltr">We're trying to install Paraview on a cluster here. &nbsp;We've used mvapich, mvapich2 and openmpi and turned off qt to do a server installation. &nbsp;After going through the iterative process with ccmake numerous times, configure shows 99% then a screen pops up that says it found the blast libs and at the bottom of the the screen it says there were errors on the last pass. &nbsp;It doesn't log any errors in any of the logs in the build directory. &nbsp;Has anybody else seen this behavior? &nbsp;Can I buy an error message?<br clear="all"><div><br></div>-- <br><div dir="ltr">Jack Vant<br>System Engineer - HPC<br>Office of Information Technology<br>Boise State University<br>208-426-4446<br>208-863-0031</div>
</div></div>

Gmane