If nek5000 stores time varying data in numbered files, you should be able to put the existing reader under a time series reader proxy to get animation etc with uniformly temporally spaced data sets. Otherwise the reader will have to be extended to provide and respond to time information requests. That is generally easy to do.<br>
<br>On Wednesday, March 5, 2014, Ravi Tumkur <<a href="mailto:ravi.tumkur@gmail.com">ravi.tumkur@gmail.com</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr"><div><div><div><div>Hi Sam, <br><br></div>Thanks for offering the help. <br><br></div>I looked at the case file you sent, NEK5000 also has a similar ASCII meta file and time-depended mesh data is saved in a set of binary files. I am not able to read this changing mesh into Paraview. <br>
<br></div>I have been using Paraview to visualize moving-mesh which are saved in VTK format for quite sometime and it was working fine. But not able to do the same with NEK5000 output files. I am guessing that this is still a bug/pending enhancement since Paraview has added the NEK5000 compatibility only recently, starting from 4.1.0.<br>
<br></div>Regards,<br>Ravi<br><div><div><br></div></div></div><div><br><br><div>On Sun, Mar 2, 2014 at 1:45 PM, Samuel Key <span dir="ltr"><<a>samuelkey@bresnan.net</a>></span> wrote:<br>
<blockquote style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
Giuseppe, Ravi;<br>
<br>
I have no knowledge of NEK-5000, nor the content of the results
files produced by NEK-5000. However, in my use of ParaView, ParaView
routinely provides moving and distorting meshes in my animations.
For me, there are at least two ways that the moving and distorting
meshes are produced:<br>
<br>
1) If there are vector displacement values supplied at the grid
points, and ParaView's Warp filter is available (that is, not
grayed-out), then the Warp filter will can be used with a scale
value to reduce, amplify, or give true position of the grid,<br>
<br>
2) If one uses the EnSight-formatted simulation results file, and
true-scale deformed mesh plots are the routine case, then EnSight's
*.geom file can contain by time-step the mesh coordinates, and then
without the use of ParaView's Warp filter, true-scale deformed
meshes are obtained automatically. A small EnSight *.case file (an
ASCII-formatted meta file the points to all of the binary files that
contain the simulation results) is attached.<br>
<br>
Sam Key<div><div><br>
<br>
<br>
<div>On 3/2/2014 9:16 AM, Ravi Tumkur wrote:<br>
</div>
<blockquote type="cite">
<div dir="ltr">
<div>
<div>
<div>
<div>
<div>
<div>
<div>Hello Giuseppe,<br>
<br>
</div>
Thanks a lot for replying to my query. <br>
<br>
</div>
I understand that you are also facing the same issue
of only one time-step's mesh geometry begin loaded
when you try to use Paraview 4.1.0 to visualize your
post-processed data.<br>
<br>
</div>
When you say using Praview 4.1.0 resolved your issue at
the beginning of your email, do you mean you were able
to visualize *.fld* data in Paraview? If so, can you
please send an example database set and the metadata
file?<br>
</div>
<br>
</div>
I am trying to visualize moving-mesh results for the basic
problem of vortex-shedding past a cylinder moving with
prescribed motion. The flow results are rendered as expected
in Paraview for each time step, but the mesh appears
stationary, corresponding to "firsttimestep", as you also
observe too.<br>
<br>
</div>
I appreciate your help, a set to example database will get me
started on using Paraview for visualizing NEK-5000 data.<br>
<br>
<br>
</div>
Regards<br>
Ravi<br>
<br>
<div>
<div>
<div>
<div>
<div>
<div>
<div>
<div>
<div><br>
<br>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div><br>
<br>
<div>On Sat, Mar 1, 2014 at 2:45 AM,
Giuseppe Pitton <span dir="ltr"><<a>giuseppe.pitton@gmail.com</a>></span>
wrote:<br>
<blockquote style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="word-wrap:break-word">Hello Ravi,
<div>although Paraview 4.1.0 solved my issue, now I am
experiencing the same problem you report with some
postprocessed data (but I have a fixed mesh). In my
opinion this happens when the dataset has mesh
information stored on each .f or .fld file. I think
Paraview expects to have mesh information only in the
first file of the dataset, and it is not behaving well
if this is not the case.</div>
<div>I suspect this because if I edit my .nek5000 file and
change the value of firsttimestep setting it to n, then
Paraview correctly visualizes the data of the n-th time
step, but again it cannot display correctly the data</div></div></blockquote></div></div></blockquote></div></div></div></blockquote></div></div></blockquote><br><br>-- <br>David E DeMarle<br>Kitware, Inc.<br>
R&D Engineer<br>21 Corporate Drive<br>Clifton Park, NY 12065-8662<br>Phone: 518-881-4909<br>