[vtk-developers] vtkUnstructuredGrid - Not rendering Overlapping Cells

Fri Feb 22 12:13:06 EST 2013

I'm guessing that you are using a vtkDataSetMapper (or equivalent) which
under the hood uses some sort of geometry filter.

When you use different nodes, the boundary operations return the number of
uses of a face as being <=1 hence a boundary face, and hence rendered. With
the same nodes, each face is used twice hence is an interior face and not
extracted, and hence nothing is rendered.

W

On Fri, Feb 22, 2013 at 12:03 PM, Bhanu Hariharan <bhanu at petrotel.com>wrote:

> Thanks Will.
> You are right. In my application,  I do use a filter BUT that is a
> VTKThreshold filter for showing only certain cells.
> If the two overlapping cells pass this threshold filter, then I don't see
> them at all (if i use structured grid, then you are right I see something
> not pretty but I see something).
>
> I did not use a VTKDataSetSurface filter. I understand if I used that then
> only the boundary cells will be rendered.
> But what I used was just a threshold filter.
>
> Infact I created a small ascii file with 2 cells and opened in Paraview as
> a Legacy VTK file.
> No filter. I cant see the two cells if ALL NODES ARE IDENTICAL.
>
> But if some nodes are identical and some different, then I see some
> surfaces.
>
> Both cases copied here.
>
> //===============ALL NODES IDENTICAL=================NOTHING RENDERED
> # vtk DataFile Version 3.0
> vtk output
> ASCII
> DATASET UNSTRUCTURED_GRID
> POINTS 8 float
> 0 0 0
> 1 0 0
> 1 1 0
> 0 1 0
> 0 0 1
> 1 0 1
> 1 1 1
> 0 1 1
> CELLS 2 18
> 8 0 1 3 2 4 5 7 6
> 8 0 1 3 2 4 5 7 6
> CELL_TYPES 2
> 12
> 12
>
>
> //===============SOME NODES IDENTICAL=================SOME FACES RENDERED
> # vtk DataFile Version 3.0
> vtk output
> ASCII
> DATASET UNSTRUCTURED_GRID
> POINTS 10 float
> 0 0 0
> 1 0 0
> 0 1 0
> 1 1 0
> 0 0 1
> 1 0 1
> 0 1 1
> 1 1 1
> 1 1 0
> 1 1 1
> CELLS 2 18
> 8 0 1 3 2 4 5 7 6
> 8 0 1 8 2 4 5 9 6
> CELL_TYPES 2
> 12
> 12
>
>
> Thanks
> Bhanu
>
>
>
>
> On Fri, Feb 22, 2013 at 10:40 AM, Will Schroeder <
> will.schroeder at kitware.com> wrote:
>
>> Rendering is not the problem. You can render a zillion things on top of
>> each other. It won't look pretty but you'll see something.
>>
>> Probably what is the problem is that you are going through some sort of
>> geometry filter. For 3D cells, the geometry filter looks at the number of
>> times each "face" is used by 3D cells. It the number of uses is <=1, then
>> it's a boundary face and is rendered. If used 2 times, then it's an
>> interior cell and not extracted (nor rendered).
>>
>> W
>>
>>
>> On Fri, Feb 22, 2013 at 10:56 AM, Bhanu Hariharan <bhanu at petrotel.com>wrote:
>>
>>> Thanks. I will look into the composite dataset but I do not have
>>> anything other than hexahedrons.
>>>
>>> I think it looks like VTKUnstructuredGrid will not render in the
>>> situation of overlapping cells and I have to find a way out outside of VTK.
>>>
>>>
>>>
>>>
>>> On Fri, Feb 22, 2013 at 9:28 AM, Sebastien Jourdain <
>>> sebastien.jourdain at kitware.com> wrote:
>>>
>>>> VTK has other type of data structure such a composite-dataset /
>>>> multi-block which let you create a dataset that is composed of multiple
>>>> dataset (i.e. Unstructured grid in your case).
>>>> And what you do is basically share the same vtkPoints across all your
>>>> unstructured grids which will then be gathered by the composite dataset.
>>>> and rendered as a single dataset.
>>>>
>>>> Hope that explain a bit more,
>>>>
>>>> Seb
>>>>
>>>>
>>>> On Fri, Feb 22, 2013 at 10:24 AM, Bhanu Hariharan <bhanu at petrotel.com>wrote:
>>>>
>>>>> Sebastian, I do not understand what you mean by multi-block. Could you
>>>>> please elaborate. Thanks.
>>>>>
>>>>>
>>>>> On Fri, Feb 22, 2013 at 9:20 AM, Sebastien Jourdain <
>>>>> sebastien.jourdain at kitware.com> wrote:
>>>>>
>>>>>> Unless it's a multi-block and then it is not the same dataset
>>>>>> anymore...
>>>>>>
>>>>>>
>>>>>> On Fri, Feb 22, 2013 at 10:17 AM, Sebastien Jourdain <
>>>>>> sebastien.jourdain at kitware.com> wrote:
>>>>>>
>>>>>>> Then I'm not sure that VTK can handle 2 cells that are based on
>>>>>>> exactly the same points ID.
>>>>>>>
>>>>>>>
>>>>>>> On Fri, Feb 22, 2013 at 10:12 AM, Bhanu Hariharan <
>>>>>>> bhanu at petrotel.com> wrote:
>>>>>>>
>>>>>>>> Thanks Sebastian. For the 2 cell case, I start with 16 points and
>>>>>>>> identify duplicates and bring it down to 8 and the two cells now map to the
>>>>>>>> same 8 nodes. In this scenario, can VTK render both cells with some flags
>>>>>>>> turned on.
>>>>>>>> Or can you please suggest another way.
>>>>>>>>
>>>>>>>> But removing duplicate nodes is a must for my application due to
>>>>>>>> memory constraints.
>>>>>>>>
>>>>>>>> -Bhanu
>>>>>>>>
>>>>>>>>
>>>>>>>> On Fri, Feb 22, 2013 at 9:03 AM, Sebastien Jourdain <
>>>>>>>> sebastien.jourdain at kitware.com> wrote:
>>>>>>>>
>>>>>>>>> duplicate the points.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Fri, Feb 22, 2013 at 9:59 AM, bhanu <bhanu at petrotel.com> wrote:
>>>>>>>>>
>>>>>>>>>> Please can anyone answer this query - is this forum active?
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> --
>>>>>>>>>> View this message in context:
>>>>>>>>>> http://vtk.1045678.n5.nabble.com/vtkUnstructuredGrid-Not-rendering-Overlapping-Cells-tp5718739p5718763.html
>>>>>>>>>> Sent from the VTK - Dev mailing list archive at Nabble.com.
>>>>>>>>>> _______________________________________________
>>>>>>>>>> Powered by www.kitware.com
>>>>>>>>>>
>>>>>>>>>> Visit other Kitware open-source projects at
>>>>>>>>>> http://www.kitware.com/opensource/opensource.html
>>>>>>>>>>
>>>>>>>>>> Follow this link to subscribe/unsubscribe:
>>>>>>>>>> http://www.vtk.org/mailman/listinfo/vtk-developers
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>> _______________________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://www.vtk.org/mailman/listinfo/vtk-developers
>>>
>>>
>>>
>>
>>
>> --
>> William J. Schroeder, PhD
>> Kitware, Inc.
>> 28 Corporate Drive
>> Clifton Park, NY 12065
>> will.schroeder at kitware.com
>> http://www.kitware.com
>> (518) 881-4902
>>
>
>

-- 
William J. Schroeder, PhD
Kitware, Inc.
28 Corporate Drive
Clifton Park, NY 12065
will.schroeder at kitware.com
http://www.kitware.com
(518) 881-4902
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/vtk-developers/attachments/20130222/a8e814a1/attachment.html>