[Paraview] Volume Render Plot - Unreasonable results with vector components or magnitude

Fri Nov 4 06:05:02 EDT 2016

Hi Alvaro,

your last advise is actually the solution!! ☺ Volume rendering works fine for structured data.

I have been working with multi-block unstructured data (Ensight files exported form Ansys Fluent CFD) which leads to the discussed problems. By the way, that is why I could not find the ‘Volume Rendering’ section on the ‘Properties’ tab since this is shown for structured data only. For unstructured data the ‘Volume Rendering’ section will not appear although a volume render plot is displayed. Maybe a comment in the corresponding section of the PV manual would be helpful.

After converting my unstructured grid to structured data (‘Resample to Image’ as you suggested), volume render works fine (both for ‘Smart’ and ‘GPU Based’ rendering mode) and is MUCH faster (100^3 sampling). Anyway, resampling to structured data is time consuming, creates a “bounding box” and leads to large grids if you want to capture small details. So I guess I keep the current limitations of volume render in mind and work with scalar fields instead which seems to work perfect with unstructured data.

Thanks to everyone! Great work!

Best wishes
Mathias

Von: Alvaro Sanchez [mailto:alvaro.sanchez at kitware.com]
Gesendet: Donnerstag, 3. November 2016 18:33
An: Dietzel Mathias
Betreff: Re: [Paraview] Volume Render Plot - Unreasonable results with vector components or magnitude

Here is a screenshot of what 'Volume Rendering Mode' looks like, the label is on the left (Properties tab).

Cheers,
Alvaro

On Thu, Nov 3, 2016 at 1:31 PM, Alvaro Sanchez <alvaro.sanchez at kitware.com<mailto:alvaro.sanchez at kitware.com>> wrote:
Hi Mathias,

thanks for your feedback.  I downloaded the ParaView-5.2-RC3 binaries and tried them out on Windows 7 and vector data seems to render correctly.  Something important that I missed to point out is that this is currently supported for structured data only, so this might be the issue.  To ensure that you are using the GPU mapper look for the 'Volume Rendering' section on the 'Properties' tab and select 'GPU Based' on  'Volume Rendering Mode'.

One thing you could try (in case your data is unstructured) is using the "Resample to Image" filter to convert your data to structured and then volume render that.

Let us know if that helps.
Alvaro

On Thu, Nov 3, 2016 at 12:35 PM, Dietzel Mathias <M.Dietzel at iab-weimar.de<mailto:M.Dietzel at iab-weimar.de>> wrote:
Hi,

first of all many thanks to the numerous and quick responses. Great work!

I did a quick test of the volume rendering in ParaView 5.2.0-RC3 64 bit (Windows 7). The issue is still the same, that is volume rendering by vector components does not seem to be fixed.

Additionally, I noticed one more issue occurring in PV 5.2-RC3: when changing from one component (e.g. x) to another one (e.g. y) in volume rendering, the title of the legend changes but nothing else. The legend keeps the same scaling parameters like min. and max. value (which may be intended) and the volume render plot is not re-rendered. One has to hit “Rescale” in order to get an update of the volume render plot (which is confusing to me).

Alvaro mentioned that I ought to use the “GPU mapper” but I was not able to find this option in the settings or properties. Actually I don’t know whether I am using the CPU or GPU or which mapper in particular. Where can I adjust that?

Let me know if I can give you any further support. Many thanks!

Best wishes
Mathias

Von: ParaView [mailto:paraview-bounces at paraview.org<mailto:paraview-bounces at paraview.org>] Im Auftrag von Andy Bauer
Gesendet: Donnerstag, 3. November 2016 14:40
An: Cory Quammen
Cc: Alvaro Sanchez; paraview
Betreff: Re: [Paraview] Volume Render Plot - Unreasonable results with vector components or magnitude

Hi,
To add to Cory's feedback -- installers and source code are available for the 5.2 release candidate if you'd like to try it out and see if it's fixed. If it's not fixed and you let us know ASAP, we may be able to get it fixed in time for the 5.2 release. You can get the release candidate and paraview.org/download<http://paraview.org/download> -- just choose the 5.2 RC3.
Best,
Andy

On Thu, Nov 3, 2016 at 9:17 AM, Cory Quammen <cory.quammen at kitware.com<mailto:cory.quammen at kitware.com>> wrote:
Hi Mathias and Andrew,

Volume rendering by vector component should be fixed in the ParaView
5.2 release, due out in the next two weeks.

See https://gitlab.kitware.com/paraview/paraview/merge_requests/1105

Thanks,
Cory

On Thu, Nov 3, 2016 at 7:06 AM, Andrew <antech777 at gmail.com<mailto:antech777 at gmail.com>> wrote:
> Hello. I noticed the same behaviour in ParaView 4.3 on CentOS 6.7. Looks
> like bug. Worked around with calculating the velocity magnitude as scalar.
> There is also another bug or feature of volume rendering: I couldn't make it
> render for all domains, it only renders one domain (for example, when I load
> a CFD study results in EnSight Gold format).
>
> 2016-11-03 13:51 GMT+03:00 Dietzel Mathias <M.Dietzel at iab-weimar.de<mailto:M.Dietzel at iab-weimar.de>>:
>>
>> Hello everyone,
>>
>>
>>
>> I am using Paraview’s Volume Render plot to visualize a velocity field
>> inside some technical equipment. In case of plotting a component of the
>> vector field or the magnitude of the vector (both from the drop-down menu)
>> with volume render I get unreasonable results, e.g. minimum or maximum zones
>> are not shown correctly or the flow direction seems wrong. However, if I
>> calculate a scalar value from the vector that corresponds to the component
>> or to the magnitude (e.g. using the calculator filter in Paraview) and then
>> plot by volume render, the fluid field is represented correctly.
>>
>>
>>
>> The data itself seems OK since a surface or slice plot leads to the same
>> (correct) results, no matter  whether I use the vector component directly or
>> the scalar value. I tested the behavior with several data sets. I am using
>> Paraview 4.3.1 64-bit and Paraview 5.1.2 64-bit – both show the same
>> behavior.
>>
>>
>>
>> I appreciate any comment.
>>
>>
>>
>> Best regards
>>
>> Mathias Dietzel
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> Powered by www.kitware.com<http://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
>>
>> Search the list archives at: http://markmail.org/search/?q=ParaView
>>
>> Follow this link to subscribe/unsubscribe:
>> http://public.kitware.com/mailman/listinfo/paraview
>>
>
>
> _______________________________________________
> Powered by www.kitware.com<http://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
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>

--
Cory Quammen
Staff R&D Engineer
Kitware, Inc.
_______________________________________________
Powered by www.kitware.com<http://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

Search the list archives at: http://markmail.org/search/?q=ParaView

Follow this link to subscribe/unsubscribe:
http://public.kitware.com/mailman/listinfo/paraview

--
Alvaro Sanchez
Kitware, Inc.
Senior R&D Engineer
21 Corporate Drive
Clifton Park, NY 12065-8662
Phone: 518-881-4901<tel:518-881-4901>

--
Alvaro Sanchez
Kitware, Inc.
Senior R&D Engineer
21 Corporate Drive
Clifton Park, NY 12065-8662
Phone: 518-881-4901
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20161104/79bf3a63/attachment.html>