[Paraview] building paraview with vtkManta

Wed Jul 23 11:55:58 EDT 2014

Actually, from your description, does not sounds like the right fit here.  I’ll keep chugging away with POVRay.  blender gave up the ghost and I haven’t learned RenderMan/Maya, which would probably be a great approach but has fairly steep learning curve

On Jul 23, 2014, at 8:40 AM, David E DeMarle <dave.demarle at kitware.com<mailto:dave.demarle at kitware.com>> wrote:

On Thu, Jul 17, 2014 at 12:25 PM, Cook, Rich <cook47 at llnl.gov<mailto:cook47 at llnl.gov>> wrote:
Thanks
What kind of experience can I expect?  I couldn’t find a description, tutorial or other presentation on the subject other than the page I stumbled across.

There isn't much to it so far. Basically vtkManta is just a drop in replacement for OpenGL rendering of surfaces.

There are a few extra controls available on the display section of the properties tab for per object display options (material type - metal, dielectric etc) and view section (restored just last week in master) for scene wide display options (enable shadows, rays per pixel, number of manta threads etc).

There are not enough controls available to make it the right option for really high quality (aesthetically pleasing) renderings. In its current level of development you are better off File->Exporting the scene and then importing it into some other system as you are doing. Given enough interest we can of course improve on it from where it sits now and turn it into such an option.

I’m hoping this will scale better than e.g blender does, where I’m hitting size limitations at disappointing scales.  I’m dusting off my old POVray stuff and I recall being seriously limited there as well in terms of size and speed but haven’t gotten it working again yet to be able to tell if it’s the same now as it was.

Scaling should be fine. Scaling is what LANL built it for (http://viz.lanl.gov/publication/presentation/L2.pdf). You can render tens of millions of triangles on a modern workstation interactively with it. In HPC settings my personal record is 10 billion triangles (from a synthetic test data set) at ~10 frames per second. That was with no secondary rays of course, and only after significant queue wait and setup wait times.

I’d love to see your build script, thanks.

Sure, will send it shortly.

— Rich

On Jul 17, 2014, at 6:55 AM, David E DeMarle <dave.demarle at kitware.com<mailto:dave.demarle at kitware.com>> wrote:

Hey Rich,

I just ran across mention of vtkManta support in paraview.

Wow, 2010? You've got some catching up to do. :)

It still compiles, runs and passes high level tests tests. The nightly linux binaries have it if you want to try it out quickly.

To compile it you need to:
* compile manta (specifically r2439)
* configure paraview with PARAVIEW_BUILD_PLUGIN_MantaView ON and MANTA_BUILD variable pointed to your manta build tree.

Let me know if you need the exact syntax for unattended builds and I'll send you a shell script.

How solid is this?

I am aware of a couple of bugs that have crept in while this has been in cold storage (i.e. no project since the initial LANL push to pick up development on it).
* Manta (r2439 specifically) needs patches for OSX 10.8+ and MSVC.
* View level options are not exposed in the GUI (very recent regression)
* off by one bug in point associated color mapping (probably old)

We did get to the "future work" item from that article about integrating manta at the Renderer level so that OpenGL and Manta can coexist and it can be used in standalone builds.

TACC is working on something that will address the limitation that secondary rays don't work in HPC contexts.

Feel free to send me or the list emails for help getting it running.

cheers,

David E DeMarle
Kitware, Inc.
R&D Engineer
21 Corporate Drive
Clifton Park, NY 12065-8662
Phone: 518-881-4909<tel:518-881-4909>

On Wed, Jul 16, 2014 at 8:14 PM, Cook, Rich <cook47 at llnl.gov<mailto:cook47 at llnl.gov>> wrote:
I just ran across mention of vtkManta support in paraview.
http://www.kitware.com/media/html/RenderedRealismAtNearlyRealTimeRates.html

Yay!  Yay?

How solid is this?  The documentation how to enable is pretty vague.
"To use vtkManta in ParaView, simply build Manta as described above, and then turn on both shared libraries and the Manta plug-in when you build ParaView. You will then have to specify the location of the Manta source and build trees as described above.”

Changing my paraview build is always a daunting experience but the idea of getting this working is very attractive.  What are the magic flags to cmake?   My builds are automated and interacting with ccmake is problematic on a build farm.

Here’s what I currently do.  Can you see why I’m confused?   :-)
Is there just a -DVTK_MANTA flag?  Do I still turn OSMESA on?

Thanks for any help.

    $CMAKE_PATH/bin/cmake \
        $CMAKE_SPECIAL_ARGS \
        -DPARAVIEW_ENABLE_PYTHON:BOOL=ON \
-DPARAVIEW_BUILD_QT_GUI:BOOL=OFF \
-DVTK_OPENGL_HAS_OSMESA:BOOL=ON \
 -DOSMESA_LIBRARY:FILEPATH='/usr/lib64/libOSMesa.so' \
 -DVTK_USE_OFFSCREEN:BOOL=ON -DOPENGL_gl_LIBRARY:FILEPATH= -DVTK_USE_X:BOOL=OFF" \
        -DQT_QMAKE_EXECUTABLE:FILEPATH=$QT_PATH/bin/qmake.real \
        -DOPENGL_glu_LIBRARY:FILEPATH="/usr/lib64/libGLU.so" \
        -DCMAKE_INSTALL_PREFIX:PATH="$INSTALL_DIR" \
        -DPARAVIEW_USE_BOOST:BOOL=ON \
        -DPARAVIEW_USE_VISITBRIDGE:BOOL=ON \
        -DVISIT_BUILD_READER_CGNS:BOOL=ON \
        -DVISIT_BUILD_READER_Mili:BOOL=ON \
        -DVISIT_BUILD_READER_Silo:BOOL=ON \
        -DCGNS_INCLUDE_DIR:PATH="${CGNS_PATH}/include" \
        -DCGNS_LIBRARY:FILEPATH="${CGNS_PATH}/lib/libcgns.so;$SZIP_PATH/lib/libsz.so;/usr/lib64/libz.so" \
        -DMILI_INCLUDE_DIR:PATH="$MILI_PATH/include" \
        -DMILI_LIBRARY:FILEPATH="$MILI_PATH/lib/libmili.a;$SZIP_PATH/lib/libsz.so;/usr/lib64/libz.so" \
        -DSILO_INCLUDE_DIR:PATH="$SILO_PATH/include;$MILI_PATH/include" \
        -DSILO_LIBRARY:FILEPATH="$SILO_PATH/lib/libsiloh5.a;$SZIP_PATH/lib/libsz.so;/usr/lib64/libz.so" \
        -DBoost_INCLUDE_DIR:PATH="${BOOST_PATH}/include" \
        -DBoost_LIBRARY_DIRS:FILEPATH="${BOOST_PATH}/lib" \
        -DCMAKE_EXE_LINKER_FLAGS:STRING="$LINKER_FLAGS" \
        -DCMAKE_EXE_LINKER_FLAGS_DEBUG:STRING="$LINKER_FLAGS" \
        -DCMAKE_CXX_FLAGS:STRING=" -I$MILI_PATH/include -I$HDF5_PATH/include  -I${MPI_PATH}/include " \
        -DCMAKE_C_FLAGS:STRING="  -I$MILI_PATH/include -I$HDF5_PATH/include  -I${MPI_PATH}/include " \
        -DVTK_USE_SYSTEM_HDF5:BOOL=ON \
        -DHDF5_ENABLE_SZIP_SUPPORT:BOOL=ON \
        -DHDF5_DIR:PATH="$HDF5_PATH" \
        -DHDF5_C_INCLUDE_DIR:PATH="$HDF5_PATH/include" \
        -DHDF5_INCLUDE_DIR:PATH="$HDF5_PATH/include" \
        -DHDF5_HL_INCLUDE_DIR:PATH="$HDF5_PATH/include" \
        -DHDF5_hdf5_LIBRARY:FILEPATH="$HDF5_PATH/lib/libhdf5.so;$SZIP_PATH/lib/libsz.so;/usr/lib64/libz.so" \
        -DHDF5_hdf5_LIBRARY_DEBUG:FILEPATH="$HDF5_PATH/lib/libhdf5.so;$SZIP_PATH/lib/libsz.so;/usr/lib64/libz.so" \
        -DHDF5_hdf5_LIBRARY_RELEASE:FILEPATH="$HDF5_PATH/lib/libhdf5.so;$SZIP_PATH/lib/libsz.so;/usr/lib64/libz.so" \
        -DHDF5_hdf5_hl_LIBRARY:FILEPATH="$HDF5_PATH/lib/libhdf5_hl.so;$SZIP_PATH/lib/libsz.so;/usr/lib64/libz.so" \
        -DHDF5_hdf5_hl_LIBRARY_DEBUG:FILEPATH="$HDF5_PATH/lib/libhdf5_hl.so;$SZIP_PATH/lib/libsz.so;/usr/lib64/libz.so" \
        -DHDF5_hdf5_hl_LIBRARY_RELEASE:FILEPATH="$HDF5_PATH/lib/libhdf5_hl.so;$SZIP_PATH/lib/libsz.so;/usr/lib64/libz.so" \
        -DBUILD_SHARED_LIBS:BOOL=ON \
        -DBUILD_TESTING:BOOL=OFF \
        $PARAVIEW_SOURCE || errexit "cmake exited with an error status. Please try again."

--
✐Richard Cook
✇ Lawrence Livermore National Laboratory
Bldg-453 Rm-4024, Mail Stop L-557
7000 East Avenue,  Livermore, CA, 94550, USA
☎ (office) (925) 423-9605<tel:%28925%29%20423-9605>
☎ (fax) (925) 423-6961<tel:%28925%29%20423-6961>
---
Information Management & Graphics Grp., Services & Development Div., Integrated Computing & Communications Dept.
(opinions expressed herein are mine and not those of LLNL)

_______________________________________________
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

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

--
✐Richard Cook
✇ Lawrence Livermore National Laboratory
Bldg-453 Rm-4024, Mail Stop L-557
7000 East Avenue,  Livermore, CA, 94550, USA
☎ (office) (925) 423-9605<tel:%28925%29%20423-9605>
☎ (fax) (925) 423-6961<tel:%28925%29%20423-6961>
---
Information Management & Graphics Grp., Services & Development Div., Integrated Computing & Communications Dept.
(opinions expressed herein are mine and not those of LLNL)

--
✐Richard Cook
✇ Lawrence Livermore National Laboratory
Bldg-453 Rm-4024, Mail Stop L-557
7000 East Avenue,  Livermore, CA, 94550, USA
☎ (office) (925) 423-9605
☎ (fax) (925) 423-6961
---
Information Management & Graphics Grp., Services & Development Div., Integrated Computing & Communications Dept.
(opinions expressed herein are mine and not those of LLNL)

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20140723/550c7e82/attachment-0001.html>