Visualization issueshttps://develop.openfoam.com/modules/visualization/-/issues2020-06-24T13:48:50Zhttps://develop.openfoam.com/modules/visualization/-/issues/1runtTimePostProcessing issues with interpolated2020-06-24T13:48:50ZMark OLESENruntTimePostProcessing issues with interpolatedAs @Prashant noted during testing, the unlikely (undesirable) combination
- a sampled cutting plane with interpolate = true
- rendering in parallel
The result looks like rubbish.
If we switch to VTK rendering in serial - works fine, so...As @Prashant noted during testing, the unlikely (undesirable) combination
- a sampled cutting plane with interpolate = true
- rendering in parallel
The result looks like rubbish.
If we switch to VTK rendering in serial - works fine, so the problem is not in the sampler.
Normally we would use 'interpolate = false' when sampling, but use 'smooth = true' for viz to get pretty pictures.
Needs more investigationMark OLESENMark OLESENhttps://develop.openfoam.com/modules/visualization/-/issues/3integration of temporal caching2021-01-15T15:45:21ZMark OLESENintegration of temporal cachingMentioned in email chain with @sbna, @IvanSpisso David Demarle and @acbauer etc.
From Dave
> My changes to update openFOAM for Catalyst v5.9 (at least the recently legacy mode) and adding in the new temporal option are available here: h...Mentioned in email chain with @sbna, @IvanSpisso David Demarle and @acbauer etc.
From Dave
> My changes to update openFOAM for Catalyst v5.9 (at least the recently legacy mode) and adding in the new temporal option are available here: https://github.com/demarle/openfoamvismodule. Also etc/casedicts/insitu/catalyst/printChannels.py needs python3 print()s.
> As described in the video I just used the propeller tutorial with a fancied up Catalyst script to do OSPRay path traced rendering and the cache-rewind-revise-playback-and-continue trick to exercise all of that beautiful memory capacity. My changes to the case files are attached.
> If you attended SC you can see my recorded presentation here:
https://www.eventscribe.net/2020/includes/tracking/clickTracking.asp?lfp=R0M2WEp6UWtPdll4czk2T1hSVFlROGUvY2ZWWTFyWklyY2ZacVgrZUk0L2pBTWdPNm1GcmF5NVpGYlpLN2hzRXVZTlo1dVkyam13NGJ0bmVKOUc5dWVyYjEvdlIxdEFpZE02OVA5M2dKUGdVZko1eEQ1SE9semljMVBJVUFCNkNVSlN4VGZnSHlSVkFwNnR4YmZLVEl5YWVKS0NVcTN4YVR1OXU5bzdhWFRYdDlEamF0NWorbkt3WE95MlorTCtzOUp2SGpQWkFwc3FCajlJeTcyU25zZG5RR2NsWXVqdXd3bmtOUXBrTHZNQT0=
> Otherwise the slides are here:
https://drive.google.com/file/d/12ilOYNBfcDYgIaOOfW4eO_o1npAi7Eib/view?usp=sharinghttps://develop.openfoam.com/modules/visualization/-/issues/4Installation issue with CUDA 10 + Gcc82021-01-15T15:49:32ZSimone BnaInstallation issue with CUDA 10 + Gcc8@mark As reported in this issue, https://github.com/LLNL/blt/issues/341, check for working CUDA compiler fails (Cuda10 + Gcc8).
I solved on Marconi100 by adding -DCMAKE_CUDA_FLAGS="-std=c++11" to cmakeOpts
cmakeOpts="-DCMAKE_INSTALL...@mark As reported in this issue, https://github.com/LLNL/blt/issues/341, check for working CUDA compiler fails (Cuda10 + Gcc8).
I solved on Marconi100 by adding -DCMAKE_CUDA_FLAGS="-std=c++11" to cmakeOpts
cmakeOpts="-DCMAKE_INSTALL_PREFIX=$CMAKE_INSTALL_PREFIX -DCMAKE_CUDA_FLAGS="-std=c++11""https://develop.openfoam.com/modules/visualization/-/issues/6Symbols not found at runtime2021-07-23T13:37:44ZSimone BnaSymbols not found at runtime@mark OpenFOAM crashes when it loads the visualization libraries (catalystFoam and runTimePostProcessing) since it does not find some python, vtk and paraview symbols.
Solved by adding explicitly the dependency on the python libraries a...@mark OpenFOAM crashes when it loads the visualization libraries (catalystFoam and runTimePostProcessing) since it does not find some python, vtk and paraview symbols.
Solved by adding explicitly the dependency on the python libraries and fixing the list of vtk and paraview libraries.https://develop.openfoam.com/modules/visualization/-/issues/8Prototype an update to Catalyst V2 APIs2021-08-10T09:16:56ZRobert Manson-SawkoPrototype an update to Catalyst V2 APIs@mark,
This is a carry over from [ParaView discourse](https://discourse.paraview.org/t/catalyst-v2-for-openfoam/6903) exchange we had a few months ago. I wasn't able to technically work on the stub I created back then, but I would like ...@mark,
This is a carry over from [ParaView discourse](https://discourse.paraview.org/t/catalyst-v2-for-openfoam/6903) exchange we had a few months ago. I wasn't able to technically work on the stub I created back then, but I would like to now return to this project and make some progress this calendar year. Moreover, we are working to secure backing from our current research programme for a 9month demo project which will cover half of my time and another colleague from UKRI. The objective is to adapt work-flows in both organisations to use in-situ and evaluate their performance on both cloud and on-prem instances.
This message is just to gauge your interest. At this stage, I am only asking whether you would like to see any scaling or performance analysis results we may come up with.
Also, I have read your [blog post](https://olesenm.github.io/2020/11/11/openfoam-code-provenance/) and in the long run we would like to be able to contribute to visualisation module (if we develop anything worth contributing of course).Mark OLESENMark OLESEN