Teleconference 2017-05-17

Jump to navigation Jump to search

Attendees: Ken Moreland (SNL), Rob Maynard (Kitware), Dave Lonie (Kitware), Sujin Philip (Kitware), Matt Larsen (LLNL), David Camp (LBNL), Mark Kim (ORNL), Ollie Lo (LANL), Dave Rogers (LANL), Tom Otahal (SNL), James Kress (ORNL)

Rob at GTC last week. Working on the tiling. Also looking at fixing some issues with the normals in Marching Cubes. Also added infrastructure to implement passing cell data in Marching Cubes. (Not complete). Ready for merge once there is a review.

Sujin has implemented implicit functions to VTK-m with virtual functions. Now working on extract cell filters.

There was an OSX dashboard failure. The machine is being rebuilt. Should be up by the end of the week.

Factoring rendering code in cxx files.

David C. came up with an issue with a buggy version of GLEW that caused him problems. He suggests we document better the specific versions of TPL have actually been tested.

Mark gave a presentation at GTC. Mark was having some issues with compiling with Visual Studio 2015. It turned out to be an issue with mixing Debug and Release modes, which are binary incompatible.

Ollie working on spatial subdivision.

LANL has a student from Ohio State who will be working on image compression and volume rendering.

Tom and Alberto did some analysis with the Intel tools for roofline plots on VTK-m benchmarks. The analysis was done with an older version of the tiling. There have some changes that have been made, but they should not affect the analysis significantly.

Tom has also been experimenting with high bandwidth memory on KNL, but that was somewhat outside of VTK-m.