Dan Ibanez

This really falls under the general category of "very low quality elements". MeshAdapt's quality control is not that strong to begin with, and you've exacerbated the situation by first creating...

Yea, one concrete direction to go from here would be to allow shape fixing to be enabled for periodic meshes, with a reduced set of operators, in this case just...

@KennethEJansen , your pictures didn't really get through, but the general idea did. You've still got terrible elements popping up. I think this is mostly due to MeshAdapt not trying...

@KennethEJansen, in terms of development time, I actually think these suggestions of yours are the fastest to implement, in order of ease: 2) "find the situations that create these pancakes...

Actually, you may be able to do this from Chef without modifying MeshAdapt by changing this input variable: https://github.com/SCOREC/core/blob/master/ma/maInput.h#L68 The chosen value should probably the the current worst quality in...

@liruipeng this is important for our CUDA-capable production code. I think a lot of time is spent just creating the preconditioner every time step, we would love to be able...

Any feedback on this?

@liruipeng thanks for the reply! Our matrix always has the same structure (non-zero sparsity pattern) and the coefficients change slowly from one time step to the next, so maybe there...

@bartlettroscoe My guess would be that every call to `Teuchos::size` needs to be fully namespace qualified. Let me see how much effort it would take to do a PR. In...

So, the fundamental difference is disk space versus runtime. The current small/large distinction is there to minimize the disk space consumed by the small test files. This is not inherently...