[aspect-devel] Checkerboard at large scale

[Wolfgang Bangerth]

On 03/29/2018 08:05 PM, Lev Karatun wrote:
> 
> I was trying to run some 3-D compression models, but all I'm getting is 
> a checkerboard pattern (see screenshot attached).

John has already asked the right questions, but I'll add this: The term 
"checkerboards" is usually applied to an *unphysical* oscillation that 
happens (i) as a result of the discretization, and (ii) has a wave 
length that scales with the mesh size. In your case, if the pattern you 
see is stable under mesh refinement, and if the picture you show is on 
an already reasonably refined mesh, then this does not look to me like a 
problem with the discretization, but with the setup of the model. In 
other words, it looks like *this* is the correct solution of the PDE, 
and that you need to dig into *why* this is the solution of the PDE. It 
would then not be artifact of the discretization.

Best
  W.

-- 
------------------------------------------------------------------------
Wolfgang Bangerth          email:                 bangerth at colostate.edu
                            www: http://www.math.colostate.edu/~bangerth/