Skip to content

Conversation

@drbergman
Copy link
Collaborator

  • bottom line: at each mechanics time step, it will suffice to push the cell in a cardinal direction to increase the shortest distance to the boundary
  • this is precisely what the current code currently does the vast majority of the time.
  • now, it will do so for the previously defined edge cases
  • in 2D sims, they looked for cells on line y=x
    • then used signs of x and y positions to determine which direction to push
    • no check for cells in upper left or lower right corners
    • the sign of the positions does not accurately convey which way to push (if all corners are negative, they would all push the same way)
  • in 3D sims, they looked for cells on line x=y=z
    • similar issues but missing more cases (x=\pm y = \pm z)
- bottom line: at each mechanics time step, it will suffice to push the cell in a cardinal direction to increase the shortest distance to the boundary - this is precisely what the current code currently does the vast majority of the time. - now, it will do so for the previously defined edge cases - in 2D sims, they looked for cells on line y=x - then used signs of x and y positions to determine which direction to push - no check for cells in upper left or lower right corners - the sign of the positions does not accurately convey which way to push (if all corners are negative, they would all push the same way) - in 3D sims, they looked for cells on line x=y=z - similar issues but missing more cases (x=\pm y = \pm z)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

1 participant