On Sun, Mar 18, 2012 at 18:45, Matthew Knepley <[email protected]> wrote:

No, I don't want the nonzeros removed, I want the values set to zero, exactly like I said. This is for robustly checking that a matrix created in parallel is the same as one made on a different number of processors. Something I do for ex62, but we obviously do not give a shit about doing.

What? Why not do a MatGetRedundantMatrix() on the parallel thing and compute a norm of the difference to the serial thing? And if you really want to stay parallel and print all the non-small values, MatDuplicate and fill up the new matrix with the thresholded values. Note, there was a request a while back to be able to make an ASCII viewer use a user-defined floating-point format string.