[Octave-patch-tracker] [patch #9282] add minres for sparse linear systems

[Marco Caliari]

Follow-up Comment #13, patch #9282 (project octave):

If you look at algorithm 9.1 on Saad's book, you see that the split
preconditioning for the conjugate gradient is realised at step 6. And it does
not matter whether the preconditioner M is M1*M2. The only thing to do is to
multiply the residual by inv(M). I see that Matlab always use a split
preconditioner:


A = toeplitz (1:4);
M = toeplitz ([2,1,0,0]);
b = A * ones (4, 1);
M2 = chol (M);
M1 = M2';
minres (A ,b ,0 ,1, M1, M2)
minres (A, b, 0, 1, M) % it should be exactly the same
M2 \ minres (M1 \ A / M2, M1 \ b, 0, 1) % equivalent formulation


I would remove the warning about too fast convergence. And some tests do not
work for me.


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/patch/?9282>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/