CONVERGENCE HISTORY FOR GMRES
 
Errors are displayed in unit:  6
Warnings are displayed in unit:  6
Matrix size:      10
Local matrix size:      10
Restart:       3
Left preconditioning
Iterative classical Gram-Schmidt
Default initial guess x_0 = 0
True residual computed at restart
Maximum number of iterations:   100
Tolerance for convergence: 0.10E-04
Backward error on the unpreconditioned system Ax = b:
    the residual is normalised by ||b||
Backward error on the preconditioned system (P1)A(P2)y = (P1)b:
     the preconditioned residual is normalised by ||(P1)b||
Optimal size for the workspace:    105
 
Convergence history: b.e. on the preconditioned system
 Iteration   Arnoldi b.e.    True b.e.
    1           0.53E+00         --
    2           0.21E+00         --
    3           0.10E+00         --
    4           0.61E-01         --
    5           0.29E-01         --
    6           0.14E-01         --
    7           0.85E-02         --
    8           0.42E-02         --
    9           0.22E-02         --
   10           0.11E-02         --
   11           0.49E-03         --
   12           0.28E-03         --
   13           0.17E-03         --
   14           0.85E-04         --
   15           0.41E-04         --
   16           0.23E-04         --
   17           0.90E-05       0.90E-05
 
Convergence achieved
B.E. on the preconditioned system:   0.90E-05
B.E. on the unpreconditioned system: 0.90E-05
info(1) =  0
Number of iterations (info(2)):    17