One step closer to an optimal two-parameter SOR method: a geometric approach (Q1011838)

This paper can be considered a continuation of another paper by the same author [Missouri J. Math. Sci., 21, No.~1, 42--55 (2009; Zbl 1169.65029)] and in the same journal issue. Actually, the first part of this paper is almost the same as the first part of the previous paper. Moreover, the geometric approach is again the chosen tool, but here, in order to go further, the author removes the hypothesis that requires the eigenvalues of the Jacobi iteration matrix to be real. This allows to provide an optimal values for the second parameter when the eigenvalues of the successive overrelaxation (SOR) method are in a certain well-defined region in the complex plane.