Reg: Math Kernel Routine

deepu_r34 — Tue, 13 May 2008 07:22:16 GMT

Hi, I'm working on the design of a routine which would to Y = A*X + Y where A is a Matrix whose elements are complex numbers
X & Y are vectors which also consist of complex numbers.

I wanted to know if there is any optimized method to multiply a complex Matrix with a complex vector so as to minimize the memory references and also the arithmetic operations. ie to multiply an element in complex matrix with an element in vector would mean multiplying two complex numbers (a + ib)(c+id) = (ac-bd) + i( ad+bc) where a &b are elements of Matrix and c &d are elements of vector X.

Thanks in advance.....

topic Reg: Math Kernel Routine in Intel® oneAPI Math Kernel Library

Reg: Math Kernel Routine