09-09-2009 10:34 AM
I am currently using the IPP for matrix maths. This IPP library contains functions that I can use to transpose matricies, multipy them, etc. The element of each matrix is of data type Ipp32f. All this is fine. I can make matricies of real numbers and manipulate them beautifully.
The problem however, is how do I make matricies of COMPLEX numbers using IPP????
This is driving me nuts.
09-10-2009 03:13 PM
IPP small matrix domain is mostly targeted for media and 3D gaming math areas where complex arithmetic is not required. I would recommend yo to use Intel MKL for complex matrix operations.
09-11-2009 10:07 PM
in addition, anote need to be added regarding using MKL - Intel MKL supports such operation with complex matrix but, please take into your account that Intel MKL is mostly optimized for middle and large size of matrix.By another words, if your matrix size will be like 3x3, 6x6 then the performance of such operations will be not very good.