Intel® Fortran Compiler
Build applications that can scale for the future with optimized code designed for Intel® Xeon® and compatible processors.

FFT in IMSL

xi__shun
Beginner
417 Views

Hello,

I am wondering if FFT subroutine provided in IMSL c_fast_dft could take double-precision complex variable? It is not mentioned in its documentation. My finding is it does not take double-precision complex variables since it is based on Fortran 77 which does not supply double-precision complex variables. Also, I am wondering if d_fast_dft and s_fast_dft subroutines are also available in IMSL, according to the documentation

"

FORTRAN 90 Interface
Generic: None
Specific: The specific interface names are S_FAST_DFTD_FAST_DFTC_FAST_DFT, and Z_FAST_DFT.

 

"

thanks

0 Kudos
2 Replies
mecej4
Honored Contributor III
417 Views

The usual naming convention in Lapack and other related libraries is that the 'c' prefix is for single-precision complex and 'z' for double-precision complex. Thus, you should try z_fast_dft().

As the IMSL Math library manual says, there is no generic interface for the FFT routines, and there are four specific interfaces, covering combinations of real/complex and single/double precision.

0 Kudos
xi__shun
Beginner
417 Views

Hi Mecej4,

mecej4 wrote:

The usual naming convention in Lapack and other related libraries is that the 'c' prefix is for single-precision complex and 'z' for double-precision complex. Thus, you should try z_fast_dft().

As the IMSL Math library manual says, there is no generic interface for the FFT routines, and there are four specific interfaces, covering combinations of real/complex and single/double precision.

thanks for your help! I went back and tried z_fast_dft, and it works. However, calling d_fast_dft or s_fast_dft gives me error LNK2019: unresolved external symbol _D_FAST_DFT. All I need is to do FFT of real signal.

thanks!

 

 

 

0 Kudos
Reply