https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-10-2-FFT/m-p/860217#M7430 Hi,<BR />In MKL 10.1, FFT is not threaded for power-of-two data.<BR />How about MKL 10.2? In User's Guide, any description of data size is not written.<BR /><BR />--<BR />Fujii Thu, 09 Jul 2009 03:47:18 GMT ks-fujii 2009-07-09T03:47:18Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-10-2-FFT/m-p/860217#M7430 Hi,<BR />In MKL 10.1, FFT is not threaded for power-of-two data.<BR />How about MKL 10.2? In User's Guide, any description of data size is not written.<BR /><BR />--<BR />Fujii Thu, 09 Jul 2009 03:47:18 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-10-2-FFT/m-p/860217#M7430 ks-fujii 2009-07-09T03:47:18Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-10-2-FFT/m-p/860218#M7431 <DIV style="margin:0px;"></DIV> <BR /><BR />In MKL 10.2, FFT performance has been improved by up to 40% for power of 2 data for sizes up to 64 for both single precison and double precision and are threaded.<BR /><BR />--Vipin<BR /><BR /><BR /> Thu, 09 Jul 2009 06:48:24 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/MKL-10-2-FFT/m-p/860218#M7431 VipinKumar_E_Intel 2009-07-09T06:48:24Z