https://community.intel.com/t5/Intel-Integrated-Performance/Who-coded-IPP/m-p/930491#M16567 <DIV>Hi,</DIV> <DIV></DIV> <DIV>You are right, we have good implementation of FFT in IPP. It was written by hand, usingassembler.</DIV> <DIV></DIV> <DIV>Currently our policy is not to provide IPP sources.</DIV> <DIV></DIV> <DIV>Regards,</DIV> <DIV> Vladimir</DIV> Fri, 29 Oct 2004 16:16:57 GMT Vladimir_Dudnik 2004-10-29T16:16:57Z https://community.intel.com/t5/Intel-Integrated-Performance/Who-coded-IPP/m-p/930490#M16566 <DIV></DIV> <DIV>I would like to know how IPP routines are coded.</DIV> <DIV>I am especially interested in the FFT routine.</DIV> <DIV>The reason is FFT consists of mainly complex multiplications.</DIV> <DIV>My experience with Intel C++ compiler is that it cannot vectorize</DIV> <DIV>complex operations properly.</DIV> <DIV>The performance level of FFT on a 2.8C GHz P4 is about 6GFlops.</DIV> <DIV>Now it is impossible that this CPU reaches that performance without full vectorization.</DIV> <DIV>Since the Intel C++ compiler cannot vectorize this code efficiently, I conclude that IPP FFT is coded in assembly.</DIV> <DIV></DIV> <DIV>Can somebody confirm this?</DIV> <DIV>If yes do we have access to the FFT code?</DIV> <DIV></DIV> <DIV>Thanks.</DIV> Fri, 29 Oct 2004 10:59:56 GMT https://community.intel.com/t5/Intel-Integrated-Performance/Who-coded-IPP/m-p/930490#M16566 ergin 2004-10-29T10:59:56Z https://community.intel.com/t5/Intel-Integrated-Performance/Who-coded-IPP/m-p/930491#M16567 <DIV>Hi,</DIV> <DIV></DIV> <DIV>You are right, we have good implementation of FFT in IPP. It was written by hand, usingassembler.</DIV> <DIV></DIV> <DIV>Currently our policy is not to provide IPP sources.</DIV> <DIV></DIV> <DIV>Regards,</DIV> <DIV> Vladimir</DIV> Fri, 29 Oct 2004 16:16:57 GMT https://community.intel.com/t5/Intel-Integrated-Performance/Who-coded-IPP/m-p/930491#M16567 Vladimir_Dudnik 2004-10-29T16:16:57Z