https://community.intel.com/t5/Intel-Integrated-Performance/matrix-vector-performance/m-p/903521#M13117 <P>Could you please also be so kind to specify what version of IPP do you use and on which platform do you run your code? Piece of code of course will simplify us investigation of that issue.</P> <P>Regards,<BR /> Vladimir</P> Mon, 06 Nov 2006 22:36:40 GMT Vladimir_Dudnik 2006-11-06T22:36:40Z https://community.intel.com/t5/Intel-Integrated-Performance/matrix-vector-performance/m-p/903520#M13116 <P><FONT face="Arial" size="2">I use ippmMul_mva_32f function to multiply 3x3 matrix by array of vectors, - rotation of 3D points. However it turned out, that this function works slower then old simpleclassic code with nine multiplies.</FONT></P> <P><FONT face="Arial" size="2">Forarray with 35000 elements, old code took 700usecs, but ippmMul_mva_32f takes about 1000 usecs! May be I do something wrong? I can postsample code, if necessary.</FONT></P> <P><FONT face="Arial" size="2">Thanks.</FONT></P> Thu, 26 Oct 2006 22:02:14 GMT https://community.intel.com/t5/Intel-Integrated-Performance/matrix-vector-performance/m-p/903520#M13116 nikolay_bit 2006-10-26T22:02:14Z https://community.intel.com/t5/Intel-Integrated-Performance/matrix-vector-performance/m-p/903521#M13117 <P>Could you please also be so kind to specify what version of IPP do you use and on which platform do you run your code? Piece of code of course will simplify us investigation of that issue.</P> <P>Regards,<BR /> Vladimir</P> Mon, 06 Nov 2006 22:36:40 GMT https://community.intel.com/t5/Intel-Integrated-Performance/matrix-vector-performance/m-p/903521#M13117 Vladimir_Dudnik 2006-11-06T22:36:40Z