<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: 2D Convolution via FFT gives all 0 in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/2D-Convolution-via-FFT-gives-all-0/m-p/941252#M14529</link>
    <description>&lt;DIV&gt;&lt;/DIV&gt;
&lt;P&gt;Thanks a lot for replying,ylatkin.&lt;/P&gt;
&lt;P&gt;Then I guess I'll need to go throughFFT - Multiplication - FFT myself. I know 1D convolution works with FFT mode. It's a very nice feature to have, giving MKL a sharp edge on winning anyone who's doing signal processing stuffs.&lt;/P&gt;
&lt;DIV&gt;&lt;/DIV&gt;</description>
    <pubDate>Fri, 10 Mar 2006 23:31:40 GMT</pubDate>
    <dc:creator>ttf</dc:creator>
    <dc:date>2006-03-10T23:31:40Z</dc:date>
    <item>
      <title>2D Convolution via FFT gives all 0</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/2D-Convolution-via-FFT-gives-all-0/m-p/941250#M14527</link>
      <description>&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;FONT size="2"&gt;
&lt;P&gt;Hi, I'm trying to use MKL to facilitate multidimensional convolution. For a same piece of code, 2D convolution works with direct mode but gives all zero with FFT mode. Is there any examples or tips on using this feature? Many thanks.&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;I was able to do a 2D convolution by&lt;/P&gt;
&lt;P&gt;... code fragment ...&lt;/P&gt;
&lt;P&gt;int mode = VSL_CONV_MODE_DIRECT;&lt;/P&gt;
&lt;P&gt;vslsConvNewTask( &amp;amp;task, mode, dim, xShape, yShape, zShape );&lt;/P&gt;
&lt;P&gt;status=vslsConvExec( task, signal, xStride, kernel, yStride, result, zStride );&lt;/P&gt;
&lt;P&gt;......&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;However, if the mode is set to VSL_CONV_MODE_FFT; the result is all zero. And the task execution returns with error: status = -2000&lt;/P&gt;&lt;/FONT&gt;&lt;/DIV&gt;&lt;P&gt;Message Edited by ttf on &lt;SPAN class="date_text"&gt;03-09-2006&lt;/SPAN&gt; &lt;SPAN class="time_text"&gt;02:08 PM&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Fri, 10 Mar 2006 06:08:02 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/2D-Convolution-via-FFT-gives-all-0/m-p/941250#M14527</guid>
      <dc:creator>ttf</dc:creator>
      <dc:date>2006-03-10T06:08:02Z</dc:date>
    </item>
    <item>
      <title>Re: 2D Convolution via FFT gives all 0</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/2D-Convolution-via-FFT-gives-all-0/m-p/941251#M14528</link>
      <description>&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;Unfortunately, multi-dimensional Fourier method is not implemented.We work on adding this functionality to future MKL releases.&lt;/DIV&gt;</description>
      <pubDate>Fri, 10 Mar 2006 15:42:25 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/2D-Convolution-via-FFT-gives-all-0/m-p/941251#M14528</guid>
      <dc:creator>Yevgeny_Latkin__Inte</dc:creator>
      <dc:date>2006-03-10T15:42:25Z</dc:date>
    </item>
    <item>
      <title>Re: 2D Convolution via FFT gives all 0</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/2D-Convolution-via-FFT-gives-all-0/m-p/941252#M14529</link>
      <description>&lt;DIV&gt;&lt;/DIV&gt;
&lt;P&gt;Thanks a lot for replying,ylatkin.&lt;/P&gt;
&lt;P&gt;Then I guess I'll need to go throughFFT - Multiplication - FFT myself. I know 1D convolution works with FFT mode. It's a very nice feature to have, giving MKL a sharp edge on winning anyone who's doing signal processing stuffs.&lt;/P&gt;
&lt;DIV&gt;&lt;/DIV&gt;</description>
      <pubDate>Fri, 10 Mar 2006 23:31:40 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/2D-Convolution-via-FFT-gives-all-0/m-p/941252#M14529</guid>
      <dc:creator>ttf</dc:creator>
      <dc:date>2006-03-10T23:31:40Z</dc:date>
    </item>
  </channel>
</rss>

