<?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: IPP Image Processing performance in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Image-Processing-performance/m-p/909877#M14012</link>
    <description>&lt;P&gt;Hi Chris,&lt;/P&gt;
&lt;P&gt;yes, that's correct. Only difference between CRT malloc/new and ippMalloc is alignment. So, if you provide the same alignment there should not be difference in performance.&lt;/P&gt;
&lt;P&gt;Regards,&lt;BR /&gt; Vladimir&lt;/P&gt;</description>
    <pubDate>Thu, 20 Sep 2007 15:43:38 GMT</pubDate>
    <dc:creator>Vladimir_Dudnik</dc:creator>
    <dc:date>2007-09-20T15:43:38Z</dc:date>
    <item>
      <title>IPP Image Processing performance</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Image-Processing-performance/m-p/909874#M14009</link>
      <description>Dear Image Processors:&lt;BR /&gt;&lt;BR /&gt;Will the performance of the ippi_* functions be significantly different if the image buffers are allocated with "new" rather than with ippiMalloc_* ?&lt;BR /&gt;&lt;BR /&gt;I had assumed NO, but maybe (hopefully) I am wrong.&lt;BR /&gt;&lt;BR /&gt;Please let me know&lt;BR /&gt;Chris&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Thu, 13 Sep 2007 21:42:42 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Image-Processing-performance/m-p/909874#M14009</guid>
      <dc:creator>chris_griffin</dc:creator>
      <dc:date>2007-09-13T21:42:42Z</dc:date>
    </item>
    <item>
      <title>Re: IPP Image Processing performance</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Image-Processing-performance/m-p/909875#M14010</link>
      <description>&lt;P&gt;Hi Chris,&lt;/P&gt;
&lt;P&gt;sometime it does, because of the modern processors can access data faster if particular alignment conditions are met.&lt;/P&gt;
&lt;P&gt;And this isthe reason why we indtroducememory allocationfunctions (which provide you aligned memory block)in IPP.&lt;/P&gt;
&lt;P&gt;Regards,&lt;BR /&gt; Vladimir&lt;/P&gt;</description>
      <pubDate>Wed, 19 Sep 2007 10:18:21 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Image-Processing-performance/m-p/909875#M14010</guid>
      <dc:creator>Vladimir_Dudnik</dc:creator>
      <dc:date>2007-09-19T10:18:21Z</dc:date>
    </item>
    <item>
      <title>Re: IPP Image Processing performance</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Image-Processing-performance/m-p/909876#M14011</link>
      <description>Hi Vladimir,&lt;BR /&gt;&lt;BR /&gt;Thank you for your reply. I have written my image class to always provide an image pitch which is a multiple of 32 bytes. In this case, I would expect the behavior of the library to be equivalent whether I call IPP's allocators or "new"&lt;BR /&gt;&lt;BR /&gt;Let me know if I'm off base here.&lt;BR /&gt;&lt;BR /&gt;Thank you,&lt;BR /&gt;Chris Griffin&lt;BR /&gt;</description>
      <pubDate>Thu, 20 Sep 2007 13:18:27 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Image-Processing-performance/m-p/909876#M14011</guid>
      <dc:creator>chris_griffin</dc:creator>
      <dc:date>2007-09-20T13:18:27Z</dc:date>
    </item>
    <item>
      <title>Re: IPP Image Processing performance</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Image-Processing-performance/m-p/909877#M14012</link>
      <description>&lt;P&gt;Hi Chris,&lt;/P&gt;
&lt;P&gt;yes, that's correct. Only difference between CRT malloc/new and ippMalloc is alignment. So, if you provide the same alignment there should not be difference in performance.&lt;/P&gt;
&lt;P&gt;Regards,&lt;BR /&gt; Vladimir&lt;/P&gt;</description>
      <pubDate>Thu, 20 Sep 2007 15:43:38 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPP-Image-Processing-performance/m-p/909877#M14012</guid>
      <dc:creator>Vladimir_Dudnik</dc:creator>
      <dc:date>2007-09-20T15:43:38Z</dc:date>
    </item>
  </channel>
</rss>

