<?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 Clarification oneDPL RNG examples in Intel® oneAPI DPC++/C++ Compiler</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-DPC-C-Compiler/Clarification-oneDPL-RNG-examples/m-p/1577060#M3525</link>
    <description>&lt;P&gt;Hi there,&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;I am currently trying to understand the use of oneDPL RNGs. There are a range of examples using this API on the internet. They all follow the same structure: within the parallel_for, we create an engine, a distribution and generate the random number as distribution(engine) for further use. See example at &lt;A href="https://oneapi-src.github.io/oneDPL/api_for_sycl_kernels/random.html" target="_blank"&gt;https://oneapi-src.github.io/oneDPL/api_for_sycl_kernels/random.html&lt;/A&gt; . Without SYCL/oneDPL one would use C++ std such that the engine and distribution are created outside of the for loop, and putting these into the loop come with a substantial performance hit. Could you clarify, maybe also in the documentation of the examples, how and why the placement and performance differs for oneDPL RNGs compared to C++ stdlib.&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Thanks,&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Fri, 01 Mar 2024 16:01:48 GMT</pubDate>
    <dc:creator>cbyrohl</dc:creator>
    <dc:date>2024-03-01T16:01:48Z</dc:date>
    <item>
      <title>Clarification oneDPL RNG examples</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-DPC-C-Compiler/Clarification-oneDPL-RNG-examples/m-p/1577060#M3525</link>
      <description>&lt;P&gt;Hi there,&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;I am currently trying to understand the use of oneDPL RNGs. There are a range of examples using this API on the internet. They all follow the same structure: within the parallel_for, we create an engine, a distribution and generate the random number as distribution(engine) for further use. See example at &lt;A href="https://oneapi-src.github.io/oneDPL/api_for_sycl_kernels/random.html" target="_blank"&gt;https://oneapi-src.github.io/oneDPL/api_for_sycl_kernels/random.html&lt;/A&gt; . Without SYCL/oneDPL one would use C++ std such that the engine and distribution are created outside of the for loop, and putting these into the loop come with a substantial performance hit. Could you clarify, maybe also in the documentation of the examples, how and why the placement and performance differs for oneDPL RNGs compared to C++ stdlib.&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Thanks,&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Fri, 01 Mar 2024 16:01:48 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-DPC-C-Compiler/Clarification-oneDPL-RNG-examples/m-p/1577060#M3525</guid>
      <dc:creator>cbyrohl</dc:creator>
      <dc:date>2024-03-01T16:01:48Z</dc:date>
    </item>
  </channel>
</rss>

