<?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>i.MX ProcessorsのトピックRe: AI/ML Performance Data with GPU and NPU</title>
    <link>https://community.nxp.com/t5/i-MX-Processors/AI-ML-Performance-Data-with-GPU-and-NPU/m-p/1518490#M194870</link>
    <description>&lt;P&gt;Hello Arrow,&lt;/P&gt;
&lt;P&gt;In case of Antutu V9 has a hard request on sampleRateShading feature(Vulkan 1.2 extension), &amp;nbsp;cannot run GPU test without this extension.&lt;/P&gt;
&lt;P&gt;This is not well compatible with Khronos standard specification, which allow the extension feature absence by GPU vendor support.&lt;/P&gt;
&lt;P&gt;&lt;A href="https://www.khronos.org/registry/vulkan/specs/1.2-extensions/man/html/VkPhysicalDeviceFeatures.html" target="_blank" rel="nofollow noopener noreferrer"&gt;https://www.khronos.org/registry/vulkan/specs/1.2-extensions/man/html/VkPhysicalDeviceFeatures.html&lt;/A&gt;&lt;/P&gt;
&lt;UL&gt;
&lt;LI&gt;sampleRateShading&amp;nbsp;specifies whether&amp;nbsp;&lt;A href="https://www.khronos.org/registry/vulkan/specs/1.2-extensions/html/vkspec.html#primsrast-sampleshading" target="_blank" rel="nofollow noopener noreferrer"&gt;Sample Shading&lt;/A&gt;&amp;nbsp;and multisample interpolation are supported. If this feature is not enabled, the&amp;nbsp;sampleShadingEnable&amp;nbsp;member of the&amp;nbsp;&lt;A href="https://www.khronos.org/registry/vulkan/specs/1.2-extensions/man/html/VkPipelineMultisampleStateCreateInfo.html" target="_blank" rel="nofollow noopener noreferrer"&gt;VkPipelineMultisampleStateCreateInfo&lt;/A&gt;&amp;nbsp;structure&amp;nbsp;&lt;STRONG&gt;must&lt;/STRONG&gt;&amp;nbsp;be set to&amp;nbsp;VK_FALSE&amp;nbsp;and the&amp;nbsp;minSampleShading&amp;nbsp;member is ignored. This also specifies whether shader modules&amp;nbsp;&lt;STRONG&gt;can&lt;/STRONG&gt;&amp;nbsp;declare the&amp;nbsp;SampleRateShading&amp;nbsp;capability.&lt;/LI&gt;
&lt;/UL&gt;
&lt;P&gt;In case of Android BSP release recommends Antutu V6 for 8MQ as V8 version requires huge memory.&lt;/P&gt;
&lt;P&gt;For other benchmark I suggest to open a new case in our system due to nda document.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Regards&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;
&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>Wed, 07 Sep 2022 11:40:48 GMT</pubDate>
    <dc:creator>Bio_TICFSL</dc:creator>
    <dc:date>2022-09-07T11:40:48Z</dc:date>
    <item>
      <title>AI/ML Performance Data with GPU and NPU</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/AI-ML-Performance-Data-with-GPU-and-NPU/m-p/1517698#M194797</link>
      <description>&lt;P&gt;Hi&amp;nbsp;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The following illustration shows that both GPUs and NPUs support algorithms by NXP i.MX 8 series.&lt;/SPAN&gt;&lt;BR /&gt;&lt;SPAN&gt;If you have AI/ML performance comparison data between GPU and NPU, please provide it.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;A customer would like to know the performance compared btw GPU and NPU.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="ArrowElectronics_0-1662463856586.png" style="width: 400px;"&gt;&lt;img src="https://community.nxp.com/t5/image/serverpage/image-id/192650i80CF677CF2B13B99/image-size/medium?v=v2&amp;amp;px=400" role="button" title="ArrowElectronics_0-1662463856586.png" alt="ArrowElectronics_0-1662463856586.png" /&gt;&lt;/span&gt;&lt;/P&gt;&lt;P&gt;Thanks.&lt;/P&gt;&lt;P&gt;Regards,&amp;nbsp;&lt;/P&gt;&lt;P&gt;James&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Tue, 06 Sep 2022 11:34:25 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/AI-ML-Performance-Data-with-GPU-and-NPU/m-p/1517698#M194797</guid>
      <dc:creator>JK-IMX</dc:creator>
      <dc:date>2022-09-06T11:34:25Z</dc:date>
    </item>
    <item>
      <title>Re: AI/ML Performance Data with GPU and NPU</title>
      <link>https://community.nxp.com/t5/i-MX-Processors/AI-ML-Performance-Data-with-GPU-and-NPU/m-p/1518490#M194870</link>
      <description>&lt;P&gt;Hello Arrow,&lt;/P&gt;
&lt;P&gt;In case of Antutu V9 has a hard request on sampleRateShading feature(Vulkan 1.2 extension), &amp;nbsp;cannot run GPU test without this extension.&lt;/P&gt;
&lt;P&gt;This is not well compatible with Khronos standard specification, which allow the extension feature absence by GPU vendor support.&lt;/P&gt;
&lt;P&gt;&lt;A href="https://www.khronos.org/registry/vulkan/specs/1.2-extensions/man/html/VkPhysicalDeviceFeatures.html" target="_blank" rel="nofollow noopener noreferrer"&gt;https://www.khronos.org/registry/vulkan/specs/1.2-extensions/man/html/VkPhysicalDeviceFeatures.html&lt;/A&gt;&lt;/P&gt;
&lt;UL&gt;
&lt;LI&gt;sampleRateShading&amp;nbsp;specifies whether&amp;nbsp;&lt;A href="https://www.khronos.org/registry/vulkan/specs/1.2-extensions/html/vkspec.html#primsrast-sampleshading" target="_blank" rel="nofollow noopener noreferrer"&gt;Sample Shading&lt;/A&gt;&amp;nbsp;and multisample interpolation are supported. If this feature is not enabled, the&amp;nbsp;sampleShadingEnable&amp;nbsp;member of the&amp;nbsp;&lt;A href="https://www.khronos.org/registry/vulkan/specs/1.2-extensions/man/html/VkPipelineMultisampleStateCreateInfo.html" target="_blank" rel="nofollow noopener noreferrer"&gt;VkPipelineMultisampleStateCreateInfo&lt;/A&gt;&amp;nbsp;structure&amp;nbsp;&lt;STRONG&gt;must&lt;/STRONG&gt;&amp;nbsp;be set to&amp;nbsp;VK_FALSE&amp;nbsp;and the&amp;nbsp;minSampleShading&amp;nbsp;member is ignored. This also specifies whether shader modules&amp;nbsp;&lt;STRONG&gt;can&lt;/STRONG&gt;&amp;nbsp;declare the&amp;nbsp;SampleRateShading&amp;nbsp;capability.&lt;/LI&gt;
&lt;/UL&gt;
&lt;P&gt;In case of Android BSP release recommends Antutu V6 for 8MQ as V8 version requires huge memory.&lt;/P&gt;
&lt;P&gt;For other benchmark I suggest to open a new case in our system due to nda document.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Regards&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;
&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>Wed, 07 Sep 2022 11:40:48 GMT</pubDate>
      <guid>https://community.nxp.com/t5/i-MX-Processors/AI-ML-Performance-Data-with-GPU-and-NPU/m-p/1518490#M194870</guid>
      <dc:creator>Bio_TICFSL</dc:creator>
      <dc:date>2022-09-07T11:40:48Z</dc:date>
    </item>
  </channel>
</rss>

