<?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: IBM vs Intel fortran compilers in Intel® Fortran Compiler</title>
    <link>https://community.intel.com/t5/Intel-Fortran-Compiler/IBM-vs-Intel-fortran-compilers/m-p/739942#M18</link>
    <description>&lt;DIV style="margin: 0px; height: auto;"&gt;&lt;/DIV&gt;
You leave a lot of dependence on guesswork, including guessing why you don't tell us if you got better agreement when you set options for numerical consistency. I believe both compilers could violate parentheses with the quoted options. Supposing that both systems use fused multiply-add so as occasionally to gain accuracy when multiplication is followed by add/subtract, it won't have the same effects in each case, and that will happen not only at -O3.&lt;BR /&gt;Plenty of expressions allow for different orders of evaluation even when standards compliance options are set, and changing from -O3 wouldn't change that. As you hint, math functions can't be expected to be identical.&lt;BR /&gt;How do you know compilers give different results? Did you use the same MPI in both cases? Can you get identical results from run to run?&lt;BR /&gt;I took some reconsideration to figure out that the Altix system must be Itanium, on the assumption that the advertised Xeon based shared memory systems haven't yet been produced.&lt;BR /&gt;&lt;BR /&gt;</description>
    <pubDate>Fri, 16 Oct 2009 05:59:33 GMT</pubDate>
    <dc:creator>TimP</dc:creator>
    <dc:date>2009-10-16T05:59:33Z</dc:date>
    <item>
      <title>IBM vs Intel fortran compilers</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/IBM-vs-Intel-fortran-compilers/m-p/739941#M17</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;I have a large CFD code written in Fortran, I compiled the code with following options using the &lt;BR /&gt;&lt;BR /&gt;&lt;B&gt;IBM compiler&lt;BR /&gt;-O3 -q64 -qwarn64 -qautodbl=dbl4 -qalign=4k&lt;/B&gt;&lt;BR /&gt;&lt;BR /&gt;and&lt;BR /&gt;&lt;BR /&gt;&lt;B&gt;Intel compiler&lt;BR /&gt;-r8 -align all  -O3&lt;/B&gt;&lt;BR /&gt;&lt;BR /&gt;I did test runs on a simple problem with &lt;B&gt;MPI for both the&lt;/B&gt; binaries above and found that both the compilers are giving different results when I compare the output.&lt;BR /&gt;&lt;BR /&gt;Now the &lt;B&gt;Intel compiled code runs on a SGI altix shared memory machine&lt;/B&gt; as an MPI process and &lt;B&gt;IBM code runs on a cluster of Apple PowerMac G5 computers&lt;/B&gt;.&lt;BR /&gt;Is this inconsistent result because of the -O3 compiling on the both the compilers or the difference in libraries or am I missing something here?&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;Amit</description>
      <pubDate>Fri, 16 Oct 2009 02:12:21 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/IBM-vs-Intel-fortran-compilers/m-p/739941#M17</guid>
      <dc:creator>amit-amritkar</dc:creator>
      <dc:date>2009-10-16T02:12:21Z</dc:date>
    </item>
    <item>
      <title>Re: IBM vs Intel fortran compilers</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/IBM-vs-Intel-fortran-compilers/m-p/739942#M18</link>
      <description>&lt;DIV style="margin: 0px; height: auto;"&gt;&lt;/DIV&gt;
You leave a lot of dependence on guesswork, including guessing why you don't tell us if you got better agreement when you set options for numerical consistency. I believe both compilers could violate parentheses with the quoted options. Supposing that both systems use fused multiply-add so as occasionally to gain accuracy when multiplication is followed by add/subtract, it won't have the same effects in each case, and that will happen not only at -O3.&lt;BR /&gt;Plenty of expressions allow for different orders of evaluation even when standards compliance options are set, and changing from -O3 wouldn't change that. As you hint, math functions can't be expected to be identical.&lt;BR /&gt;How do you know compilers give different results? Did you use the same MPI in both cases? Can you get identical results from run to run?&lt;BR /&gt;I took some reconsideration to figure out that the Altix system must be Itanium, on the assumption that the advertised Xeon based shared memory systems haven't yet been produced.&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Fri, 16 Oct 2009 05:59:33 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/IBM-vs-Intel-fortran-compilers/m-p/739942#M18</guid>
      <dc:creator>TimP</dc:creator>
      <dc:date>2009-10-16T05:59:33Z</dc:date>
    </item>
    <item>
      <title>Re: IBM vs Intel fortran compilers</title>
      <link>https://community.intel.com/t5/Intel-Fortran-Compiler/IBM-vs-Intel-fortran-compilers/m-p/739943#M19</link>
      <description>&lt;DIV style="margin:0px;"&gt;
&lt;DIV id="quote_reply" style="width: 100%; margin-top: 5px;"&gt;
&lt;DIV style="margin-left:2px;margin-right:2px;"&gt;Quoting - &lt;A href="https://community.intel.com/en-us/profile/367365"&gt;tim18&lt;/A&gt;&lt;/DIV&gt;
&lt;DIV style="background-color:#E5E5E5; padding:5px;border: 1px; border-style: inset;margin-left:2px;margin-right:2px;"&gt;&lt;EM&gt; You leave a lot of dependence on guesswork, including guessing why you don't tell us if you got better agreement when you set options for numerical consistency. I believe both compilers could violate parentheses with the quoted options. Supposing that both systems use fused multiply-add so as occasionally to gain accuracy when multiplication is followed by add/subtract, it won't have the same effects in each case, and that will happen not only at -O3.&lt;BR /&gt;Plenty of expressions allow for different orders of evaluation even when standards compliance options are set, and changing from -O3 wouldn't change that. As you hint, math functions can't be expected to be identical.&lt;BR /&gt;How do you know compilers give different results? Did you use the same MPI in both cases? Can you get identical results from run to run?&lt;BR /&gt;I took some reconsideration to figure out that the Altix system must be Itanium, on the assumption that the advertised Xeon based shared memory systems haven't yet been produced.&lt;BR /&gt;&lt;BR /&gt;&lt;/EM&gt;&lt;/DIV&gt;
&lt;/DIV&gt;
&lt;/DIV&gt;
&lt;BR /&gt;Tim,&lt;BR /&gt;&lt;BR /&gt;Thanks for your reply.&lt;BR /&gt;I figured it out that I had something else wrong and I need to fix it before I can proceed any further with this line of enquiry.&lt;BR /&gt;&lt;BR /&gt;Thanks again,&lt;BR /&gt;Amit&lt;BR /&gt;Amit&lt;BR /&gt;</description>
      <pubDate>Mon, 19 Oct 2009 14:48:19 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Fortran-Compiler/IBM-vs-Intel-fortran-compilers/m-p/739943#M19</guid>
      <dc:creator>amit-amritkar</dc:creator>
      <dc:date>2009-10-19T14:48:19Z</dc:date>
    </item>
  </channel>
</rss>

