topic Re: floating-point underflow occurs in numerical computing on Windows, but not on Linux in Intel® oneAPI DPC++/C++ Compiler https://community.intel.com/t5/Intel-oneAPI-DPC-C-Compiler/floating-point-underflow-occurs-in-numerical-computing-on/m-p/1717466#M4578 <P>Hi, could you provide a minimal sample code to reproduce the issue ?</P> Wed, 17 Sep 2025 16:36:34 GMT yzh_intel 2025-09-17T16:36:34Z floating-point underflow occurs in numerical computing on Windows, but not on Linux https://community.intel.com/t5/Intel-oneAPI-DPC-C-Compiler/floating-point-underflow-occurs-in-numerical-computing-on/m-p/1715381#M4543 <P>The latest Intel compiler is used to build a numerical code the code throws out an exception of&nbsp;floating-point underflow. The location shows a multiplication of two variables of 1.0e-40 triggers exception in Visual Studio on Windows. Tried to set&nbsp;</P><P>/fp:strict /Qftz-</P><P>and no help. How to solve the issue?</P><P>&nbsp;</P><P>The same code does not have any issues on Linux.</P> Fri, 05 Sep 2025 17:54:42 GMT https://community.intel.com/t5/Intel-oneAPI-DPC-C-Compiler/floating-point-underflow-occurs-in-numerical-computing-on/m-p/1715381#M4543 newcfd 2025-09-05T17:54:42Z Re: floating-point underflow occurs in numerical computing on Windows, but not on Linux https://community.intel.com/t5/Intel-oneAPI-DPC-C-Compiler/floating-point-underflow-occurs-in-numerical-computing-on/m-p/1717466#M4578 <P>Hi, could you provide a minimal sample code to reproduce the issue ?</P> Wed, 17 Sep 2025 16:36:34 GMT https://community.intel.com/t5/Intel-oneAPI-DPC-C-Compiler/floating-point-underflow-occurs-in-numerical-computing-on/m-p/1717466#M4578 yzh_intel 2025-09-17T16:36:34Z