https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/C-C-speed-optimization-bible-resources-pointers-needed/m-p/915094#M12494 Unless you can profile and show the code you want to speed up, your best bet is to narrow down your question by Google etc. and see if any MKL functions are applicable. You make it sound contradictory; you have super-fast code which isn't super-fast, and you don't even want to use multi-core. Performance can't be a "huge concern" if ability to run on a 5 year old CPU is more important. Yes, I have 3 of those. Intel compilers are capable of SSE3 vectorization of complex code, which of course requires a CPU less than 3 years old.<BR /> Fri, 27 Jul 2007 12:42:00 GMT TimP 2007-07-27T12:42:00Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/C-C-speed-optimization-bible-resources-pointers-needed/m-p/915093#M12493 C/C++ speed optimization bible/resources/pointers needed!<BR /><BR />Hi all,<BR /><BR />I am in the middle of programming to solve an engineering problem<BR />where the speed is huge concern. The project involving lots of<BR />numerical integration and then there are several loops/levels of<BR />optimization on top of the function evaluation engine. As you probably<BR />know, the key to a successful optimization is a fast underlying<BR />objective function evaluator. The faster it is, the more promising the<BR />optimization result(perhaps global optimal). However our project<BR />requires many numerical integrations which prohibits us from making it<BR />super fast. At the heart of the numerical integration is a smart<BR />integrator and a super-fast integrand function evaluator. Even worse,<BR />our function evaluation is in complex-domain. So the kay point is how<BR />to arrange our C/C++ code to make it highly efficient in every aspect.<BR />Could anybody give some advice/pointers on how to improve the speed of<BR />C/C++ program? How to arrange code? How to make it highly efficient<BR />and super fast? What options do I have if I don't have luxury to use<BR />multi-threaded, multi-core or distributed computing? But I do have a<BR />P4 at least. Please recommend some good bibles and resources! What can Intel's libraries do for me specifically for my problems? Thank<BR />you!<BR /><BR /><BR /> Fri, 27 Jul 2007 05:09:11 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/C-C-speed-optimization-bible-resources-pointers-needed/m-p/915093#M12493 losemind 2007-07-27T05:09:11Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/C-C-speed-optimization-bible-resources-pointers-needed/m-p/915094#M12494 Unless you can profile and show the code you want to speed up, your best bet is to narrow down your question by Google etc. and see if any MKL functions are applicable. You make it sound contradictory; you have super-fast code which isn't super-fast, and you don't even want to use multi-core. Performance can't be a "huge concern" if ability to run on a 5 year old CPU is more important. Yes, I have 3 of those. Intel compilers are capable of SSE3 vectorization of complex code, which of course requires a CPU less than 3 years old.<BR /> Fri, 27 Jul 2007 12:42:00 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/C-C-speed-optimization-bible-resources-pointers-needed/m-p/915094#M12494 TimP 2007-07-27T12:42:00Z