July 2004 - Posts

• Method for getting /Op like consistency in MS C++ 14.0

  (From Microsoft Visual C++ Floating-Point Optimization ) Many C++ compilers offer a "consistency" floating-point model (through a /Op or /fltconsistency switch) that enables a developer to create programs compliant with strict floating-point semantics. Read More... Posted Wednesday, July 28, 2004 4:53 PM by ericflee | 1 Comments

• Method for retaining intermediate results in extended precision

  I thought I’d post a brief entry on how to retain intermediate results in extended precision. Consider the following direct summation algorithm: double sum(double[] a, int c) { 1: double result = 0.0; 2: for (int i=0; i<c; i++) 3: result = result Read More... Posted Wednesday, July 21, 2004 5:30 PM by ericflee | 1 Comments

• Overloading the semantics of the register keyword (followup)

  John Albert asks two questions: What would "register float" mean? How would this work under fp:fast? The basic guideline is that "intermediate expressions will be computed in as high as precision as is practical". For instance, suppose for a particular Read More... Posted Friday, July 02, 2004 6:24 PM by ericflee | 0 Comments

• No inline assembly on AMD64

  Igor Abramov wrote: “Well, I was in desperate need for 80-bit long doubles. Now I coded this type as set of asm inlines and application prototype shipped in such form. But the production version will be done for amd64, in 64-bit mode for performance Read More... Posted Friday, July 02, 2004 6:18 PM by ericflee | 2 Comments

• Confused about VC++ version numbers? Me too...

  So I may have been confusing people when I talk about VC8.0. When I refer to VC8.0, I'm referring to Visual C++ in Visual Studio 2005. The product version is 8.0, but the compiler itself (i.e. cl.exe) is “version 14“. I'll try to be more consistent Read More... Posted Friday, July 02, 2004 6:06 PM by ericflee | 0 Comments