Urgent: "atof" function cannot outcome precise result

Hot Topic

Capturing Screen
Program termination conditions
Advantages to Using C++.Net 2005?
An STL Problem: Reading Unicode File in VC++ 2005
Autoscroll
Set VC++ Directories in Options dialog (Visual C++ Express)
aspell compilation problems
ON_WM_NCHITTEST (and other MFC macros) won't compile in whidbey
build error
debugging: where are the variables?

Visual C++

NMAKE : fatal error U1052: file date.mak not found
disable manifest stuff?
MessageBOX
VC 2005 RC with Loki small object allocator, std::lower_bound, and CompareFixedAllocatorSize
What happened to the C++ language profiler?
VS2005 IDE Bug with static lib dependency
Why does adding a MessageBox make so much difference?
ultra fast binary writes
Thread-Safe increment
cfiledialog

Urgent: "atof" function cannot outcome precise result

The following test code is compiled and debuged with VS.NET2003 IDE, the calculation is always wrong. The real calculation in my debugging process are

d1: 21.329999999999998

d2: 191.96999999999997

n: 19196

The expected correct is supposed to be

d1: 21.33

d2: 191.97

n: 19197

The test code is from my project algorithm code which outcome is seriously complained from QA team. I urgently need help to solve it.

Thanks


The test code is:

#include <math.h>

int main(int argc, char* argv[])
{
    char str[] = "21.33";

    double d1 = atof(str);

    double d2 = d1*9.0;

    long n = (long)(d2*100.0);

    return 0;
}


 



Answer this question

Urgent: "atof" function cannot outcome precise result

  • Marlon Chen

    Hi,

    Take a look at the following article:

    http://docs.sun.com/source/806-3568/ncg_goldberg.html

    Ronald Laeremans
    Visual C++ team

     

  • Dmitry Goncharenko - MSFT

    I read a double type value from GUI like 4.56 and when check it from debugger it became 4.5599999999999996 and such difference always cause my following calculation uncorrect result.

    Urgent help is appreicated!

    XZH
  • Urgent: "atof" function cannot outcome precise result
Answer this question