以下是是程序不对，还是由于精度根本比不出来？这是为什么呢？

Question

就是比较一个int变量和double变量的问题，我的程序如下：#include <iostream>using namespace std;int main( void ){int i_var1;double d_var2;cout<<"Please input two numbers( The 1st is int, the 2nd is double )"<<endl;cin>>i_var1>>d_var2;double d_var_temp = (double)i_var1;if ( ( d_var2 - d_var_temp ) == 0.0 ){cout<<"double number = int number"<<endl;}else if ( ( d_var2 - d_var_temp ) >= 1e-06 ){cout<<"double number > int number"<<endl;}else{cout<<"double number < int number"<<endl;}return 0;}测试了几组数据，发现i_var1 = 1，d_var2 = 1.000001的结果不对，应该是d_var2大，但输出是‘double number < int number’，是怎么回事啊？是程序不对，还是由于精度根本比不出来？

Answer 1

double是16位精度
把1e-06 改成 1e-15，扩大应用范围
另外对于浮点数不要用 == 来比较
判断a,b相等，用
if(fabs(a-b)<1e-15) printf("a==b");

Answer 2

问题在else if ( ( d_var2 - d_var_temp ) >= 1e-06 )
1.000001 输入后，跟踪一下，就知道是1.0000009999999999,
当然就是double number < int number
避免：

#include "stdafx.h"
#include <iostream>
#include <math.h>
using namespace std;
#define MIN_EQ_THHOLD 1e-12
int main( void )
{
double c = 0;
c++;
int i_var1 = 0;
double d_var2 = 0;

cout<<"Please input two numbers( The 1st is int, the 2nd is double )"<<endl;

cin>>i_var1>>d_var2;

double d_var_temp = (double)i_var1;

if ( fabs( d_var2 - d_var_temp ) <= MIN_EQ_THHOLD)
{
cout<<"double number = int number"<<endl;
}
else if ( d_var2 > d_var_temp )
{
cout<<"double number > int number"<<endl;
}
else
{
cout<<"double number < int number"<<endl;
}

return 0;
}