big float computation confusion with MPIR

[ztdep]

#include <iostream>
#include<iomanip>
using namespace std;

#pragma warning(disable: 4800)
#include "mpirxx.h"
#pragma warning(default: 4800)

int main (int argc, char *argv[])
{   

   mpf_class sum(0.000000000000000000000000000000000000000000000000000000000000000,200);
  mpf_class con(0.0000001000000000000000000000000000000000000000000000000000000000,200);
   for(int i=0;i<10;i++)
      { sum=sum+con;
         cout <<scientific<<setprecision(50)<<sum<<"  "<<endl;
   }

the out put:

9.99999999999999954748111825886258685613938723690808e-08
1.99999999999999990949622365177251737122787744738162e-07
2.99999999999999986424433547765877605684181617107242e-07
3.99999999999999981899244730354503474245575489476323e-07
4.99999999999999977374055912943129342806969361845404e-07
5.99999999999999972848867095531755211368363234214485e-07
6.99999999999999968323678278120381079929757106583565e-07
7.99999999999999963798489460709006948491150978952646e-07
8.99999999999999959273300643297632817052544851321727e-07
9.99999999999999954748111825886258685613938723690808e-07
9.999999999999999547481118258862586856139387236908078193664551e-07

Dear friends:
I use MPIR(http://www.mpir.org/index.html) for the big float computation, but the result is very confusion.since i have use a big precision of 200 bits(it corresponds to about 64 bits precisions), but the results is still not the exact values. could you please give me some help. where is wrong.