Everyone whose taxable income is not less than a certain value m, has to pay income tax on that taxable income at x% rate. But this often creates a hazard if proper incentives are not given. For example lets say a person has to pay tax at the rate of 10% of taxable income if his taxable income is not less than 150000 BDT (Bangladeshi Taka). Now consider two persons A and B whose taxable incomes are 145000 BDT and 155000 BDT respectively. Now it is obvious that A do not have to pay tax. But B pays 155000*.10 = 15500 BDT as tax. So after paying the tax the amount that remains in his hand is 155000-15500=139500 BDT. So even though B earns more than A, his effective earning is less than A. Given the value of m and x, you will have to find the value of the maximum income v, which is effectively (after deducting the tax) less than someone earning less than v. You must assume that income of anyone is always an integer and always positive. However, income tax is a real number and so is the effective income. Note: In this problem we are assuming that a person pays tax on his whole taxable income. In real life the rules are a bit different and so such hazard do not occur. Input The input file contains at most 20000 lines of inputs. Each line contains two decimal integers m (0<m<1000000001)andx(0≤x≤100). Themeaningofmandxisgivenabove. Input is terminated by a set where the values of m and x are zero. This set should not be processed. Output For each line of input produce one line of output. This line contains a positive integer which denotes the value of v, whose meaning is given in the problem statement. If such a value of v is not found then print ‘Not found’ instead. Sample Input 20 10 2300 4 00 Sample Output 21 2394