CODEWITHSUNDEEP
javaroundingcoin-change
Dave555
Dave555

Reputation: 167

Rounding Error?

Hi I'm having an issue while creating a change calculator for a school assignment. It essentially is to calculate the lowest amount of change needed for a certain amount of money.

I've stated all of my variable to be doubles so I can calculate the values and the totals together. It seems to work fine on whole numbers (5.00, 6.00, 7.00) but messes up whenever I add a decimal place. Like when I say $5.25 it should say 2 toonies 1 loonie and 1 quarter. I think it could be an error in rounding or something wrong with my calculations. Any Help is appreciated. Here is the calculations of the code:

//Rounding to one number           
DecimalFormat oneDigit = new DecimalFormat ("#,0");

//Ask user for input
String moneyinput = JOptionPane.showInputDialog ("Welcome to the Change Caluculator. "
                + "Please Enter your amount of money in Dollars ($): ");

//Take user input and create into string.
totmoney = Double.parseDouble (moneyinput);

//Calculate number of toonies
numtoonies =  (totmoney/toonieval);

System.out.println ("There is a total of " + oneDigit.format (numtoonies) + " toonies.");

//Find new amount
totmoney = (totmoney%toonieval);

//Calculate the number of loonies
numloonies = (totmoney/loonieval);

//Find new amount
totmoney = (totmoney-numloonies);

System.out.println ("There is a total of " + oneDigit.format (numloonies) + " loonies.");

//Calculate number of quarters
numquarters = (totmoney/quarterval);

//State the about of Coins
System.out.println ("There is a total of " + oneDigit.format (numquarters) + " quarters.");
}

Upvotes: 1

Views: 313

Answers (3)

marme
marme

Reputation: 496

Do not use floating point numbers when you need absolute accuracy. The IEEE floating point specification which java follows even states that there will be a loss of precision because the floating point spec can not properly represent all floating point operations it can only approximate them. The solution for this is that you need to store the value in 2 ints, one for the amount left of the decimal and one for the value to the right of the decimal

Upvotes: 1

twain249
twain249

Reputation: 5706

I don't quite understand why you are using the DecimalFormat at all. You should be able to do solve this with only mod % and division /.

This is how I would approach this problem:

  1. Take the input from the user (as a double in the format 5.33)
  2. Save that as int by moving the decimal place (int value = cost * 100)
  3. Find the number of toonies by using division (numToonies = value / toonieval)
  4. Find the remaining amount of money by (value = value % toonieval)
  5. Repeat steps 3 and 4 for the other denominations

Note: you will have to modify the values of the toonies to reflect the fact that the price you used was multiplied by 100.

Upvotes: 3

Attila
Attila

Reputation: 28762

I don't think this is working properly for floating point numbers:

totmoney = (totmoney%toonieval);

Try

totmoney = totmoney - toonieval*numtoonies;

instead.

Also, be aware that floating point values are generally not suitable for handling monetary values (because of the possiblity of rounding errors). Use fixed point decimals instead (basically, store the value in cents and calculate everyting on a cents base).

Upvotes: 1

Related Questions