Why does value change when converting floating point value to int

Question

I am declaring the amount as a float. When I multiply the value by 100 and plug it into an "int" variable the value changes to 419... why is this happening?

#include <stdio.h>
#include <cs50.h>
#include <math.h>

float amount;
int cents, coins, quarters, dimes, nickels, pennies, Q, D, N, P;
int main(void){

do
{
    amount = get_float("How much we talkin?\n");
}
while (amount < 0);
        printf("cents = %.55f\n", amount);
cents = amount * 100;
        printf("cents = %i\n", cents);
quarters = cents % 25;
    Q = cents / 25;
        printf("quarters = %i\n", quarters);
dimes = quarters % 10;
    D = quarters / 10;
        printf("dimes = %i\n", dimes);
nickels = dimes % 5;
    N = dimes / 5;
        printf("nickels = %i\n", nickels);
pennies = nickels % 1;
    P = nickels / 1;
        printf("pennies = %i\n", pennies);
coins = Q+D+N+P;

    printf("%i\n", coins);

}

~/workspace/pset1/cash/ $ ./cash
How much we talkin?
4.2
cents = 4.1999998092651367187500000000000000000000000000000000000
cents = 419
quarters = 19
dimes = 9
nickels = 4
pennies = 0
22

... that link helped. Thank you!

Include actual code. Not a link to code. And especially *not a link to an image* of code.. — user2864740, Jun 28 '18 at 02:47
You need to post your code *here,* not on some site that could disappear in future. — paxdiablo, Jun 28 '18 at 02:48

score 1 · Answer 1 · 2018-06-28T03:56:31.800

Thank you, I resolved this issue by taking the input from the user as a float, multiplying that value by 100, then rounding to the nearest integer. The code below works, not the cleanest solution in the world.

#include <stdio.h>
#include <cs50.h>
#include <math.h>

float amount;
int cents, coins, quarters, dimes, nickels, pennies, Q, D, N, P;
int main(void){

do
{
    amount = get_float("How much we talkin?\n");
}
while (amount < 0);
    // Print float input to see value is not precise
    printf("amount = %.55f\n", amount);

// Multiply amount by 100, then round to nearest
cents = roundf(amount * 100);
    printf("cents = %i\n", cents);

// *Quarters*
quarters = cents % 25;
    Q = cents / 25;
    printf("quarters = %i\n", Q);

// *Dimes*
dimes = quarters % 10;
    D = quarters / 10;
    printf("dimes = %i\n", D);

// *Nickels*
nickels = dimes % 5;
    N = dimes / 5;
    printf("nickels = %i\n", N);

// *Pennies*
pennies = nickels % 1;
    P = nickels / 1;
    printf("pennies = %i\n", P);

// Add up all the coins
coins = Q+D+N+P;

    printf("%i\n", coins);

}


~/workspace/pset1/cash/ $ ./cash
How much we talkin?
4.2
amount = 4.1999998092651367187500000000000000000000000000000000000
cents = 420
quarters = 16
dimes = 2
nickels = 0
pennies = 0
18

Rather than `cents = roundf(amount * 100);`, consider `cents = lrint(amount * 100.0);` `long lrint( double arg )` available since C99, does the rounding and conversion to an integer in one call. Using a `double 100.0` insures no rounding in the product calculation which may occur with `some_float * 100`. — chux - Reinstate Monica, Jun 28 '18 at 12:49

score 0 · Answer 2 · answered Jun 28 '18 at 07:39

To understand this issue, you need to understand how floating point numbers are stored in memory, in their binary form.

You also need to understand that C's type conversion TRUNCATES a float when converting to an integer.

The number 4.2 is stored as:

1 x 4
0 x 2
0 x 1
0 x 0.5
0 x 0.25
1 x 0.125
1 x 0.0625
0 x 0.03125
0 x 0.015625
1 x 0.0078125
1 x 0.00390625
etc

Which sums to 4.199 (and eventually approximates to 4.199999 - but NEVER 4.2)

Multiplying by 100 gives you 419.999(etc) - and truncating this gives you 419 not 420.

The solution (as you have found) is to make sure you ROUND rather than TRUNC

Why does value change when converting floating point value to int

2 Answers2