Double and Float format displaying different results

Question

I thought that the difference between double and float were the precision of the decimals. However, I am getting strange results with using double and float and they are no where close to one another.

The first segment of code is used with the float format producing the correct results:

#include <stdio.h>

#define ABSOLUTE_VALUE(number)      ( ((number) < 0) ? -(number) : (number) )

int main (void)
{
    float number, absNumber;


    printf ("What number do you want to check the absolute value for? : ");
    scanf  ("%f", &number);

    absNumber = ABSOLUTE_VALUE(number);

    printf ("The absolute value of %.2f is %.2f\n", number, absNumber);

    return 0;
}

Output:
What number do you want to check the absolute value for? : -3
The absolute value of -3.00 is 3.00

The second segment of code is used with the double format producing incorrect results:

#include <stdio.h>

#define ABSOLUTE_VALUE(number)      ( ((number) < 0) ? -(number) : (number) )

int main (void)
{
    double number, absNumber;


    printf ("What number do you want to check the absolute value for? : ");
    scanf  ("%d", &number);

    absNumber = ABSOLUTE_VALUE(number);

    printf ("The absolute value of %.2d is %.2d\n", number, absNumber);

    return 0;
}

Output:
What number do you want to check the absolute value for? : -3
The absolute value of -03 is 2147344384

Solution 1

In the second example, %d is used for integers. (d isn't short for double, but for decimal)

scanf  ("%d", &number);

should be

scanf  ("%lf", &number);

The printf is incorrect too, as %fis used for double

printf ("The absolute value of %.2d is %.2d\n", number, absNumber);

Instead, this should be:

printf ("The absolute value of %.2f is %.2f\n", number, absNumber);

---

Notice that the format specifier for double is different for scanf and printf, this C FAQ article has some good explanations.

Solution 2

In your second program, you have used %d format specifier for double which is wrong. And that's why you are getting error. You should use %lf (long float or better called 'double') for printing it. For other format specifier, I think this might help you a link
You can also search different format specifiers on google by typing 'format specifiers'.

Solution 3

Your last printf format string is wrong. Should be.

 printf ("The absolute value of %.2f is %.2f\n", number, absNumber);

and if you enable all compiler warnings (e.g. compile with gcc -Wall -g) you'll have a warning about that. Always enable all warnings in the compiler.

Your scanf are also wrong, should be scanf("%lf", &number); and with all warnings the compiler would have warned you.

BTW, scanf returns the "number of successfully matched items" and you should test its result.

You really need to carefully read the documentation of printf(3) and of scanf(3). (In a linux terminal, you could type man 3 printf to get it).

Notice that arguments are converted to double and the conversion specifier for double-s is %f in printf and %lf in scanf

And you should learn how to use a debugger (like gdb).

Solution 4

Make 2 changes for using double data-type in c :

use %lf in scanf
use %f in printf

//remember : %d is for int
             %f is for float
             %c is for char

View more solutions

Comments

[Basile Starynkevitch]

Not %lf (this is for long double) but %f

[LinhSaysHi]

So %d is used for integers? Oh darn :( I don't know why I thought it was used for decimals as well

[Basile Starynkevitch]

This is more a comment than an answer.

[Yu Hao]

Wrong for scanf, %f is used for float, %lf is used for double. Its format is different from printf. My answer is correct.

[Yu Hao]

@BasileStarynkevitch No, in the format of scanf, %f is used for float, %lf is used for double. Its format is different from printf.

[Jigyasa]

but you can go to the link and can know which format specifier is for which purpose in a more tabulated and easy to read & understandable form.

[Jigyasa]

@Basile telling him this is not for that etc, I personally feel may confuse him. So, I told what I think is better from my perspective

[LinhSaysHi]

Is a debugger something that will be useful to me in the long run as I keep programming?

[Basile Starynkevitch]

Using a debugger is in practice absolutely essential when coding, specially in C or C++ ... You really need to learn it...

[Yu Hao]

No, the format specifier for double is different for scanf and printf, see my answer. It's %f in printf.

[Ed Heal]

@LinhSaysHi - I meant in the sense that the format is incorrect. Should have been more explicit

[Moberg]

%lf is also available for print f, it is the same as %f in this case so it can be used.