Floating-point types

Learn float, double and long double types.

Floating-point types

C++ has three floating-point types: float (single precision, ~7 digits), double (double precision, ~15 digits) and long double (extended, ~18+ digits).

When writing a float value, use the f suffix: float temp = 36.5f;. Without the suffix, 36.5 is a double.

For most calculations double is recommended; precision loss is small and there is no performance penalty on modern CPUs. Choose float when memory savings matter.

float and double precision

float holds about 7 significant digits; double about 15. So printing float x = 3.14159265f; may lose some digits.

In scientific and financial calculations, small errors accumulate; that is why double is the safe choice. Do not trust floating-point equality: 0.1 + 0.2 may not exactly equal 0.3.

You can control the number of printed digits with std::setprecision (from <iomanip>); you will see it at an advanced level.

Exercise Print float and double

Inside main, declare a float (value 36.5f) and a double (value 3.14159265). Print both on separate lines with std::cout using std::endl.

Show example solution

#include <iostream>

int main() {
    // Declare float and double
    float temperature = 36.5f;
    double precise = 3.14159265;
    // Print each with std::cout
    std::cout << temperature << std::endl;
    std::cout << precise << std::endl;
    return 0;
}

Test yourself

Double-precision variable

____ precise = 3.14159265;

Show answer

double precise = 3.14159265;

Which is the highest-precision floating type?
- int
- float
- double
- long double
Show answer

long double — long double offers the widest precision.
Match floating types

float

~7 digits
double

~15 digits
long double

~18+ digits
f suffix

float marker

Next lesson: bool and char

We Are Enhancing Your Experience

Cookie Preferences

Strictly Necessary Cookies

Analytical Cookies