Home Original page

std::cbrt, std::cbrtf, std::cbrtl - cppreference.com

From cppreference.com

Defined in header <cmath>

(1)

float cbrt ( float num ); double cbrt ( double num ); long double cbrt ( long double num );

 (until C++23)

/*floating-point-type*/ cbrt ( /*floating-point-type*/ num );

 (since C++23)
(constexpr since C++26)

float cbrtf( float num );

 (2) (since C++11)
(constexpr since C++26)

long double cbrtl( long double num );

 (3) (since C++11)
(constexpr since C++26)

SIMD overload (since C++26)

Defined in header <simd>

template< /*math-floating-point*/ V > constexpr /*deduced-simd-t*/<V> cbrt ( const V& v_num );

 (S) (since C++26)

Additional overloads (since C++11)

Defined in header <cmath>

template< class Integer > double cbrt ( Integer num );

 (A) (constexpr since C++26)

1-3) Computes the cube root of num. The library provides overloads of std::cbrt for all cv-unqualified floating-point types as the type of the parameter.(since C++23)

A) Additional overloads are provided for all integer types, which are treated as double.

 (since C++11)

Parameters

num - floating-point or integer value

Return value

If no errors occur, the cube root of num (3√num), is returned.

If a range error occurs due to underflow, the correct result (after rounding) is returned.

Error handling

Errors are reported as specified in math_errhandling.

If the implementation supports IEEE floating-point arithmetic (IEC 60559),

  • if the argument is ±0 or ±∞, it is returned, unchanged.
  • if the argument is NaN, NaN is returned.

Notes

std::cbrt(num) is not equivalent to std::pow(num, 1.0 / 3) because the rational number

is typically not equal to 1.0 / 3 and std::pow cannot raise a negative base to a fractional exponent. Moreover, std::cbrt(num) usually gives more accurate results than std::pow(num, 1.0 / 3) (see example).

The additional overloads are not required to be provided exactly as (A). They only need to be sufficient to ensure that for their argument num of integer type, std::cbrt(num) has the same effect as std::cbrt(static_cast<double>(num)).

Example

#include <cmath>
#include <iomanip>
#include <iostream>
#include <limits>

int main()
{
    std::cout
        << "Normal use:\n"
        << "cbrt(729)       = " << std::cbrt(729) << '\n'
        << "cbrt(-0.125)    = " << std::cbrt(-0.125) << '\n'
        << "Special values:\n"
        << "cbrt(-0)        = " << std::cbrt(-0.0) << '\n'
        << "cbrt(+inf)      = " << std::cbrt(INFINITY) << '\n'
        << "Accuracy and comparison with `pow`:\n"
        << std::setprecision(std::numeric_limits<double>::max_digits10)
        << "cbrt(343)       = " << std::cbrt(343) << '\n'
        << "pow(343,1.0/3)  = " << std::pow(343, 1.0 / 3) << '\n'
        << "cbrt(-343)      = " << std::cbrt(-343) << '\n'
        << "pow(-343,1.0/3) = " << std::pow(-343, 1.0 / 3) << '\n';
}

Possible output: 

Normal use:
cbrt(729)       = 9
cbrt(-0.125)    = -0.5
Special values:
cbrt(-0)        = -0
cbrt(+inf)      = inf
Accuracy and comparison with `pow`:
cbrt(343)       = 7
pow(343,1.0/3)  = 6.9999999999999991
cbrt(-343)      = -7
pow(-343,1.0/3) = -nan

See also