Method fdim

samples/common/half.h:1735–1742

Positive difference implementation. \param x first operand \param y second operand \return Positive difference stored in single-precision

    /// \param y second operand
    /// \return Positive difference stored in single-precision
    static expr fdim(float x, float y)
    {
#if HALF_ENABLE_CPP11_CMATH
        return expr(std::fdim(x, y));
#else
        return expr((x <= y) ? 0.0f : (x - y));
#endif
    }

    /// Fused multiply-add implementation.
    /// \param x first operand