def q_large_x_complement(a, x, with_error = false)
          q = q_large_x(a, x, with_error)
          with_error ? [1.0 - q.first, q.last + 2.0 * Float::EPSILON * (1.0 - q.first).abs] : 1.0 - q
        end