In the paper https://doi.org/10.1016/j.physleta.2011.08.072 authors propose a modification of the conventional delayed feedback control algorithm, where time-delay is varied continuously to minimize the power of control force. Minimization is realized via gradient-descent method... However, the derivation of the gradient with respect to time-delay is not accurate. In particular, a scalar factor is omitted. The absolute value of the scalar factor is not crucial, as it only changes the speed of the gradient method. On the other hand, the factor's sign changes the gradient direction, therefore for negative value of the multiplier the gradient-decent becomes gradient-ascent method and fail power minimization. Here the accurate derivation of the gradient is presented. We obtain an analytical expression for the missing factor and show an example of the Lorenz system where the negative factor occurs. We also discuss a relation between the negativeness of the factor and the odd number limitation theorem. (read more)