A focusing signal generator sends an instruction signal to a voice coil driver. The driver drives voice coils to move an objective lens in accordance with the instruction signal. Under this state, a laser beam emitted from a laser head is irradiated on a first photoelectric converting element and a recording surface of a recording medium via an optical system, respectively. The laser beam irradiated on the recording surface of the recording medium is gathered in the recording medium in a non-focusing state. As the non-focusing state is maintained, a first signal processor provides an arithmetic processing to the electrical signal, which was output via the first photoelectric converting element, whereby the output power of the laser beam from the laser head is detected. The detected data signal is output to the laser driver. The laser driver controls the laser head, so that the laser beam having the recording output power is emitted. When such control is finished, an end signal is output to the voice coil driver from the first signal processor. The voice coil driver drives the voice coils in accordance with the end signal, and returns the objective lens to the initial position.