Abstract: Method of deriving a quality signal for a binary read signal having bit cells of constant length, and reading and recording devices which employ that method. During a reading process, an information pattern of optically detectable effects, representing certain of the bit cells, is scanned. A read signal is generated therefrom in which first read signal portions that exceed a reference value correspond to the effects. The quality signal is derived (in an analysis circuit) from the first read signal portions based on a measure of the deviation of first read signal portions corresponding to a particular length or mutual differences between length deviation of first read signal portions of different lengths. The quality signal can be used for adapting a recording strategy used in the recording of information, or for setting an equalizing filter.

Abstract: In an optical recording device wherein an information signal is recorded in digital form as recording areas and intermediate areas, an optimum write beam intensity is determined. The resulting recorded pattern is then read, and the read-out signal is analyzed to derive an analysis signal (V.sub.a) indicative of deviation of the DC component of the read signal from an optimum level in relation to the minimum and maximum levels of the read signal, which optimum level is defined by the duty cycle of the information signal. At such optimum level the average ratio of the lengths of the recording areas to the intermediate areas will correspond to the duty cycle of the information signal. The write beam intensity is then adjusted to reduce such deviation.