This bug was introduced in the last versions. If a calculation of crosscorrelation from Confocor 2 raw data was interrupted with the error "Flow is incorrect" (this error is shown if next photon arrives before the current one), most probably you met this problem. In some cases depending on data some photons might be accounted in a wrong channel (respectively with assignment of arrival time which differ from the actual one in 1 - 3 clock-time events). In spite of this bug the calculation of ACF, PCD, etc. from one channel was correct.
Please ask for a free update of the Measurements database if you are working with two channel Confocor 2 data.
See the whole list of updates here