The threshold probability for differences between observed and theoretical (based on observed mean, sigma) CDFs.
The Kolmogorov-Smirnov test is used to evaluate the probability that the measured deviation between observed and theoretical CDFs is a chance occurrence (i.e., that the two distributions are actually the same).
In general, any significant probability indicates that the two distributions compare very well. When the K-S probability exceeds the TUNE_DPRBTHR threshold, clipping stops and the fit is accepted.
The default value is 0.9. (i.e. 90%).
Martin Clayton