Point taken! However, given that I'm looking for ways to control the 'silliness' of a dissociated-press algorithm, critical accuracy isn't necessary. :-)