Oh that is interesting!.. and indeed I did consider doing a 'hash compression' where I would generate offsets to numbers that might be similar, then do direct c

Oh that is interesting!.. and indeed I did consider doing a 'hash compression' where I would generate offsets to numbers that might be similar, then do direct comparison on those ones that are similar. I'll take a crack at doing this - It could certainly be an improvement and would just be fun to try out.