I am working aggressively to merge all my data from MyHeritage and Ancestry into one RM7.5 managed database. I linked RM to my tree on Ancestry and created a fresh database with that data. I then did a full GEDCOM export from MyHeritage and imported it. Obviously that created a few thousand duplicates. I used the search duplicates function first with a more exact match and walked through each of those one at a time. There were only a few hundred. I then did the "sound similar" and got a few hundred more.
There are still hundreds of duplicates out there that the only way I seem to be able to find them is walking through the list in alpha order. That's a pretty slow way to work through over 8000 names. I can't understand why George Beharr Crewson and George Beharr Crewson Sr with the same dates of birth and death don't show up as a dup. Likewise when George's wife Elsie is listed on one as Elsie Crewson (married name) and Elsie <blank> (no maiden name) don't show up.
It seems like the dup checking misses hundreds of obvious duplicates. It's not like I'm expecting RM to automatically merge them, but I am expecting RM to find them.