yes, this is definitely necessary. the current version has trouble with large datasets and large files. it simply freezes all the time after hashing phase on two of my machines with different datasets of several TB. i left it running for 1 month (!) on both machines. it doesn't crash but hangs...
