File names with similar sized files are sorted as the same, but are totally different files. It would be useful if the files hash code was shown as well. This would enable a more definite comparison.

According to the program description, it compares files byte-by-byte, which would be more accurate than hashes (however unlikely it is to find two matching hashes, it's still a nonzero possibility) and, well, faster too. Either you've found a bug and should report it as such to the developers (I couldn't reproduce it myself, you'll probably need to show them the files or something), or you maybe misread the output of the program.