MCI scans selected directories and reads the musical content of the files. Thats right! Its not just doing it the old fashioned way by reading the tags and file names or calculating hashes. It reads the content and extracts salient features of the recordings (NOT comparing files byte by byte). The files dont have to be identical. If they contain the same record they will be identified as duplicates even if the files are different due to different compression, different quality, or even presence of some noise. This feature is unique for the applications of this class.
In addition to finding duplicates MCI has a built-in tag editor and a file manager. MCI also compiles a full report on quality of the files in your music collection with respect to quality of the file formats and completeness of metadata in tags.