Exactly that came to my mind as well. Which is why a detailed analysis would be advisable in this case. To really dig into this, one would have to decode everything to PCM, synchronize all the stuff, subtract it from each other and check what will remain.

Neither do I want to know which so-called "lossless" tracks, repackaged in Dolby TrueHD and DTS-HD MA are actually retrieved from lossy sources. Not to speak of converted pitches (25 fps, 23.976 fps, etc.) and other stuff.

Yet another question for me remains unanswered: How much better is DTS @1536 kbps really compared to AC3 @384 kbps? Most comparisons dealing with that question are invalid because they look at the end result only whereas often no one takes the effort to check if identical data was fed to the encoders.