I went through a period where I exhaustively tested lossy (lame mp3 vbr -0 and cbr 320) vs lossless. I would even purchase HD tracks in FLAC format and use SoX to convert the file to lossy and than convert the lossy back to the same bit depth and sample rate of the FLAC file to ABX. Every time I tried, I was unable to identify a difference. There were several times where I would be listening to an MP3 or streamed song and swear there was an anomaly or glitch that was different. If I didn't already own the CD to rip, I'd purchase the CD. I did this probably a couple of dozen times, and still nothing. I performed well over 100 ABX trials. It was only after passing the Philips Golden Ear Challenge that I felt comfortable that my hearing was not just deplorable. I even have a verifiable Foobar ABX test in Sound Science somewhere that shows I was able to ABX the Tidal FLAC/ALAC with their AAC 320 version. In this forum thread, it was revealed that something in the way that Tidal was converting the AAC 320 files was creating a noticeable difference. I explained in my post that I was unable to reproduce the same difference when I converted the same song. I could not pass the ABX between my ripped FLAC, Tidal's streamed FLAC, or my converted AAC file. Only the Tidal AAC 320 file was different, and only for the 3 test files that were available on their website to demo. I don't know if this was an issue with the actual streaming files from their library of music. I even grabbed the Sammy David, Jr. - The Decca Years CD to see if I could hear what @bigshot was hearing at AAC 192. I didn't hear anything different, though I am not too familiar with Sammy's catalog.