I've just done a blind listening test using my wife and a friend as guinea pigs. I used some tracks ripped from CD in compressed FLAC (level 8), uncompressed FLAC and WAV. Using a blind test involving the three file types and the original CDs, my test subjects consistently picked-out uncompressed FLAC and the CD as being equal and best, albeit by a small margin. I used XLD on my MAC to create the files and they were all accurately ripped. In theory, there should be no difference, and I think that some of the perceived differences could well be down to the decompression algorithms on the player that I was using for the test, a Samsung BDF-F6500 blu-ray player. However, the latest XLD update also introduced a later version of the FLAC encoder which I used to create the uncompressed FLACs and WAV files, (I already had the CDs ripped as compressed FLACs using an older version of XLD with and earlier FLAC engine) so that could also have accounted for some of the differences perceived, although that wouldn't be relevant for the WAV files.
As I've only just started ripping my CDs, in readiness (probably) for use with a Sonos Connect in my a/v system, I'm going to (re)rip using uncompressed FLAC. Although there is an increase in disk space usage (for my files, the increase in storage for uncompressed FLACs seems to be ~35%), having [potentially] to negate decompression routines seems like a sensible trade-off if the results of my testing are accurate, which I'm confident that they are given the blind tests. When I listened I slightly preferred the CD playback by a whisker, but couldn't tell that much difference between the other formats though, with perhaps the WAV files sounding slightly brighter. However, given the many limitations of WAV storage, uncompressed FLAC is my friend.
HTH
Regards
Clem