Dean,
Have you done the same experiment with your speakers instead of the headphones? That's something that I have planned to do for a while but never got to it.
Basically it was harder to tell the mp3s from the FLAC using the speakers vs the headphones though this was with my M22s and not M80s nor was the room setup very good. I should rerun the test using my M80s which were boxed up for moving at the time to get a better “full range” comparison. Several posts down is a brief description of the results I got using my M22s.
Here’s my original post on the test I did showing the results of the 128 and 320 bit comparisons:
http://www.axiomaudio.com/boards/ubbthre...true#Post241538The type of stereo encoding used to rip the mp3 had a profound impact (at least as much or more than bit-rate) on the SQ and ability to distinguish is from a lossless track. Listed from best to worse:
“Stereo” 320bit indistinguishable from FLAC in blind tests.
“Joint Stereo” 320bit only distinguishable from FLAC in blind tests on tracks w/discrete instrument placement.
“Dual Channel” 320bit easily from FLAC in blind tests.
Note that lower bit-rates did fare worse but the stereo encoding method was most easily detectable.
