I repeated my experiment, this time with an analog cable loop... Your looped MP3 and a recording of the original with Deemphasis applied do sound very similar. This doesn't explain why an analog recording of a CD would sound bad. Does this affect all the line inputs? What about microphone signals?
It seemed to be the analog line ins having the problem (not the FF pre's).
But I didn't do testing to confirm that 100%. I was short on time to get it working before a session. I could be wrong.
I didn't test the digital I\O either.
D\A stage seemed fine other than the limiter distortion when the level was pushed.
I was on the newest driver when this happened, so I'm a little scared to update it again until I have a couple days free to make sure everything is cool.
I imagine if I replaced the system32 files after it would be fine though.
@RME Support, do you know if the op-amps in the upcoming ADi-8 QS have been changed from the DS?
Guess so, but I don't know the exact details of the board layout... You can expect quite an improvement, I'd say (not that the DS was all that bad... )