Yup, depending on the kind of DAC and filtering, I can certainly appreciate where you're coming from. IMO, you don't even need to bother with a filter when you have 192kHz data. Imaging artifacts would already be >96kHz so the natural roll-off of the "system" (ears, speakers, possibly amps) good enough to not cause trouble. In the blind test I ran years ago (linked above), it was actually interesting to find that folks (151 total respondents, "audiophile" types) seemed to have a preference for the MP3 version of the heavy metal track. I did not see this preference with the Pink Floyd or vocal/pop/country tracks. Indeed I wondered if the MP3 processing might have just toned down the "hash" and especially higher frequency content in music like that which could be perceived as "dysphonic". [Just a note, most MP3 blind tests are run with high quality productions like classical! I think this might be why we don't see more results showing the potential benefits of MP3 with nasty sounding music. ;-] Audiophiles in general never want to say that they prefer MP3 (other than you @Doctor Fine ;-); after all, we want *ultimate* fidelity, right? So it might be surprising for some to consider this possibility and be more open minded that MP3 is indeed its own form of DSP which in some instances might render the output more "euphonic"!