That is absolutely not true. People can be trained to hear that difference when the testing is not blinded, however in triangle tests that ability vanishes.
Maybe not anyone. I could be biased with sensitive hearing.
I know that in grade school, part of the requirements for joining band, due to the size of my school and overwhelming demand, was passing an audiometry test, where we were evaluated on a few different contexts related to ability to discern detail in audio, such as pitch and volume. I remember being pulled away into the principal's office where some of the test administrators were present, and they accused me of cheating and demanded to know how I did it.
Apparently, I was the only student in the entire state to get a perfect score on that test, at least for that particular year. Unsure if they were implying I was the first ever, but that seems ridiculous to me because passing the test boiled down to just paying close attention.
So I really don't know. Maybe the average person can't hear it, but I know just what to look for in the high-end and usually guess even 320kbps mp3 correctly from my own self-tests wherein I would randomly select between different encodings of a music file. I'm confident I would do well in an administered ABX test if I'm simply being tasked with finding the difference between mp3 lossy encodings and a lossless reference.
I don't claim to have golden ears but I am a picky listener. I've been pretty satisfied with good MP3 encodings like 320kb/s though I'm not sure I'd agree they're indistinguishable.
I had an experience I mentioned elsewhere in this discussion, about hearing a difference between 24bit 96 kHz and CD-quality in a sound studio. I don't know whether the operator screwed up or what. It doesn't mean I don't think CD quality is "good enough". But that's not the same question as "non-discernable".
On the other end, I have had a few encounters with satellite radio (SiriusXM) and I cannot understand how that product exists. In each car I've encountered it, I've had a visceral reaction and wanted it turned off after a few minutes of trying to listen to it. It put me somewhere between anhedonia and dysphoria.
I'm not sure I get you. If someone can't tell the difference in a blind test, that means they can't tell the difference. The result of a non-blinded test is of no consequence.
You said the same thing that I said but with more words.
There are a ton of people in the audiophile world who swear that knowledge of what you’re listening to makes a physiological difference and therefore it’s not the placebo affect. It’s insane, but then so is most audiophile marketing.
There is a clear distinction between audiophile snake oil, and detecting artifacts from a lossy compressed mp3.
You shouldn't conflate that with things like telling the difference between a 44.1KHz and 96KHz sample rate, which is bollocks.
That said, it's a good practice for any archivist to source the highest quality digital versions of albums possible, in case they end up befriending or needing to barter with aliens with much more sensitive ears.
