I have been meaning to comment on this but keep forgetting. What do you mean "well-engineered MP3?" MP3 is not hardware so engineering does not enter the equation. But perhaps you mean using best-in-class encoder? If so, even the best of the best won't achieve transparency to the PCM source. It was stated that way when the codec was designed, and it is true today. I am often challenged to show this is the case at the highest bit rate of 320 Kbps and I have shown it. Here is the most recent occasion:
These are encoded using the professional encoder in my Adobe Audition software. Clips were from another test being discussed then, ironically related to the work of Bob Stuart showing audibility of resampling of high resolution music (see my article here
http://www.madronadigital.com/Library/High Resolution Audio/High Resolution Audio Matters.html). So no hand picking of the worst case scenario. Granted, I have lots of training to hear compression artifacts but still, transparency is either there or it is not. If a human can hear the artifacts, the mission is failed and transparency not achieved.
That aside, would you please clarify how you are OK with MP3 artifacts but worry about vibrations making an audible difference in gear other than mechanical analog equipment?
First, as I recall the deviation started with Orb claiming a superior sound from active speakers, my response to that, then Don chiming in with his opinions.
Amir, are you saying there's no such thing as an inferior-engineered MP3 recording? I realize it's down-converted to MP3 format, but somebody still has to engineer that process, right? If 2 different companies where to down-convert the same recording to MP3, does not one outcome stand to possibly be superior or inferior to the other? What am I missing here?
As for your hearing tests, as I recall these tests were all done on your computer, correct? Not to say the results wouldn't be the same, but that's not same as performing the tests on your playback system where it really counts. And to the best of my knowledge your playback system is going to induce potentially far more distortions (much higher noise floor) than your desktop or laptop.
That aside, whoever said I was ok with MP3 or its artifacts? If I said anything remotely close to that it was that I said something like, because of the distortions induced into our playback systems raising the noise floor to such a high level that a good percentage of even a well-engineered high-rez recording remains inaudible so that you and others may well be hearing no more music info content than what is contained in a "well-engineered" (think down-converted) MP3 recording (without any distortions masking any of the MP3's music info).
That was about the time I quoted Harley, regarding the sound quality we hear, who speculated in the Mar/Apr 2009 TAS issue that he believes something catastrophic must be occurring at the recording mic's diaphragm that prevents much of the music from ever making it to the recording. About a year prior to that, Jonathan Valin said something similar when he claimed, "We are lucky if even our very best playback systems are able to capture just 15% of the "magic" of the live performance."
I was saying that even though I don't put much stock into anything they say, I was essentially in agreement with their lowly performance claims and that's where my comment of the MP3 entered.
Does that clarify things a bit?
That reminds me. To the best of my knowledge neither Harley nor Valin have rescinded their statements so I assume they stand even today. If that's the case, why the frick is Harley backing MQA when he's convinced the problem lays with the recording mic's? FWIW, Harley made his speculation based on the findings of a somewhat silly experiment conducted by Ed Meitner he described in that same article.
Does that make Harley a sell out too?
