Neil Young: great songwriter, dumb engineer
I love Neil Young's music, but he is a complete idiot with regard to analog vs. digital technology. He, along with just about everyone else who touts vinyl as inherently superior to CD, should spend some quality time at Hydrogenaudio.com. That sure cured me.
On that site people get banned if they repeatedly make claims that cannot be scientifically verified, and they offer the proper procedures and software to run properly controlled experiments with. Over the last several years of objective and fair double-blind ABX testing, the results tell us:
1. No human population can reliably distinguish 44.1k/16 bit music playback from any other "hi-fi" resolutions/bit depth. That's what hundreds of studies, all with duplicatable parameters and objective criteria confirm. Not even professional mastering engineers can. 48/88.1/96/192k sample rates cannot be distinguished from 44.1k, and 20 and 24 bit depth cannot be distinguished from 16 bit--so long as the tests are a) all at a consistent volume level, and b) properly controlled and double-blind, so that neither the person giving the test or taking it have any idea which example is which.
Higher sample rates really only give you a higher ceiling for top frequency, and only the pure virgin ears of young children can typically hear anything above 22k.
And while it is absolutely true that very low volume program material will have better resolution at higher bit depths, the benefit is negligible at "normal" listening volumes. In other words, you can hear the slightest inhale breath of a piccolo player before he/she begins to play a
ppp melody line if you crank up the volume on your playback system to the max, but if you leave the volume setting there your ears will be bleeding and ringing with tinnitus when the
fff orchestra hits are played. When volume levels are set for normal listening levels below the pain threshhold, the super-quiet passages are so soft and close to the noise floor that no one can consistently distinguish the 20/24 bit ones from 16 bit.
You can call that my "opinion," but the results of hundreds of ABX test verify this. In fact, it is safe to say that for all intents and purposes, 44.1/16 bit IS "ear drum" resolution for human beings. The bats in your attic and your family dog might be able to hear additional content in a 96k recording, but you can't.
"But I DO hear a difference!" you claim. Well, what you might be hearing are issues with how the D/A conversion is occurring within your playback equipment. The easy way to control that factor is to set up a test like this:
a) Take a native 96k/24 bit sound file and make a copy.
b) Downsample it to 44.1/16 bit. Much of your data is lost in the process
c) Uprez it back to 96k.
d) Play both the original and the uprez'd back on 96k/24 playback equipment
e) Compare
The second file will essentially be at 44.1 16 bit resolution (as far as available information goes) but processed identically on the same playback equipment as the native 96k file, ruling out potential differences with how the playback system's D/A converters handle different sample rates. From everything I've read, you'll fail any ABX discrimination test between the two (i.e., you'll only guess right half the time given enough trials).
2. Vinyl is in no quantifiable way superior to CD. With only 80 db of s/n range, CD's 98 db totally smokes it. In listening tests, people easily identify vinyl from digital sources, but only for the texture of the noise and distortion--the pops, hiss, and other goofy idiosyncracies of the medium. But make a 44.1k/16 bit recording of the vinyl, complete with the pops and hisses, and people can't discern the difference between real vinyl and 44.1k/16 bit recordings of vinyl in ABX testing.
_____
Having said all that, I would like to offer some possible theories to what is really going on in the minds of people like Mr. Young:
1. Early CD engineering sucked. In the early days of CD, engineers were pretty excited to have the technology, and some of the mixes were brittle and harsh because one could have more highs without increasing hiss. Back then, IIRC, clear, articulate high end was the tell-tale sound of "hi-fi". Also, engineers quickly discovered that digital distortion was entirely unforgiving, unlike analog distortion, which can be subtle and sometimes desirable, so they learned to avoid 0 db like the plague. The result was soft and shrill use of the medium. When Neil Young heard that back in the 80's he thought it sounded "digital" and knew it sucked compared to rich, fat vinyl. But what he was really hearing was poor engineering, not a poor medium.
2. Vinyl's idiosyncracies have a very familiar, warm, nostalgic sound. I love the sound of vinyl--the way the mixes were/are mastered for the medium, the hisses and pops, everything. I also really like the way one operates a turntable--putting that big plastic disk on a spinning drum and placing a diamond in the groove to hear the resulting noise from amplifying the sound of it dragging through the channel. That's wonderfully low-tech! Also, in an odd way, vinyl is potentially more permanent, as CDs, HDDs and SSDs have limited lives but properly stored vinyl could be playable centuries from now. But while vinyl is a splendid blast from the past, I would never make the claim that it is any quantifiable way superior or even close to equal to CD. If we were to believe the claims of some, then even a wax cylinder Victrola, being "analog", would be superior to a CD!
Those are only my guesses. I sympathize with Young's frustrations, but his initial diagnosis was totally wrong. And all of the deep-pocketed audiophiles who swear by vinyl? I hope they continue to enjoy it, but I also hope they can be honest and disavow themselves from the snake-oil explanations as to why it is better than CD. It is only clearly better in the manner it strikes upon the chords of nostalgia within the hearts of its proponents.
