I’m old and I’m trying (not very successfully) to get rid of our vinyl. My son, who just turned 50, is mostly into new vinyl releases that are collectible. They’ve got arty covers and labels, and the discs themselves are colorful and have special effects. He also shops for older collectible albums.
I gave away a boxful of jazz albums to a local public radio jazz station and took the tax write-off. I sold some fairly collectible jazz albums on ebay a few years ago and did pretty well. The Japanese are (or were) big on that stuff.
I guess I’m the first to mention that there’s a lot of music available on old LPs that never made it over to CD. Especially if you’re into jazz; you’d be surprised how much is out there.
I agree: they (and you know who I mean… ) should make compact disks with those errors embedded, call them hyper-realistic and they would sell like bagels fresh from the oven.
Here’s a simplified version that helped me get my head around the idea. It doesn’t fully explain it, but it helps.
Assume for a moment that you have a speaker that has only two states: on and off. If you switch it on and off fast enough, you get a tone, a frequency of sound.
Imagine you can switch it from off to on 880 times a second. What would be the highest pitch you could get? Or, to put it another way, what’s the most number of times you can turn it on? 440 times, per second, or 440Hz.
If you think about that in reverse, you can see why the maximum pitch you can represent if you sample 880 times per second is 440hz. You can always sample the same state for longer, but not shorter.
This is not the full thing, of course. The real example has all the states between on and off, like in a sine wave. But, for me, at least, I find it gives me an intuition of how the Nyquist theorem can be true.
I seem to recall reading a quote by Neil Young saying the LP “Harvest” was better than the CD of it. He said with an LP, it’s like if you look through a screen window and examine each little square of the screen, and you can see a mixture of shapes and colors etc. in each. CDs, he said, reduce the small squares to the predominant shape, color, etc.
Maybe my hearing is shot, maybe I need better speakers, whatever: I can’t hear much difference between a .wav file and an .mp3 or a .wma. I grew up listening to mono AM radio so it all sounds pretty good now.
Vinyl doesn’t sound better than CDs: as a matter of fact, it sounds slightly worse to me, all else being equal. However, CDs are redundant with ripped digital because they sound the same.
Vinyl sounds different from digital, so it is not redundant to listen to stuff on vinyl and digital. Plus, the album artwork is larger.
You could lose the word “digital” and this would still be true, of course. And it’s an important part of the psychology of the debate I think.
With digital, one is making a deliberate decision to sample at a certain rate, where sampling at a higher rate would lose less information. This bothers people despite the reality being that the rate is entirely adequate.
With analog, there are always going to be limitations on the medium such that information is lost. But at no step is a pragmatic decision made to lose that information.
Somehow this allows analog enthusiasts to kid themselves their preference is more “pure” and (ironically) non-lossy.
Meh, doesn’t help. Even in Nyquist you’re still taking slices. Many discrete readings at a moment in time which are trying to represent an infinite continuum. It really doesn’t fit my understanding of the physics. I trust the mathematicians on it, but I’d really need someone to explain it. Maybe Numberphile will do a video.
This is a good post. Really, with vinyl, you got dust gathering on the needle or in the grooves that distorts the signal massively, and that’s just one issue, which you never have with a digital medium (ok, a CD can also have errors because of dust, but I’m talking about best circumstances for playing digital recordings, such as streaming lossless audio files).
And don’t get me started on room acoustics, as some others already pointed out. The layman has no conception how much of an influence this makes on hearing music. Most music today is heard on crappy hardware and speakers/headphones anyway.
Waaay before you even get to that there are limitations on how faithfully the groove reproduces the sound and on how faithfully the needle tracks the groove. The smoothness and resolution of all the mechanical processes involved have limitations regardless of any “maintenance” issues like dust. The grooves are made by an imperfect process, and are imperfect, and are then traced out imperfectly by an imperfect needle. Losses all the way down.
So consider a simple sine wave of a frequency of X. How many times do you need to sample it to guarantee perfectly recreatevthe wave? The answer is twice the frequency, because if you sample anything less you risk starting the sample at a point in the wave where you can not capture the whols thing.
Now remember that any complex wave can be broken down into a series of sine waves of different frequencies. That’s what a Fourier transform does.
So now you have a bunch of pure sine waves of different frequencies and amplitudes. Each one can be perfectly digitized by sampling at or above its Nyquist frequency. Once you have digital representations of all of them, you could recreate the original wave perfectly by converting each sampled wave to analog, then combining them.
This isn’t exactly what happens, but it might shed some light on what is going on. So long as you are are sampling at a frequency higher than the Nyquist limit of any of the sine waves that go to making up the signal, you can perfectly recreate it.
As I said, dust is not the only issue with vinyl. Really, with digital recordings you cut so much of the issues of hearing vinyl out, so the only reasons for preferring vinyl to anything else is the artwork. And this I miss too.
) should make compact disks with those errors embedded, call them hyper-realistic and they would sell like bagels fresh from the oven.