Why does vinyl (analog) sound better?

This is the basic problem with digital. The industry doesn't really care how it sounds. The guys who developed SACD are rewarded by being left sharing a coffee machine.

 

The strange thing is that scientific explanations are often given by those who want to prove that vinyl is superior.

 

I find it a bit hypocritical that people are sensitive about euphonic distortion claims. CDs are full of euphonic distortion. Heck, master tapes are full of the same euphonic distortion. It's called "compression". This distortion of the original signal, when applied to a pure unadulterated mix of the music, yields a signal that is, for better or worse, more pleasing to the ears of many people. Similarly, distortions of frequency response (equalization) and harmonic distortion (hard-hit tape recordings, guitar amps..) are pleasing to many people.

How is that any different from the distortions that are attributed to vinyl? And moreover, why is wanting to know about these distortions somehow wrong, or impossible? It's certainly neither. It might even change your listening habits for the better (to or from vinyl). As for myself, knowing all the voodoo that goes on when pressing and playing a record makes me love vinyl more.

 

I think the only thing scientific explanations can (try) to do is explain why one form of reproduction sounds different from another or from the source. It does nothing in the way of convincing me that one format really does sound better than another one because of what's on a particular graph, especially as far as my wallet is concerned.

It makes me think of the old (1970s) AR ads. One of them read:

"We at AR call it 'truth in listening,' and when you think about that you realize that's what hi-fidelity is all about."

That is what hi-fidelity is all about -- what sounds good. Anything else is hot air.

Dale

 

Yes.

Of course, such explanations are only valid if they are based on *science*, rather than what might be intuitive.

 

I assume a good mastering as most of my 400 SACDs are. You have to assume the same level of mastering to have a meaningful discussion.

The noise shaping occurs so far out in DSD I can't honestly hear it and neither has John Atkinson. I think it's a myth that it is audible to even a critical listener.

 

That is only true Luke if science can fully explain audio phenomena. I and many others don't believe that is correct.

Of course that gets into a subjective/objective debate which is best saved for another time.

 

You're not understanding me correctly.

Science can help us understand differences in sound. However, it is disingenuous to *make up* "science" in an effort to explain why something is different. I see that quite a bit.

 

WHAT? Science is intuitive!!! But intuitive hypothesis' are just the first step. Then you gotta test em. There in lies the rub. No doubt a lot of vinyl enthusiasts put forward some wild and unsupported explinations for what they percieve as superior sound. But it comes as no surprise in the vast absence of any attempt to research what causes these perceptions by those who are capable of doing the research. It is aoften written off as a mere bias effect. I think that is a bunch of B.S. You'd think some one would wise up, investigate the phenomenon and then put it to good use.

 

I would contend that Nyquist is not intuitive for most people.

 

also some of us are just laymen, and not scientists, and we try to understand the phenomena as best we can, based on our own empirical evidence

we are not making up science, just going by what we've been taught. frequently the answer we hear is something like "oh you're crazy that's not how it works" and then when we ask "how does it work then?" we are pointed to sources that are WAY above our heads - or, more disturbing, do not seem, to a layman, to refute what we already thought.

people with scientific knowledge can sometimes forget the vast gap between what they know and what others know.

all that having been said: my friend Mike is a pretty high-level scientist, and he has explained to me that by Nyqust you should not be able to hear the difference between DVD-A at 192/24 and Redbook at 44.1/16

my empirical evidence is different from that - I can hear the difference, and I imagine most other people could hear it as well.

so what gives? obviously a layman is going to try to analyze it as best he can and try to figure out why there is a difference

 

I take it by "making up science" you mean untested hypotheses asserted as baseless conclusions. Just how would one find a scientific explanation if not by starting with conjecture? I'm not clear about just what you mean by science here. Usually what I see in these typical pleas for using "science" are attempts to debunk and discredit rather than to really find out why the sonic differences exist. The energy is concentrated much more in nay-saying, skepticism, and easy dismissal than in any kind of process of discovery.

 

Some differences.

Agreed, but I don't see where we are making up science here. We're just discussing what we hear under different formats.

If you believe only some differences are explained by science then it follows that subjective listening tests are valid and essential.

 

If you hear a difference, that's fine. Just don't try to come up with a *reason* for that difference that isn't grounded in scientific reality. Nyquist *isn't* totally intuitive. So saying "it makes sense that 16/44 wouldn't be able to capture this" isn't necessarily true.

 

No, that's not what I mean. Example:

1) This CD doesn't sound as good as my LP. I think it's something with the transients.
2) It must be because Nyquist only applies to sine waves, not transients! 16/44 simply can't capture those! Look at what it does to square waves!

Unfortunately, that isn't true. Does the LP sound better? It very well might. That isn't the issue, though. If those transients are audible, 16/44 will capture them. Saying otherwise is just a shot in the dark.

Does that help?

 

Once again you are being negative, instead of trying to figure out why they sound different. Its much easier to tell someone they're wrong than it is to come up with something correct yourself I guess.

If you are dealing with an equivanent mastering job, and you compare LP to CD...why will there be a difference? is still the question. I think we touched on it breifly with distortion characteristics of each format, and the distortion that the necessary playback gear adds. But this can be explored much more with science I have a feeling... and it would be quite constructive.

 

But what's worse? Telling someone they are wrong, or letting a false theory spread like wildfire?

I *don't* have all the answers. Don't claim to. But I'd much rather say "we don't know" than try and come up with reasoning that is clearly wrong.

I don't disagree at all. I'd *love* to finally figure out some scientific explanation for why some people prefer vinyl. Perhaps it could improve CDs. Again, though, the key word is "scientific". It's fine to say "I prefer the sky in Montana over the sky in New York". It's *not* fine to say "I prefer the sky in Montana over the sky in New York because the sky in New York is lime green".

 

I'm not sure I agree with your premise, which is quite biased. But, since I've had similar experiences with many (not all) Lps seemingly sounding better than their CD equivalents. I've listed the potential reasons. I'm not endorsing them. They are speculations.
Possible causes:
1) Placebo effect
2) Needle in vinyl groove makes a physical sound, which is then amplified. This may actually produce the emotional connection people talk about
3) Most CDs poorly mastered. Early ones lacked the best A-D conversion while later ones included digital compression and other market-driven bad behaviors
4) Digital lacks high frequency time decay that compromises its ability to emulate real sound
5) Most LPs are mastered from better, earlier/fresher tapes than the subsequent CD release - this is one reason given when earlier LP sound better than audiphile rerelease

Hope this helps. I enjoy discussion more than argument. It's a twenty-year question, one I've never heard answered to complete satisfaction.

 

Let's agree for a second that a transient is a drum crack.

Now let's assume we are deaf - and we really have no way
whatsover of hearing what a drum crack sounds like

BUT

we have a magic machine that lets us see what the
corresponding sound wave looks like.


Now lets assume we have a way to capture the image
that the magic machine creates. But each image
costs a specific amount of money to keep in storage.


Now lets assume we have several different ways to
store the visual image. But unfortunately the less
costly mechanisms alter the image And the more
pricey mechanisms are outside our budget.


Is this still sounding like SCIENCE or is it sounding
like an economic decision ?


I wasn't in the room in 1977 when the redbook
designers agreed to use 44.1/16, but common
sense tells me that there was a heated debate,
and the final decision was a compromise between
cost and quality. Why ??? Because that's
how ALL things are engineered.

The idea that 44.1/16 was selected first -
because it applies for all human ears - just
doesn't seem likely. And its seems to be
a fundamental building block for a lot of
the ""cd is great - we don't need anthing
else" type arguments.


I'll agree that the redbook standard wasn't
developed by deaf people and their magic
machine, if we can agree that the people who
DID develop the redbook standard were probably
not the same people who decided what ALL
human ears are capable of hearing (or not
hearing).

 

Again, that isn't the argument. It's possible there *is* a difference that we can hear. I simply take issue with *why* there's a difference. It's misleading to say 16/44 can't capture something when it clearly can. Or saying things like "it isn't frequency based, it's time based".

There are many possible causes for differences. Let's debate those, instead of the ones that aren't possible.

 

The brain does a great deal of processing of aural information without conscious thought on the part of the listener. For example, if you hear a sound to a lesser extent with your left ear than your right, your brain automatically locates the sound toward the right side. Thus, without conscious thought on the part of the listener, the brain uses audible cues -- some of which are extremely subtle -- to determine the direction and distance of the sources of sounds, the direction in which the source is moving, the relative positions of sources, the likely identity or composition of the source, etc. Also without conscious thought on the part of the listener, if expected cues are missing, or the cues are different from what the brain expects to hear in relation to other cues, the brain will automatically register the sound as being unrealistic. IMO, and at least in my experience, my brain automatically and quite consistently registers sounds reproduced from vinyl as more realistic than digital even in instances in which the digital is reasonably well-mastered and has good tonal balance.

Because the process occurs without much conscious thought, it may not be easy to "reverse engineer" why my brain deems a particular digital source to be "phony" or deems an analog source to be more like the "real thing." In many instances, I think the problem is that some of the subtle cues are missing rather than distorted. In particular, I think subtle cues found in low level detail help the brain sort out distance and position of sources (essentially, the image of the soundstage). I think playback of redbook digital lacks some of this information, your brain recognizes it, and your brain determines -- again, without conscious thought on the part of the listener -- that the reproduction is not authentic or persuasive. Even if the listener cannot reliably identify why his brain is transmitting the message that the sound is "bogus," he can nonetheless reliably report that he is receiving the message and, as a consequence, is dissatisfied with the listening experience.

Additionally, I must add myself to the ranks of those whose listening experience indicates that digital playback is not fully successful with hard transients like drum "thwacks" and that high frequency sounds such as cymbals often come across as electronic noise rather than vibrating metal.

I don't have the scientific explanation for the lack of low level detail, the weakness with hard transients or the poor reproduction of high frequency sounds, although I don't dispute that there should be a scientific explanation for all that I hear. Nonetheless if you focus your listening on the foregoing defiencies of digital playback, they are clearly audible (or missing). Thus, I believe there must be a scientific explanation -- even if neither I nor anyone else can currently offer a fully satisfactory explanation. What I do know is that even if I'm not consciously listening for these deficiencies while listening to a digital source, and even if I'm not conscious that I'm listening to a digital source, my brain often reports back "phony" when listening to digital sources and I become disengaged.

 

The #1 giveaway of digital right there. All in the drums. Even if its a piece of vinyl that has been digitally mastered, I can tell. fake fake fake.

This is what leads me to believe it isnt just distortions in vinyl that make things sound good...that it has a lot to do with digital vs analog.

And chances are, that if its happening with drums and cymbals, its happening everywhere, just maybe to a lesser degree (to our ears)...but still something noitceable when it comes to the "real" factor.

 

That has been my experience as well Hal. I honestly wonder if transients are indeed fully captured by Nyquist. I think the problem here is that Nyquist in a way describes two of the three dimensions of an audio event. The question is, what is the missing dimension or dimensions?

Could it be data or time related? That may explain why higher sampling seem to capture transients better.

Cymbals sound very poor at 16/44. Hey that rhymes.

 

It helps me understand what you meant in your post, but beyond that, no.

I don't believe it has been proven by science or anything else that 16/44 sampling will capture everything perceptible by human hearing. You asserting this to be true doesn't prove it no matter how many times you repeat. You have accepted this to be true based on certain studies or analyses by certain individuals, which you take to be authoritative. That's all there is to it. I'm not saying those studies aren't valid, by the way - they may very well be the best information extant. But then again they may not. Historical validity in science is all about repeatability and design of experiments and the strength and durability of a consensus on particular ideas within the scientific community. Everything is provisional, and I think ideas concerning audio and human perception are more provisional and complex than many other areas of measurement and theory.