Vinyl vs. CD @ HD. Score sheets and music from A.C.A.

This is vinyl vs 44.1khz/16bit - not vinyl vs. digital. The only way to make a valid comparison is with vinyl sourced analog captured digitally at 44.1khz/16bit. If anyone has done this they know that 44.1khz/16bit does not capture the full quality of vinyl.

Tell that to Stan Lipshitz.

Good science does not have an expiration date.

One test with a few individuals is hardly definitive.

Sure, but it's a test that, as Lipshitz himself says, has been done many times, and always with the same result. And where is the test that refutes it?

I've actually tried this. I have some LP's that definitely sound better than their CD counterparts. I then ripped the LP to my computer (I ripped at 24/44.1, normalized the signal to get max dynamic range of digital signal, then interpolated to 16/44.1).
I could not tell the difference between the original LP and the digitized copy of the LP.
To me this suggests the problem is not digital, the problem is how the CD was mastered.

You left out "blind." Which demonstrates that you're the one who does not realize what scientific means in listening.

Then maybe you better read Lipshitz's article again.

When it comes to listening tests, if it's not blind, it's not scientific. Period.


That's mature.

Bottom line is, listen to what you like and ignore the pedants around here.

That is one nice record player !

I have done this. And by the same rationale you are likely using (sighted comparison) I 'know' you are wrong. There is nothing audible coming off the vinyl, that cannot be captured in full quality, by Redbook.

Yes, the mastering for vinyl is usually different, and vinyl has its own 'sound' (euphonic distortions -- I'm sure you can work out the etymology of 'euphonic' ) that some like. Both can be easily transferred to the digital domain by recording the LP output to digital. So?




Again, when there is likely to be difference (as is the case between a commercial CD and a commercial LP version of same recording), there is usually preference. But preference too is subject to bias. Your analog/digital comparison was laughably inadequate to the task of filtering out bias.

And the continuing silence as this dangling question still awaits a reply speaks volumes.

Has anyone been able to make heads or tails of the CDP comparison method used on the page the OP links to? As best I can tell, they attached two players to the preamp, did some vague level matching, and had someone switch between them using a remote control (making this at best a single-blind test). I don't see any information on how the presentation order of A and B was randomized, how many trials were done per comparison, and what the 'score sheet' means, and how the the results were analyzed statistically. And the videos have no sound when I try to view them (Firefox).

The room , btw , looks like it could be a sonic nightmare, with its bare wooden floors. Though packing it with seated human bodies might've helped.

To your ears. That's nice - I'm happy for you.

It is very hard to accurately test for, and there are many arguments against the validity of A B testing even when done double blind in the best possible way.

Also, just because most or all the people tested so far can't tell doesn't mean that no one can. Until I or anyone else who claims to hear a difference takes such a test and fails, you have no case.

Even further, even if a test were show someone who could tell, many would choose to simply not believe it or doubt that it was done properly (just like it happens when it shows a negative result).

In the end, it's a subjective judgment.

Actually, to my ears, as influenced by other senses and prior knowledge. Just like you. And to really determine if they sounded different, I'd have to do a double-blind level matched comparison. And so would you.

Get my point now?

THere seem to be only a few, actually, and none of them are very good.
Which is why double-blind methods are used when actual science is applied to the question of audible difference.


Wrong. THe case is also strongly supported from established limits of human hearing. DBTs can validate audible differences down to the known physical limits of human hearing. If you claim to have extraordinary hearing beyond those limits, YOU are the one who has to prove it.



Wrong again. Science has no problem accepting three things as objective fact: 1) that 'sighted' audio evaluations are subject to bias, and thus by themselves, inadequate evidence of difference and 2) that the way circumvent that bias is a double-blind method and 3) when a DBT indicates audible difference, there is a measurable reason. When DBTs have shown
such difference, they have been accepted -- I don't know any anyone who has argued against the validity of the positive difference results from these ABX tests, for example:

http://www.provide.net/~djcarlst/abx_data.htm

Furthermore, positive controls for DBTs consist of...wait for it...differences that really are predicted to be above threshold of hearing. The correlation is very good.

Your argument that objectivists are rejecting difference results from ABX tests is bogus. THe problem is NOT that 'objectivists' are rejecting DBT results that show difference - they are rejecting SIGHTED comparison results.
It's the 'subjectivists' who routinely reject DBT results for typically trumped-up reasons, when the real reason is: it threatens their audio 'world-view'. What if the emperor really DOES have no clothes?

No I don't. The burden of proof lies with me only if I desire to convince you and the rest of scientific community. Unfortunately, I don't have the means to run the appropriate experiment or I would.

You are perfectly within reason to doubt me. The point I'm making is that in order to prove me wrong, you'd have to at least subject me to a well designed test in which I fail to demonstrate my claim.


Of course sighted comparison results are not scientifically valid. The problem is there have been so few accepted scientifically valid experiments (and on such a small number of people), and even those experiments were not perfect. It's true the few experiments done so far have come up mostly negative (people could distinguish at loud volumes in the recent Meyer/Moran study - an often overlooked but revealing piece of evidence). I would also add that there have been people who can hear pure frequencies up to 24khz - 4khz above the CD frequency limit.

The bottom line is that much higher frequencies penetrate our ears and bodies in any real life sound event. These frequencies can be captured during recording and reproduced during playback. There is no way to reliably test in enough different ways with enough different people to show with absolute certainty that they have zero effect on how some of us may hear sounds. For this reason, it is ultimately a subjective judgment.

My problem is with those that feel the need to obsessively poo poo the listening experiences of others. They could simply state that they themselves don't hear a difference and know of no studies that have conclusively shown anyone else has either. State their skepticism and move on.

It's not like this is some ridiculous audiophile superstition like magical disc cleaning solutions that "open up" the sound - those things can be proven to not even exist in any measurable way. Higher frequencies not only exist, but they can be measured, captured and reproduced. They exist in every real life sound event. No one knows exactly how the human auditory system works or to what degree it may vary from person to person.

I would also like to say for the record that I've never heard the slightest difference between audio interconnects and power cables and I've experimented a little with both. As much as I wanted to hear an improvement, I couldn't hear squat. Everything sounded exactly the same to me. Now audio cables can and often do have small measurable differences and many claim to hear differences between cables. Do I go around poo pooing their experiences? No.

Looked at another way:

1. I hear a difference.

2. I know that it's a fact that frequency components well above 20khz are produced when an LP is played.

3. Frequency components above 20khz are not captured on the CD format and don't exist when the CD recording of the LP is played back.

4. The dynamic range of the LP is definitely less than the 96db limit of the CD format, which safely rules out limited dynamic range as the possible culprit.

5. LPs transferred to higher sampling rate/frequency response digital formats sound a little truer to the original raw analog sound of the LP to me.

6. Years of my general listening experiences with higher frequency response digital formats are consistent with the LP to CD experiences listed above.

7. Even before these higher resolution format existed, I sensed that there was something missing with LPs transferred to CD sound. I felt it had a slight hollowness and harsness and still had some digital glare. It didn't sound transparent.

8. As much as I'd like to conduct a scientifically valid ABX experiment, I have no means of doing so.

Am I supposed to deny my experiences just because I don't have the ability to scientifically prove them to myself or anyone else? It would be like trying to pretend something I witnessed didn't really happen because I have no proof.

Actually, that's not entirely true. While the frequency response os Redbook audio is up to the challenge(theoretically), the temporal resolution is not. What I'm saying is Redbook audio is not able to fully convey the spatial information.

Here's something to chew on. The human brain uses time to figure out where a sound is coming from, right? The sample rate of 48kHz, slightly higher than Redbook, is only able to reproduce a sample once every 48000th of a second. The human brain is able to discern timing differences smaller than that. The average person can detect the difference of about 15 microseconds, while some are able to discern 3-5 microseconds. A single 48k sample is 20.8 microseconds. Hmm... A single 96k sample is 10.4 microseconds. The appreciation of 96 kHz (or higher) audio, compared to 48 or 44.1 kHz audio, is a binaural (2-ear) phenomenon. If we plug one ear, it is unlikely that anyone would be able to distinguish a 96 kHz recording from a 48 kHz recording. The converse of this is that when listening with both ears, everyone can distinguish 96 kHz recordings from 48 kHz recordings, and everyone prefers the 96 kHz recordings. Going one step further, the reason they prefer the recordings is not because steady-state tones from 26 kHz to 48 kHz can be represented, but probably because some kind of time-domain resolution between the left and right ear signals is more accurately preserved at 96 kHz.

I have taken hi-res audio that I have recorded in the studio, and made 48k, 44.1k, MP3 and AC-3 down-sampled conversions of it. Even my friends with 'lead' ears are able to tell the difference. True, they can't tell me what they are hearing, but they prefer the Redbook audio over the MP3/AC-3, and prefer the hi-res(192k/24bit) over the PCM down-sampled versions.

That being said, vinyl is able to reproduce l/r timings at least equal to the HD PCM, if not greater. Of course, this requires the TT setup to be rather exact, and to not pass through any lo-res digital processing on it's way to your ears. I'm in the process of converting my entire vinyl library to 192k/24bit audio, to play back via DVD-Audio.

Also remember, that once a signal is digitized, the resolution(both frequency and temporal) have now been quantized. Simply putting that signal onto a piece of vinyl isn't going to magically recover any of the lost information. Of course we can't forget the mastering process either. Typically, different masters are made for the Redbook and vinyl versions. Vinyl requires very specific eq to be cut and track properly. Also in the mix is the RIAA phono eq curve. The reason I bring this up, is that unless you were involved in the mastering, you can't be sure what changes were made for the different releases. The only way to perform a true A/B listening test to judge the technical merits, is if you are intimately involved through the entire process of recording and mastering/pressing. Now subjective judging, is easily performed.

Which do you like better? That's all it really boils down to. What is your preference? The digital vs analog debate will rage for a very, very long time, but it still comes back to 'what do you like best?'

So... I'm done. Anyone else want this soapbox?

And you're wrong.



Hmmm... you seem to be under the impression that subsample time delays can't be represented in a discrete time signal. You're wrong. Yours the sort of ostensibly 'common sense' reasoning by analog champions, that digital audio confounds again and again. It's been debunked before. Digital audio is often counterintuitive. So work your way through this thread on 'time resolution' and you'll be enlightened.



http://www.hydrogenaudio.org/forums/...howtopic=49043


As for the rest of your post, the bottom line facts that would be readily demonstrable under scientific/bench test conditions, are, 1) LP is objectively more distorting in every dimension of the audible spectrum than CD , 2) CD rates and resolutions can, objectively speaking, fully and accurately capture any LP output in the audible range; 3) sighted comparisons between sample rates and formats are useless as indicators of relative difference or quality.