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

[DrWho]

No contest if the best of each medium is used given that the specs of 16/44.1 cannot match those of the best of analog

I recall a recent experiment where an entirely analog system was presented with a ADC to DAC conversion inline with the output from a preamp and could be bypassed with a switch setup for perfectly volume matched comparison. The study was done in the living room of some crazed "golden ear audiophile" who was extremely adament about the destruction digital would impart on his system...

Nobody at the gathering was even able to identify when the digital system was engaged. The digital stage was running at 44.1kHz and the Bitrate was adjustable....they got down to around 10-Bits before anyone could identify when it was switched in. One of the caveats was the gain through the ADC/DAC had to be optimized...and was the main point of the article.

Anyways, the point is that redbook audio should be more than adequate to be unidentifiable if properly implemented. I would welcome anyone that wants to swing by a studio to come by for a demonstration. I think we might still have some tape and a reel to reel laying around...and then we can do straight up comarisons against the original sonic event (the way these comparisons should be done).

[DrWho]

Oh, did anyone notice the fan off to the side of the right speaker? I don't recall reading any comments regarding the impact that it had on the sound quality...

[Mallette]

OK, Mike, I was trying to be a bit diplomatic rather than scientific. However, I DID mention the piano recording I did at 16/44.1 which has yet to be judged by anyone of any persuasion who has heard it as not superior to any released vinyl of piano.

I remain both mystified by this as well as pleased. OTOH, I also remain convinced that I am still quite able to hear significant superiorty from the higher resolution digital formats, both PCM and DSD.

In any event, my only issue with such thing is why anyone rejects music that sounds good, regardless of format.

There are 10s of thousands of perfectly dreadfuly records, and the number of equally dreadful digital releases continues to grow. Quit blameing the glass and start blameing the vintner.

Dave

[maxg]

"Only one out of 10 of my lp's is good. Only one of a 100 is great, and one of a thousand is awesome. "

Oh good- at least I buy vinyl better than you. []

Although it isnt always easy to asses I would guess that of my collection:

95% is better than CD (maybe more - not less).

60% is really good (forget its playing and just listen to the music).

About a third of the above is superb (stuff you would use if an audiophile drops in to listen).

And around 20% of those are jaw dropping.

Translation: In 100 records, 95 are better than CD, 60 of those are good, 20 are great and 4 are astonishing.

[maxg]

Forgot about the piano:

Rachmaninov
Piano Concerto 3
Vladimir Ashkenazy
London Symphony / Andre Previn
London
KIJC-9204
Stereo
Japan
Super Analogue

Its about as good as it gets.

[Mallette]

Max:
I said "my" collection and should have said "in general." My own collection fares significantly better than I said, though it goes back 30 years to well before I got really picky so it's nowhere near your levels. I would stand by my statement when applied to releases as a whole.

I'll have to check out the piano you referenced. I have many of the greats and not a one would fool my cat. The best are well recorded but there is always some vestige of surface noise that ruins the illusion. There is no room for noise of any kind in a piano recording.

May be a while, especially if that is not a current release and I have to send out bots looking for one, but I'll post when I've had a chance to hear it.

Dave

[skonopa]

In any event, my only issue with such thing is why anyone rejects music that sounds good, regardless of format.

There are 10s of thousands of perfectly dreadfuly records, and the number of equally dreadful digital releases continues to grow. Quit blameing the glass and start blameing the vintner.

Dave

A-friggan-men! I guess I never really "got" all these silly Analogue vs Digital arguments. I guess, to me, it has always been about the music, thus I was never really that anal about the particular medium it is on or if one was "better" than the other. As far as I am concerned, if it sounds good on my own rig, and I am enjoying what I am hearing then I am friggan happy as a lark. Yes, I even (gasp []) listen to MP3s (mostly eMusic downloads) on my system as well. I guess I'll be crucified by the snooty, anal-retentive, "audiophile" community for that one! [6]

And for the record, those loud heavy metal concerts has not killed my hearing (I use good quality Etymotic earplugs when I go to them. Stupid me went to one without earplugs once; never will make that mistake again!), as, even at 39, I can still hear those annoying 14khz ringtones some of these kids sometimes use on thier cellphones (one guy played one while we were in Atlanta asking if we could hear it. Me and one of my friends, also the same age, told him to turn that d@mn thing off! It was annoying as hell). So yes, I do still have perfectly good ears and can certainly appreciate good music on a well recorded pressing when I hear it.

[Mallette]

>...even at 39, I can still hear those annoying 14khz ringtones some of these kids sometimes use on thier cellphones...

Interesting. When I was 21 I could hear out to 22khz. I first experienced it by purchaseing a "silent" dog whistle when I was perhaps 13. I disposed of it disappointed that it was not only not "silent," but loud and annoying. I assumed everyone could hear it (except my stupid dog!). I did not realize the implications, however, until I was at Fort Benjamin Harrison in AIT and paid a visit to the local art museum. I was making my way with a buddy through a line when suddenly it felt as though someone had shoved an icepick into my ears. I ducked and rolled and everyone looked at me as though I'd taken leave of my senses. The guard was momentarlily startled, then said "Oh, another one." I asked what he meant and he pointed up at a transducer on the cieling "Silent burglar alarm," says he, "we just disarm the annunciator, but that thing stays on and radiates 22khz at 130db. Perhaps one of a thousand through here can hear it." I experienced this several times in banks and such, but at least I knew what was happening.

I am now 59 and the last time I "self tested" one ear could hear a 15khz sine wave, though barely, and the other could hear a 15khz warble tone (barely) but nothing with the sine wave. Probably even worse now, but I believe there is some acoustic brain training that critical listeners develop that somehow still can sense the presense and quality of higher freqs even you if don't really hear it.

Dave

[Mallette]

Not talking to myself here (I hope), but this is more on topic and unrelated to the last one so I am making it separate.

In a discussion over at computeraudiphile.com I have really added a lot to my own knowledge and experience in matters digital. I felt this post was relevant here so I quote myself:

"It would appear you might disagree with my statement about the Panny's "clinical like SS but musical like tubes" statement. However, I am not so sure. It MAY be that the chip amp is simply accurate and that SS is simply harsh.

I don't know. I DO know that I was thrilled to get my first SS amp back in 1970 to replace my Dyna SCA-30, and then equally thrilled to get back to a Dyna SCA-30 that came my way about 10 years later. It was like the first time I played an LP after a steady diet of CD's for nearly ten years. Revelation that I had been being musically starved.

It was that later revelation that led me down the path of studying digital recording techniques to find out why so little of it nourished the soul like fine analog (from any source).

This thread has confirmed most of my own findings as well as added to them.

I can now confidently state that, at least to my own satisfaction:

1. The FORMAT of any recording is irrelevant.
2. The ENGINEERING of any recording is all important.
3. Neither of the above matter if the playback chain is not right.

So, if you don't like the wine, don't blame the glass, blame the vintner."

That is where my head is now. Digital vs. Analog comparisions are like whether a great Bordeaux is better from Waterford or Crystal de Arque.

Dave

[Mallette]

Typical, Mark. I go on for a couple hundred words, and you sum it up in two sentences.

I MAY be mistaken, but I think you just had the last word here.

Dave

[DrWho]

there ain't no mo'.

Yea there is... []

EDIT: Just to expand, we know more about the original sound before it became digital, and can reproduce every sound with"infinite resolution" as long as it is below a given frequency (for lack of a better description).

In other words, compare the input and output sinewave going into a digital system and they will be identical within the bounds of the noise floors.

[skaloumbakas]

Years ago, I had the listening experience with the following CD's (I supose you have heard them...) and not with my current analogue set-up...

I had those records in CD's and when came in my hands the vinyl pressings, I imediatelly exchanged them with the vinyls...

They are 16 bit - linear cuts, as I read on the records' cover technical notes... I have an explanation why this is happening... but I would prefer to ask if any other fellow had the same experience with me with digitally remastered LPs...

[tigerwoodKhorns]

What's the finest resolution you can have with a Redbook CD?

What is the finest resolution you can have with an analog LP?

My CD Player and my DAC both upsample. I also find that the enginering seems to matter most.

[DrWho]

Upsampling is unrelated to native resolution. Upsampling is part of a filtering technique, that's all.

Agreed

Why did film manufacturers constantly seek finer grained film?

Why do digital camera makers constantly seek to increase "megapixels?"

Why do publishers seek finer resolution printing presses?

I would say because our eyes can see the better resolution.

In the case of digital audio, I don't know of anyone that can hear a 96dB noise floor....in fact, the resolution of digital is already better than most analog resolution. []

Those claiming that analog (especially vinyl) has "infinite" or "better resolution" than digital are missing the fact that the resolution is not better. As a parallel analogy, take a line....it can be defined by two points. Defining it with more than one point does not improve the line's definition...it's still one line that is equally defined by two points. Just because analog could be argued to have a lot of points defining a line, does not mean that line is better defined. In the case of vinyl where the noise floor is higher than digital, the line definition is actually worse...because all the points trying to define a single line are defining many different lines (aka, noise).

[DrWho]

A "line" is not a parallel analogy for
resolution, which is a real world phenomenon. A line is a mathematical
abstract: A line can be described as an ideal zero-width, infinitely long, perfectly straight curve, containing an infinite number of points. It occupies no real space.

Ok, that's great and all, but totally irrelevant.

A single line is completely defined and unique with only two points to describe it...the implication being that the resolution
of the description for that line is no worse with 2 points defining it
versus a million or infinite number of points defining it. In the case
of discrete systems, you're not losing any resolution if you choose
your finite number of points correctly.

As you mention, there
is always error associated with discrete systems, but in the case of
digital audio, that error (noise floor) is 96dB....how is that any
worse than a system that is accurate to only 60dB?[^o)]

Would CDs have been better if the RedBook spec was 32-bit and 100kHz sample rate?

Better
perceived quality or better numbers on paper? Why not pick other
meaningless numbers and go overboard with them? Like how about a 10
billion dB noise floor? Or how about 1Hz to 5Ghz
+-0.000000000000000001dB frequency response? Better on paper or better
in perception? Good proper engineering only cares about the
perception...and is likely why redbook hasn't been pushed to a next
level (it's not like current technology couldn't pull it off).

The
only thing extra bit-depth gains you is more headroom for improperly
using digital. A higher sampling rate only yields cheaper filters. But
if you want to be as obnoxious as the rest of the audiophile world,
then why not go all out with a topology instead of complaining about
stupid cheap implementations? That would be like me comparing an
automotive cassette deck to reel-to-reel...

[DrWho]

No hostility or belligerence over here....my apologies if it came across that way.

[ZAKO]

Thats my world..

[Mallette]

Mike, I can't quite follow your analogy between a line and audio or video resolution. Certainly a vector line has only two points and infinite resolution, but a bitmap line does not. Seems to me audio is more of a bitmap line, and the more points the higher the resolution.

Dave

[DrWho]

Mike, I can't quite follow your analogy between a line and audio or video resolution. Certainly a vector line has only two points and infinite resolution, but a bitmap line does not. Seems to me audio is more of a bitmap line, and the more points the higher the resolution.

I wasn't trying to imply the line had anything to do with audio directly....I am merely pointing out how a discrete system can have very good (sometimes perfect) resolution.

In the case of audio, we've got discrete time and discrete amplitude. It's fairly straightforward math to know the resolution limits imposed by going discrete in those two dimensions. It gets a bit more complicated when you take into account that the process of converting between the continuous and discrete world isn't perfect. In other words, when your discrete representation isn't what it thinks it is, you run into problems. Distortion imparted by a DAC, or an inconsistent clock (jitter) would be perfect examples.

[DrWho]

Here's a link
to the Meyers/Moran AES paper of last year wherein the claim is that
listeners can not tell when a standard 16-bit/44.1 A/D/A loop is
inserted into a higher bit rate, higher sample rate audio stream, such
as from a SACD or DVD-A player. Does this prove something interesting
about digital audio, or about AB/X testing?

I think one of the reasons that inline A/D/A conversions are
inaudible has mostly to do with the fact that the A/D and D/A are being
driven directly from the same clock and the trace lengths are very
short.

I also think higher sampling rates tend
to sound cleaner because the low pass filters don't have to be as
steep....which means less ripple through the passband. The amplitude of
the ripple isn't so much the issue as it is the ghosting that it
creates. However, that doesn't mean good performance can't be achieved at redbook rates...it's just more expensive.



The point being that the discreteness of redbook should be able to
exceed the capabilities of human perception....it's just hard to
realize that potential in the real world.