Why CDs May Actually Sound Better Than Vinyl

[Al M.]

Al, when Peter takes me to visit you, before we get to drinking, you will have to try to explain this to me, because I do not understand this at all. (And this post does not even make any sense to me.)

1) To plagiarize an earlier post in this thread (Peter A, I think) if the dCS gears allows you to conclude that digital theory, properly implemented, gets you a perfect reproduction of the musical waveform (and not as I think an approximation thereof) then how can digital ever be improved upon from here?

Ron,

as I had stated, the sound might have been even better, but it was convincing enough. I had had real worries that there might be some unresolvable issues with Nyquist sampling after all, but on the most critical timbres that I was listening for I found proper believability. So no, I did not claim the sound to have been perfect, but it was convincing enough for me to to deem it audible proof that the underlying theory really was perfect, i.e. flawless, otherwise I might still have heard substantial and worrisome issues with the most critical timbres that just wouldn't go away. I didn't. Certainly not compared to analog reproduction in great systems (of course nothing quite approaches the sound of live music, for multiple reasons.)

2) A higher sampling rate must get digital closer to reproducing a complete waveform. How can it not?

As long as sampling rate is at double the frequency, the waveform is represented completely. Thus, at 20 kHz frequency the waveform is represented as completely by a 44.1 kHz sampling rate as it is by a 192 kHz sampling rate. I suggest you watch again the video in the link that I already posted earlier, it shows the 20 kHz sine wave on the oscilloscope:

https://www.xiph.org/video/vid2.shtml

Only when you go in frequency beyond half a given sampling rate, that frequency cannot be properly re-presented anymore by that sampling rate. For example, a 30 kHz tone cannot properly be represented anymore by the CD sampling rate of 44.1 kHz, but it falls comfortably within the range of representation by a 192 kHz sampling rate. But we cannot hear above 20 kHz.

Aah, but what about a tone that is not a sine wave? Amir explains it very well here:

http://www.whatsbestforum.com/showt...ral-resolution&p=334892&viewfull=1#post334892

Why can sometimes high-res sound better? For reasons of practical implementation. It's just easier with more data points along the curve, but not theoretically necessary. The dCS gear did just fine on just Redbook CD, which sounded more convincing and real than hi-res on other digital playback.

PS: I look forward to your visit!

 

[FrantzM]

Hi

Let us try not to make ad abusurdo and ad hominem devolution of this discussion. A theorem needs to be demonstrated. A demonstartion remains a mthematical exercise. The theorem physical applications raise it above speculation, it become a fact. Nyquist Theorem is applied and is beyond speculation. it is a fact proven by our phones, our video and in that case by CD...

We go on saying that analog simply copies reality. It does not, it is a reconstruction as much as digital.. You take sound waves then transform it into electricity and then do the reverse. A lot is lost .. Reconstruction by any other names , let's be nice and call it so. Digital does a better job. A lot less is lost, that some people like analog it better is fine , some, again prefer Tang to Orange Juice and not to make it too much of the artificiality of Tang, for many people bottled , processed OJ is what they know to be Orange Juice. When confronted with the Real freshly squeezed OJ some don't like it.. And that is Ok too when the facts of the matter come into play, that bottled commercial same-taste-year-after-year-processed OJ is far from the juices of Oranges... Some of us like and may prefer it . The fact remains that Fresh Squeezed is a better construct, a better approximation of the orange juices.
It would be nice if some would not call something superior because they like better. Let' s just start using the less confrontational prefer.

 

[Mike Lavigne]

Pending Al's effort to disabuse me of my misunderstanding of digital theory, this is what I believe, too.

over many, many, years of listening to the highest levels of digital and vinyl, digital theory falls short of reality. something gets lost in the translation to art. particularly when you hear how completely perfect redbook was suppose to be (and is now claimed to be in this thread), and then you hear how much better higher rez PCM sounds. it's either perfectly fine, or some short of that level.....not both.

how can dxd sound better than what was suppose to be 'perfect'? the answer, of course, is that all this theory is just that.

I just invested huge dollars in the Trinity dac to give redbook it's best chance at being perfect. I got convinced it can be without the intrinsic PCM'ness of edginess and come up to the level of dsd. but all they way there to analog? no.

and this is with thousands of PCM sourced recordings of various resolutions. most of which sounds wonderful, if far short of analog. clearly higher rez sourced recordings better than lower rez sourced recordings; using my analog as the reference. so redbook, even native, is not the final answer.

I just don't rely on math for my music.

 

[Fiddle Faddle]

A higher sampling rate must get digital closer to reproducing a complete waveform. How can it not?

Just from a practical perspective rather than a theoretical one, the higher the sampling rate, the higher the demand is on accurate clocking. I have rarely heard consumer level equipment, for example, that can reproduce bowed acoustic instruments better at 192 kHz than it can at 96 kHz. The only equipment I have heard that does this better is either incredibly high quality to begin with and / or uses external clocking hardware to achieve near-perfect time domain stability. Then there is the other issue of very high sampling rates reproducing HF noise that would not be reproduced at lower sampling rates. That noise would obviously not be directly audible in itself but may effect the sonic integrity of what we do hear.

You will find some engineers prefer to stick by 96 KHz as the ideal sampling rate (i.e Abbey Rd) even though obviously much higher rates and multiple-rate DSD are alternatives. I do recall reading an article a few years ago that suggested something along the lines of 60 KHz was "ideal".

 

[Al M.]

We go on saying that analog simply copies reality. It does not, it is a reconstruction as much as digital.. You take sound waves then transform it into electricity and then do the reverse. A lot is lost .. Reconstruction by any other names , let's be nice and call it so.

Well said. I have also heard the argument that the magnetic particles on analog tape are well, just particles, i.e. dots in time as the tape moves along the head that also represent a signal discontinuously. And vinyl usually is made from tape...

 

[Al M.]

Just from a practical perspective rather than a theoretical one, the higher the sampling rate, the higher the demand is on accurate clocking. I have rarely heard consumer level equipment, for example, that can reproduce bowed acoustic instruments better at 192 kHz than it can at 96 kHz. The only equipment I have heard that does this better is either incredibly high quality to begin with and / or uses external clocking hardware to achieve near-perfect time domain stability. Then there is the other issue of very high sampling rates reproducing HF noise that would not be reproduced at lower sampling rates. That noise would obviously not be directly audible in itself but may effect the sonic integrity of what we do hear.

You will find some engineers prefer to stick by 96 KHz as the ideal sampling rate (i.e Abbey Rd) even though obviously much higher rates and multiple-rate DSD are alternatives. I do recall reading an article a few years ago that suggested something along the lines of 60 KHz was "ideal".

Thanks for that post, very interesting!

 

[amirm]

I don't know what to make of your comment about Peter Breuninger's listening abilities. Perhaps the system he used to listen to that tape for his review was not as revealing as your system.

What then to make of any other evaluations he makes with that system when hiss and intermittent buzz was not audible to him??? I tend to think his system is fine. It is that like most people, he doesn't have critical listening abilities. I would bet any amount of money that I can go and listen to the same tape on his system and not only hear the same problems, but teach him to hear it too.

We really need to accept that we are bad listeners. I can give you data point after data point. Here is another.

In the last two pacific northwest society when demoing tape and NADAC, both presenters used fairly loud playback levels. This caused the left channel amplifier that was set to slightly higher gain to clip. This was plainly audible to me and so it was to Bruce and Gary. This was Gary's discontinued amplifier and as he explained, it just doesn't have enough power to fill that large room at those levels. Out of the entire group that was there, only one other person beside us heard that noise! Just one. The rest raved about what they heard, what nuances were revealed to them, etc. Yet, clear clipping and distortion in the left channel went inaudible to them. If we can tolerate that, then sure, any and all objective flaws that we can easily demonstrate and prove with LP likely is not audible to people.

So the notion that "in so many years LP has always sounded better to me" doesn't sit well with me. These faults are there beyond shadow of doubt. We can show them in measurements and demonstrate clearly that they should be audible. The article in OP provided ample evidence that the format is not transparent or anything close to it. And it not like it is fashionable to be that frank these days. Folks want to hear analog is better not the other way around.

In contrast with digital we can't demonstrate these objective flaws. Yet we want to be convincing that it has more issues to our ears where we can't hear orders of magnitude more distortions in analog?

To me, it is that blindness to distortion that allows analog to sound so good to us. Combine good mastering with that and that is why so many people like LP. I too very much enjoy good LPs when I listen to them. Great music makes you melt in the experience and I am there. Until there is a pop or click and then I want to smash the thing to very little pieces .

 

[ddk]

Hi

Let us try not to make ad abusurdo and ad hominem devolution of this discussion. A theorem needs to be demonstrated. A demonstartion remains a mthematical exercise. The theorem physical applications raise it above speculation, it become a fact. Nyquist Theorem is applied and is beyond speculation. it is a fact proven by our phones, our video and in that case by CD...

The application of the theorem isn't in question what's contested is the claimed sonic superiority and accuracy.

We go on saying that analog simply copies reality.

No we don't!

It does not, it is a reconstruction as much as digital.

.

No its not, the electric signal is kept and transferred to other medium, in digital the same signal is digitized and later reconstructed, there's no equivalent A2D and D2A process in the analog realm.

A lot is lost .. Reconstruction by any other names , let's be nice and call it so

.

Actually loss & reconstruction are completely different you can't call one by the other, no matter how nicely we try!

Digital does a better job. A lot less is lost, that some people like analog it better is fine

How do you figure that Frantz? This is the part where you kicked "nice" to the curb! Have you quantified the losses of both formats and figured out a universal value system? What about the accuracy of the digital reconstruct, how's that been verified?

It would be nice if some would not call something superior because they like better. Let' s just start using the less confrontational prefer.

I'm afraid that boat has sailed Frantz, "Perfect Sound Forever" campaign and declaration of superior digital accuracy saw to it. I don't think anyone here is trying to argue anyone else's preference. Some of us prefer straight talk to any veiled PC reconstruct!

david

 

[FrantzM]

over many, many, years of listening to the highest levels of digital and vinyl, digital theory falls short of reality. something gets lost in the translation to art. particularly when you hear how completely perfect redbook was suppose to be (and is now claimed to be in this thread), and then you hear how much better higher rez PCM sounds. it's either perfectly fine, or some short of that level.....not both.

how can dxd sound better than what was suppose to be 'perfect'? the answer, of course, is that all this theory is just that.

I just invested huge dollars in the Trinity dac to give redbook it's best chance at being perfect. I got convinced it can be without the intrinsic PCM'ness of edginess and come up to the level of dsd. but all they way there to analog? no.

and this is with thousands of PCM sourced recordings of various resolutions. most of which sounds wonderful, if far short of analog. clearly higher rez sourced recordings better than lower rez sourced recordings; using my analog as the reference. so redbook, even native, is not the final answer.

I just don't rely on math for my music.

Mike

Those are subjective impressions and I have no doubt you sincerely believe so. Confronted with similar experiences I don't however share you views. One could say that my headphones-based system is not to the level of yours and that would be their right. I, for one would be one of those. It still doesn't make analog to be superior because you find it to be so. It is a preference as valid as mine but not more so.
We are past the perfect sound forever we know it was ill-advised and in hindsight premature. As for DSD I am one of those who don't care about it. It sounds nice but nothing in the way of more fidelity.. a nice patina which doesn't please me... I repeat me.
I haven't found Hirez to be systematically superior to Redbook, A different opinion. Enough to not care to purchase any more Hi-Rez until they , to my ears, trounce Redbbok. Is the Berkeley not optimized for Hirez?Could well be.. I have heard many other DACs among them the famed MSB TOL and some DCS can't remember which one and on Hi-Rez .. they were good but nothing to change my point of view. Again notice the regular use of "I".. as in "me"... Under the conditions of casual listening and with all the variables that this entails , I haven't been swayed.
It wasn't my opinion ten years ago that digital had arrived. It is now, IMHO, digital can match analog in most things that I care about in term of reproduction and in many that appeal to me surpass it. Plus once my digital is where it needs to be I can concentrate on what at this point in my life is paramount: Enjoying Great Music.

 

[PeterA]

Hi

Let us try not to make ad abusurdo and ad hominem devolution of this discussion. A theorem needs to be demonstrated. A demonstartion remains a mthematical exercise. The theorem physical applications raise it above speculation, it become a fact. Nyquist Theorem is applied and is beyond speculation. it is a fact proven by our phones, our video and in that case by CD...

We go on saying that analog simply copies reality. It does not, it is a reconstruction as much as digital.. You take sound waves then transform it into electricity and then do the reverse. A lot is lost .. Reconstruction by any other names , let's be nice and call it so. Digital does a better job. A lot less is lost, that some people like analog it better is fine , some, again prefer Tang to Orange Juice and not to make it too much of the artificiality of Tang, for many people bottled , processed OJ is what they know to be Orange Juice. When confronted with the Real freshly squeezed OJ some don't like it.. And that is Ok too when the facts of the matter come into play, that bottled commercial same-taste-year-after-year-processed OJ is far from the juices of Oranges... Some of us like and may prefer it . The fact remains that Fresh Squeezed is a better construct, a better approximation of the orange juices.
It would be nice if some would not call something superior because they like better. Let' s just start using the less confrontational prefer.

All in good fun and no offense intended....

Fresh squeezed orange juice: Live unamplified music in a real space
Bottled commercial same-taste-year after year processed OJ: analog reproduction
Add pulp to this commercial processed OJ: Master tape or direct to disk 45 rpm LP
Frozen OJ concentrate: Quad DSD, dCS current models
Tang: Typical Digital

 

[Al M.]

I haven't found Hirez to be systematically superior to Redbook, A different opinion. Enough to not care to purchase any more Hi-Rez until they , to my ears, trounce Redbook.

I have also repeatedly heard that recording quality and mastering both trounce format, i.e. a great recording on Redbook will definitely sound better than a mediocre one on hi-res (that is also my limited experience). Also, the mastering for hi-res is often different, so then differences in sound quality may not necessarily be related to the format either.

Is the Berkeley not optimized for Hirez?Could well be..

The Berkeley definitely is intended to be optimized for that, since the Berkeley designers find hi-res digital the absolute best recording format, i.e. also superior to analog tape.

 

[Mike Lavigne]

Mike

Those are subjective impressions and I have no doubt you sincerely believe so. Confronted with similar experiences I don't however share you views. One could say that my headphones-based system is not to the level of yours and that would be their right. I, for one would be one of those. It still doesn't make analog to be superior because you find it to be so. It is a preference as valid as mine but not more so.
We are past the perfect sound forever we know it was ill-advised and in hindsight premature. As for DSD I am one of those who don't care about it. It sounds nice but nothing in the way of more fidelity.. a nice patina which doesn't please me... I repeat me.
I haven't found Hirez to be systematically superior to Redbook, A different opinion. Enough to not care to purchase any more Hi-Rez until they , to my ears, trounce Redbbok. Is the Berkeley not optimized for Hirez?Could well be.. I have heard many other DACs among them the famed MSB TOL and some DCS can't remember which one and on Hi-Rez .. they were good but nothing to change my point of view. Again notice the regular use of "I".. as in "me"... Under the conditions of casual listening and with all the variables that this entails , I haven't been swayed.
It wasn't my opinion ten years ago that digital had arrived. It is now, IMHO, digital can match analog in most things that I care about in term of reproduction and in many that appeal to me surpass it. Plus once my digital is where it needs to be I can concentrate on what at this point in my life is paramount: Enjoying Great Music.

Frantz,

i have made the investment and done the work to be able to know where things are.

how good can redbook sound? top level interconnects? Herzan TS-150 anti-vibration shelf? full tilt boggie server, linear power supplies. 7 terabytes of PCM.

is the system up to showing differences? i think so.

i was not fooling around. subjective? sure....i did use my ears. but i put my money and effort where my mouth was, and took a no holds bared approach.

and redbook sounded great but fell far short of both high rez PCM and farther short of analog.

case closed.

 

[Andrew Stenhouse]

Amir, you may be onto something with this comment in bold. For some time now I have heard what I describe as digital glare, to a greater or lesser degree, from every system that has a digital source component. Some people, but not all of them, and rarely the owner, hears this artifact. Some do hear it, but it does not bother them the way it does me. If it is bad enough, I quickly get fatigued and cut the session short. Interestingly, I have not hear this artifact in my recent exposure to three specific digital players/DACs: the dCS Vivaldi, the Rossini, and the Nadac (only with quad DSD). These three units had remarkable levels of resolution with absolutely no sense of fatigue.

I can not explain why some other listeners in the room could not hear this same distortion when I described it to them. Some could, but others could not.

I don't know what to make of your comment about Peter Breuninger's listening abilities. Perhaps the system he used to listen to that tape for his review was not as revealing as your system.

Spot on. Not ever one hears the same or has the same sensitivity. In my circles we speak of having 'a good ear' for music which broadly means an deep understanding of how natural instruments sound and how close the facsimile we listen to approximates them.

If I may, listen if you get a chance to the Pathos Endorphin, another CD player that sounds "right' to my ears, at least with classical.

I cringed when I listened to pcm through the Nadac - it was the usual digital I dislike with a passion until I got to 24/172 - DXD, the latter was when I found pcm indistinguishable to DSD 256. I didn't mind pcm at all when converted to DSD 256 by HQP, but felt native DSD 128 and DSD 256 were better again (having the qualities you describe).

But of course you were experiencing the Nadac through a system many, many times more resolving than mine, so the differences would have been much more apparent than they were to me.

Still, it wan't subtle. It is an interesting discussion this - what we all hear, and how that accounts for the debates we have.

Mike

It still doesn't make analog to be superior because you find it to be so.

Ahhh, there I agree with Mike. Yes it does, because ultimately the litmus test has to be what our ears tell us not what our brain instructs us should be the case. But that is the difference in a nutshell between engineers and non - engineers, or objectivists and subjectivists. I haven't met an engineer yet that is comfortable accepting a proposition without a rationale explanation.

 

[ddk]

In all honesty I'm trying to figure out what you're debating with me in this post Al. My argument is the claimed superiority of the digital format in retention of original signal and its attributes over analog. We're not even debating the difference in or preference for sonic qualities of the formats and agree on CD playback vs hi rez computer source. If you can at some point compare a 44 recording to higher sampling rates 192 and above, there is an audible difference.

david

Ok, you have stated your opinions, David, and I'll leave it mostly at that. Yet let me answer the issue about sampling rate.

It has been claimed that the CD sampling rate was a (significant) compromise. Yet I am afraid this is engaging in revisionist history from current hindsight and audiophile consensus. Yes, it may be that at the origin of digital audio some engineers said a much higher sampling rate is needed, but that does not appear to have been the consensus. In fact, the consensus appears to have been more or less Nyquist (double sampling rate over frequency range presented), and still is to a large part (outside of audiophile circles), see AES guidelines below.

From:
Telarc, Frederick Fennell, and an Overture to Digital Recording

Telarc, founded in Cleveland, Ohio, in 1977 by Jack Renner and Robert Woods, both of whom were classically trained musicians and educators, made its first two recordings in the then-typical direct-to-disc format. At the same time, Renner and Woods were inspired by the new digital recording technology of Tom Stockham’s Salt Lake City-based Soundstream, Inc., the first commercial digital recording company in the United States. Stockham, whom Renner calls “the father of digital signal processing,” had developed a 16-bit digital audio recorder using a high speed instrument magnetic tape recorder and demonstrated the recordings at the fall of 1976 AES convention. Renner and Woods formed a partnership with Stockham. They requested that he increase his digital system’s high frequency response, from 17 kHz to 22.5 kHz at a sampling rate of 50 kHz, an unprecedented level.

As you can see, Nyquist (obviously with a few extra kHz sampling space to allow for filtering).

From: https://en.wikipedia.org/wiki/44,100_Hz

In the early 1980s, a 32 kHz sampling rate was used in broadcast (esp. in UK and Japan), because this was sufficient for FM stereo broadcasts, which had 15 kHz bandwidth.

Again, Nyquist.

The current AES recommended practice for professional audio is 48 kHz, which is not too far from 44.1 kHz (2008, revised 2013):

Abstract: A sampling frequency of 48 kHz is recommended for the origination, processing, and interchange of audio programs employing pulse-code modulation. Recognition is also given to the use of a 44.1-kHz sampling frequency related to certain consumer digital applications, the use of a 32-kHz sampling frequency for transmission-related applications, and the use of a 96-kHz sampling frequency for applications requiring a higher bandwidth or more relaxed anti-alias filtering. This revision further quantifies the preferred choices for higher sampling frequencies. (8 pages)

Again, basically Nyquist at 48 kHz, with a few extra kHz sampling space to allow for filtering.

Yes, there may have been a debate about 48 kHz vs. 44.1 kHz even in the beginning (and 44.1 kHz won because of available video equipment for recording), but here we are not talking about vastly different sampling rates -- that was not a 192 kHz vs. 44.1 kHz discussion, not by a long shot.

***

And again, given my experience on the dCS gear with plain 44.1 kHz Redbook CD that delivered sonically superior results than the hi-res that I have heard elsewhere, and very life-like timbral believability, it has become clear to me that Nyquist must be correct. It is all about the practical implementation to make the approach to theory as close as possible, and that implementation is superior on the dCS gear, to my ears.

Higher sampling rates do not appear to be strictly necessary. These are just a practical vehicle to more easily approach theoretical perfection, but not required by theory.

 

[BruceD]

+1

And frankly coming from a person who used to do photography on a 4 x5 ... Digital is far better than anything anlog photography has ever dreamed of doing .. We are at a point where a smartphone takes incredible pictures.. an iPad has HDR ...

@everyonel

Yes firstly I am also "proud" of the transformation to balanced and forthright discussion that has prevailed over the last weeks on WBF--especially this thread --the ability to show preference without the incursive dissent that plagued earlier threads is welcoming and indeed makes for pleasurable

and honest discussion

I too applaud Steve and all who strove to make WBF right itself to its even keel

Yes Frantz re you analogy of the Digital Photography in this case --you have a good point--while I agree Digital snapping has great aspects over Analog (Film) with the stop latitude up to 9 stops over film last time I looked!

the one fact remains in my book one of the deciding factors holding back Digital Photography is still the fact that no chip in any Digital Camera can out put Raw files larger than 300DPi-- surely this must enlarge at some stage in the near future

look at the ranges of Digital Audio--not so in the camera modes though.

the only "upside" to Film is the fact that it can be scanned to 8000 dpi from the base product--maybe more today--its awhile since I had my Imacon.

Sorry for the transgression

I run both CD (Redbook only ) and Vinyl and am in the camp of both parties--dependant on how the mood strikes!

Carry on chaps--great Thread

BruceD

 

[Al M.]

I cringed when I listened to pcm through the Nadac

Me too, it had this weird glassy hardness of sound. That went away with DSD, native or upsampled via HQPlayer from PCM. But then, it is known that this type of DAC (Sabre DAC) performs better on DSD.

 

[FrantzM]

Frantz,

i have made the investment and done the work to be able to know where things are.

how good can redbook sound? top level interconnects? Herzan TS-150 anti-vibration shelf? full tilt boggie server, linear power supplies. 7 terabytes of PCM.

is the system up to showing differences? i think so.

i was not fooling around. subjective? sure....i did use my ears. but i put my money and effort where my mouth was, and took a no holds bared approach.

and redbook sounded great but fell far short of both high rez PCM and farther short of analog.

case closed.

To you Mike..

It remains an opinion regardless of the amount of money invested. And this is fine.

 

[FrantzM]

All in good fun and no offense intended....

Fresh squeezed orange juice: Live unamplified music in a real space
Bottled commercial same-taste-year after year processed OJ: analog reproduction
Add pulp to this commercial processed OJ: Master tape or direct to disk 45 rpm LP
Frozen OJ concentrate: Quad DSD, dCS current models
Tang: Typical Digital

Love it

 

[Al M.]

In all honesty I'm trying to figure out what you're debating with me in this post Al.

Very simple, David, your original insinuation (emphasis added):

As there are with digital! How much is left on the so called cutting floor when you convert the analog signal to digital? How is lost to down sampling to 16/44? Is 16/44 the ideal signal for our ears to evaluate or is it a commercial compromise? We already know that higher rates sound better to our ears.

 

[FrantzM]

I am off for now I will go listen to some Redbook PCM , Arvo Part's "In Principio", then "Tabula Rasa "and Krzysztof Penderecki's String Quartet No. 3.. Find myself a little bit adventurous and will follow on a piece I have never cared for until recently when Roon told me I had it Karlheinz Stockhausen's "Gruppen for 3 Orchestras"..

Arvo Part works I like a lot. it requires for me not no intellectual effort, flows with my mind. So does Penderecki's works especially his chamber peices, Stockhausen is different, I have to take something out of what I expect from classic Classical Music. He's a study and truly a discovery .. Acquired taste? maybe and that thanks to digital