Why CDs May Actually Sound Better Than Vinyl

[Andrew Stenhouse]

Well, if I may, it is all a commercial compromise in the sense that what we hear at home is only ever what is given to us in edited form: I am with FF - format is much less important than what is done with the tape/digital recording in the editing process. And it isn't all bad - editing is actually very necessary for anything above a couple of guitars and some voices.

In 2007 I had a long and (at times) unpleasant debate on Stereonet on accuracy vs musicality. The proposition put forward was if something was accurate it was, by that fact, musical. My position based on my listening experience of the system used to support the proposition was just because a system was "accurate' to the recording mic (and how exactly that could be determined was another question entirely), it didn't make it musical ie emotionally engaging.

The system listened to was astounding - it really was extraordinary in what it could do - but it sounded utterly unconvincing to me as a reproduction of a musical event. And to another set of ears I trust - a former professional violinist with the SSO. It was reminiscent of early dCS, regardless of whether it was digital or analogue being played.

I digress, but perhaps we are speaking of the same thing, in a different way.

 

[ddk]

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

The only claim I made is that there are losses with digital formats too, are you saying that there aren't any? The rest are unanswered questions challenging the superiority claims.

david

 

[Al M.]

The only claim I made is that there are losses with digital formats too, are you saying that there aren't any?

Perhaps there are, but engineers were fed up with the losses from the analog chain, which is why they went for digital, which they judged to be more faithful to the original signal. Along these lines, I may remind you of the following post on this thread by someone in the business of recording himself:

It is not the "analogueness" which is at issue here but the accuracy of the "transport" mechanism. The sound picked up by the microphones is pressure fluctuations in the air. There are both analogue and digital ways of recording the electrical output of the microphones and storing it. These methods allow later reproduction of something like the original pressure fluctuations in the air by decoding the recording and powering loudspeakers.
As a recording and reproduction method digital will certainly both store a more accurate recording of the original electrical output of the microphone, and certainly reproduce it more accurately, even than analogue tape which is much more accurate than LP can ever possibly be.
If the output sounds more "real" to somebody that is not because it is a better reproduction but because the inaccuracies in reproduction appeal to the taste of the listener, which is fair enough.

In the most exigent data recording cases I have experienced no analogue method was good enough to get, store and then reproduce the data well enough to be much use at all. Digital recording then reproduction were the only way that anything useful was learned.

 

[microstrip]

Describe, not recreate! Not all theories are proven either!
(...) david

IMHO most of these debates about perfect digital forget the implementation. In reality the imperfections are already encoded in the digital stream. Then we add imperfections in the digital playback. Are we supposed to believe it is perfect enough just looking at Stereophile DAC measurements?

In reality we still can not correlate our standard measurements with digital sound quality. Each of us is presenting his preference for some digital equipment and we have no "theory" why it sounds like we are describing.

Some people in the digital business openly admit they can not understand it completely - I remember reading the DCS documentation that DCS admits that they do not understand why the Upsampler used between the transport and DAC improves the sound quality. But as they say when listening people have found it really improves!

 

[Al M.]

IMHO most of these debates about perfect digital forget the implementation. In reality the imperfections are already encoded in the digital stream. Then we add imperfections in the digital playback. Are we supposed to believe it is perfect enough just looking at Stereophile DAC measurements?

Precisely. I have repeatedly made the distinction here between the theory, which I believe to be correct, and imperfect practical implementation, but that distinction is often lost among digital proponents.

Conversely, digital opponents attack the theory just based on the audible results which they judge inferior. It's not the theory folks, it's the practical implementation which is the problem.

In reality we still can not correlate our standard measurements with digital sound quality. Each of us is presenting his preference for some digital equipment and we have no "theory" why it sounds like we are describing.

Some people in the digital business openly admit they can not understand it completely - I remember reading the DCS documentation that DCS admits that they do not understand why the Upsampler used between the transport and DAC improves the sound quality. But as they say when listening people have found it really improves!

And the folks at dCS are probably among the engineers that understand digital best...They are at least honest, for sure, as I also pointed out earlier in the context of them talking about filter settings and the shortcomings vs. advantages of each.

 

[amirm]

This one is I am afraid or practically the entire world around you would cease to work.

??? How so Amir?

I will give you another example. The radio in your cell phone is completely digital. It digitizes the incoming RF band using a very high-speed ADC and completely demodulates the signal in software. It is called Software Defined Radio. It is able to do the job more reliably, more flexibly (hence the ability to work on multiple networks), and stably. The output is analog audio you hear. If sampling theory did not work, your cellphone would stop working. Or underperform analog which it does not.

Highschool Physics, says otherwise.

High-school physics can indeed show the failings of a physical format like LP. The faults are that elementary. AP math may be also most of what you need to understand sampling theorem.

But I will play. What physics are you referring to?

 

[Lee]

IMHO most of these debates about perfect digital forget the implementation. In reality the imperfections are already encoded in the digital stream. Then we add imperfections in the digital playback. Are we supposed to believe it is perfect enough just looking at Stereophile DAC measurements?

In reality we still can not correlate our standard measurements with digital sound quality. Each of us is presenting his preference for some digital equipment and we have no "theory" why it sounds like we are describing.

Some people in the digital business openly admit they can not understand it completely - I remember reading the DCS documentation that DCS admits that they do not understand why the Upsampler used between the transport and DAC improves the sound quality. But as they say when listening people have found it really improves!

This is exactly right. We are still discovering what is important in digital.

Think about jitter. Julian Dunn, rest his soul, thought the answer to the question "what minimal timing factor can our ears hear" was in single digit nanoseconds. Then papers were written around hundreds of picoseconds. Now we know that single digit picosecond timing errors are perceptible.

 

[Lee]

To paraphrase the incredibly sanctimonious Neil Degrasse Tyson:

"The thing about science is it is always changing, so we never know what is true."

Science is not about absolute truths. It is about experimenting, making observations, and refining/improving our understanding. Audio measurement is just another area where the truth is a moving target.

It's like THD on amps. There are a lot of half-truths out there.

 

[fas42]

I believe that particular respondent was refering to a legacy DCS line of a decade ago when using that descriptive , a period that I agree produced a somewhat cold and over analytical presentation from DCS products, you were pronouncing upon the latest Vivaldi line in a similar vein that would place you distinctly in the minority I would say.

Yes, amusing that it was Vivaldi playing on the Vivaldi?? - probably not! What was "terrible" was the string tone, it had the quality such that I would have had to get up and walk out if it kept going, it was irking me so much!

 

[Lee]

I think the article does not tell us anything new. I think it is a typically non-introspective exposition of the view that since CD measures better than vinyl then ipso facto CD sounds better than vinyl, and that if vinyl people choose to delude themselves into thinking otherwise, because they like nostalgia and the feel of LPs, they are simply wrong.

I'm not sure what article you are referring to Ron but this came after my reference to mixonline which is actually suggesting that in at least one area LP measures better than CD.

In any event, I think many digital era people believe CD measures better than LP but in fact it only measures better in some areas and in others LP measures better.

I also believe through my pro recording background that LP is really on par with 24/96+ dvd-audio resolution so I think a strong argument can be made that CD is inferior to LP.

However, given that the mastering plays such a big role in the final outcome, a true lover of fine sound should probably be format-neutral for the best listening.

 

[Lee]

If you think that digital is intrinsically incomplete or missing something, or only accurate 'at points in time', then I suggest you learn a bit more about sampling theory.

This video might help:

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

Xiph's videos have all sorts of mistakes and misrepresentations.

I have detailed these mistakes on another forum and it is tedious. It would be a mistake to put them up as any kind of an authoritative source.

 

[Argonaut]

Yes, amusing that it was Vivaldi playing on the Vivaldi?? - probably not! What was "terrible" was the string tone, it had the quality such that I would have had to get up and walk out if it kept going, it was irking me so much!

My apologies Frank should I be missing the obvious here, Are you saying that the irksome rendition that you are refering to was via a legacy DCS product, Or via the latest Vivaldi line ?

 

[SoundAndMotion]

Xiph's videos have all sorts of mistakes and misrepresentations.

I have detailed these mistakes on another forum and it is tedious.

I'm interested in seeing that. To avoid extra tedium here, can you give a link?

 

[fas42]

My apologies Frank should I be missing the obvious here, Are you saying that the irksome rendition that you are refering to was via a legacy DCS product, Or via the latest Vivaldi line ?

Certainly not the latest dCS stack, this was 10 plus years ago, I could track down what it was if I spent some time checking it out, if people desired - call it legacy for now :b !

The thread appears to be derailing just a touch at the moment - would be a shame if the temperature rose any more than what it is now. To ease things off a bit, I would make the point that something I've found to be key, is that at a certain level of refinement of individual components then the overall integrity of the system need to be sufficient - what some people would currently call "synergy". Unfortunately, this is more critical with digital than analogue, in part because the potential level of detail that can be revealed is greater with digital recordings - I have spent years dealing with these issues, know them well.

The analogy, since we are also talking photography , is having a lens with a huge aperture - the focus then needs to be precisely right, otherwise there is always a degree of blur - when the focus snaps into the absolutely right position then you see every fine detail, without effort - so it is for digital playback.

 

[Fiddle Faddle]

Science is not about absolute truths. It is about experimenting, making observations, and refining/improving our understanding.

And that is something that annoys me incredibly when I get involved in any sort of scientific discussion because, well, it's 2016. Just like we now know far more in 2016 than we did in 1916, in 2116 we will know a lot more than we do today! I've gotten into this style of debate with medical people from time to time, arguing that just because such and such a procedure may be state of the art, doesn't mean it is actually any good! But it may well be in another 50 or 100 years time. And infact some of those procedures are now being progressively abandoned as our scientific and medical understanding - together with empirical experience - matures.

I have already said numerous times then when audio measurement in the main correlates to what I actually hear, I will start to afford it due respect. But there are only certain measurements that do correlate and even those cannot predict what the final sound will be like. For example, you cannot scientifically predict how a certain jitter noise pattern is going to effect the actual audio program material playing some 110 plus dB above the worst of it. You still have to hear it to make that sort of judgement. You can show me two jitter noise floors across the audible 20to 20K spectrum and one will measure better than the other. That doesn't mean the one that measures better will sound better. It might have noise spikes at certain frequencies that are particularly offensive. Same reason that I can have two 16 bit shaped noise floors. One might measure fantastically between 1Khz and 8 KHz and another might measure worse there but better and less lumpy at low frequencies. Which one is worse? The only way to know is to listen to the 24 bit and 16 bit versions side by side and make a subjective judgement as to which is more faithful to the original. And even then the result is not a "given". It depends on the actual recording too, so you might get a different result each and every time.

 

[mauidan]

I've really enjoyed this thread, especially the posts by audiophiles that listen via headphones to DACs which regardless of the original sample rate, convert it to a datastream sampled at 110kHz, and have no idea what native sample rates sound like, and those with Redbook only transports that have no idea what HI-REZ PCM sounds like on their systems.

All we need now are comments by audiophiles with hearing aids who suffer from tinnitus.

 

[Rodney Gold]

The part of this thread that resonates with me is the lack of hearing accuity or lack of training re critical listening

I have visited many audiophiles who have spent huge amount on the system and are anal about every small thing..who PLAINLY cannot hear properly

I will give you some instances

1) 20 ppl in demo at a local hifi purveyors room , only 2 besides myself picked up on weird phasey effects .. the rest sat there and nodded

2) audiophile pal that spends fortunes on cables , plugs , tweaks that did not pick up a -6db difference between 2 channels - he had forgotten to switch to lower gain on one channel when swapping balanced to single ended cabling

3) audiophile with very expensive multiway speakers that did not pick up that one tweeter was MIA

4) Audiophile pal that spent a month listening to his system out of phase

5) audiophile with a treated room and uberbuck system.. did not pick up that the room had loud buzzes and rattles and that the amp was clipping

6) hifi dealer didnt pick up that one woofer was foobarred

7) Audiophile at my house .. I have G1's and 4 subs , enough bass to crack my ceiling(tis true) and who asked me "where is the bass" on a bass heavy track , you could feel the pressure waves and feel the visceral bass..he didnt

8) 2 folk who had left/right swapped and never picked it up .. classical music lovers both of them

That was just in the last year.....the list goes on

 

[BruceD]

The part of this thread that resonates with me is the lack of hearing accuity or lack of training re critical listening

I have visited many audiophiles who have spent huge amount on the system and are anal about every small thing..who PLAINLY cannot hear properly

I will give you some instances

1) 20 ppl in demo at a local hifi purveyors room , only 2 besides myself picked up on weird phasey effects .. the rest sat there and nodded

2) audiophile pal that spends fortunes on cables , plugs , tweaks that did not pick up a -6db difference between 2 channels - he had forgotten to switch to lower gain on one channel when swapping balanced to single ended cabling

3) audiophile with very expensive multiway speakers that did not pick up that one tweeter was MIA

4) Audiophile pal that spent a month listening to his system out of phase

5) audiophile with a treated room and uberbuck system.. did not pick up that the room had loud buzzes and rattles and that the amp was clipping

6) hifi dealer didnt pick up that one woofer was foobarred

7) Audiophile at my house .. I have G1's and 4 subs , enough bass to crack my ceiling(tis true) and who asked me "where is the bass" on a bass heavy track , you could feel the pressure waves and feel the visceral bass..he didnt

8) 2 folk who had left/right swapped and never picked it up .. classical music lovers both of them

That was just in the last year.....the list goes on

Ha!--Good ones Rodders ol chap!

You left out the one where the Friend comes over to hear your system and brings out the Most disgusting sonic record in his collection for you to play

Read " K-tel Favs of the 70s' or "Ace of Clubs bargain Pop of 81"

Sounds like Sh*te on anything

BruceD

 

[Rodney Gold]

Yeh .. I hate that .. folk come over for a listen... Im always asking them to play what they know etc...they either are like a rabbit in the headlights thinking of something.. and invariably they pick something terrible or hotel california , dire straits , dianna krall or something compressed to death and distorted as hell
Kinda like test driving a Ferrari on a dirt road...

 

[Al M.]

Ha!--Good ones Rodders ol chap!

You left out the one where the Friend comes over to hear your system and brings out the Most disgusting sonic record in his collection for you to play

Read " K-tel Favs of the 70s' or "Ace of Clubs bargain Pop of 81"

Sounds like Sh*te on anything

BruceD

Or the guy in The Netherlands where I used to live whose favorite music on his expensive mega system was "Die Schluempfe":

https://www.youtube.com/watch?v=Yoc976WQNmA