• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Why do records sound so much better than digital?

Status
Not open for further replies.

Alexium

Active Member
Joined
Sep 22, 2019
Messages
223
Likes
144
Location
Ukraine
Vinyl = distortion and compression. If you read up about mastering for vinyl, you may be amazed by how compromised it is (I know I was).

I agree with what was said above, though - for its time, and for such a simple technology, it does play music very very well.

I myself like some albums on vinyl better than on CD. I listen to them as digital rips from vinyl, which, of course, sounds exactly like the original but doesn't wear out and doesn't need dusting off. Perhaps, the limitations of vinyl make the mixing and mastering engineers think harder about what really matters in music. But not every release sounds better on vinyl than on CD, far from it.
 
Last edited:

weesch

Active Member
Joined
Apr 28, 2021
Messages
107
Likes
23
Location
Paris
ok
i see but why is the meaning of this document ?
best regards...
hi
i have tried protools at 192 khz with 01v yamaha at 96 khz it work with rme adi 8 dd
i hear that the sound is better in the yamaha EQ (better high better bass) at 96 khz.
also all the protools rtas plugins have a better sound resolution at 192 khz.
this let me think that this higher sample rate you use , better is the sound .
 
Joined
Oct 13, 2022
Messages
50
Likes
98
Funny, this love for vinyl. I love it too, when doing graphic work. Layout for LP is much more rewarding than a 14x12 cm box !

But knowing the recording side, and how extremely compromised contemporay vinyl mastering/pressing is, even low rate mp3 sounds better. (hang on to your vintage pressings !)

In the Eu turnaround for vinyl is 9 months, unless you are a really big client. Cd 10 work days…
in the Eu there are only a handful of facilities with direct metal mastering lathes, all the rest is in fact not suited for classical music.
in the USA there are apparently only two DMM cutters left, owned by the scientology church.

Speaking as a pianist, analogue is laughable. I never heard an analogue recording, played back on vinyl, which represents a grand piano with enough pitch stability to make it sound credible.

On a final note, I am amazed even on this forum there are those who think there are gaps of missing information between those 1 and 0 ? Seriously ? I would read up on the technology, it is literally getting old !

This myth once again proves that you hear what your brain thinks you must.
Funnily, according to some digital wizards (i guess it’s a quote by Bruno Putzeys) digital is in fact an analogue system.
 

Alexium

Active Member
Joined
Sep 22, 2019
Messages
223
Likes
144
Location
Ukraine
Speaking as a pianist, analogue is laughable. I never heard an analogue recording, played back on vinyl, which represents a grand piano with enough pitch stability to make it sound credible.
What are the best / most impressive recordings of a piano that you know?
 

earlevel

Addicted to Fun and Learning
Joined
Nov 18, 2020
Messages
545
Likes
776
Funnily, according to some digital wizards (i guess it’s a quote by Bruno Putzeys) digital is in fact an analogue system.
Yes, mainly because analog and digital are different levels of specificity. I could come up with a better word, I'm sure, but distracted listening to the Prokofiev you recommended; digital is very specific—a numerical representation of a signal, whereas analog basically means a representation of a signal. Analog—"Something that bears an analogy to something else; something that is comparable." A mic produces an electrical signal that is analogous to the sound waves hitting it, grooves are cut that are analogous to the electrical signal, which get converted to an electrical signal that's analogous to the groove, which drive a speaker that pushed air analogous to the original air pushing.

The true difference is discrete time versus continuous time. Digital is always discrete time, but it's just a digital representation of an analog sampling (zero-level connecting the samples in the continuous domain). And since numbers are an idea, we can't really store them without making an "analog" of them (representing them as bits stored as voltage levels in chips, or converting them to an analogous continuous voltage for SP/DIF, etc.) The subtle difference is that these are not strictly analogs of the signal, but of the discrete-time—sampled—signal. That's a profoundly different reference point. Once sampled, it's no longer the original signal, it's been modulated and is no longer the original signal, so my point of view is that claiming digital is also analog is going too far. I'm not going to fight anyone about it, I just don't think they do anyone a favor by claiming it, it jsut makes discussion more difficult.

That's a long way of saying that while people tend to consider analog and digital to be mutually exclusive but equivalent descriptors, they actually describe different things altogether. They don't have the same relationship as, for instance, continuous and discrete. And to complicate discussion, a signal that has never been digitized can be either analog or discrete. Of course, discrete is almost always digital, because it's so much easier to store and process if digitized. The bottom line is that most people use analog to mean continuous analog, and digital to mean quantized discrete sampled (usually PCM), so it just muddies to water to claim digital is analog. Yes, it's an analog of something, just not that of the original sound like what "analog audio" is.

I somehow feel less clean for going through that explanation. :p
 

Gringoaudio1

Addicted to Fun and Learning
Forum Donor
Joined
Sep 11, 2019
Messages
585
Likes
743
Location
Calgary Alberta Canada
Yes, mainly because analog and digital are different levels of specificity. I could come up with a better word, I'm sure, but distracted listening to the Prokofiev you recommended; digital is very specific—a numerical representation of a signal, whereas analog basically means a representation of a signal. Analog—"Something that bears an analogy to something else; something that is comparable." A mic produces an electrical signal that is analogous to the sound waves hitting it, grooves are cut that are analogous to the electrical signal, which get converted to an electrical signal that's analogous to the groove, which drive a speaker that pushed air analogous to the original air pushing.

The true difference is discrete time versus continuous time. Digital is always discrete time, but it's just a digital representation of an analog sampling (zero-level connecting the samples in the continuous domain). And since numbers are an idea, we can't really store them without making an "analog" of them (representing them as bits stored as voltage levels in chips, or converting them to an analogous continuous voltage for SP/DIF, etc.) The subtle difference is that these are not strictly analogs of the signal, but of the discrete-time—sampled—signal. That's a profoundly different reference point. Once sampled, it's no longer the original signal, it's been modulated and is no longer the original signal, so my point of view is that claiming digital is also analog is going too far. I'm not going to fight anyone about it, I just don't think they do anyone a favor by claiming it, it jsut makes discussion more difficult.

That's a long way of saying that while people tend to consider analog and digital to be mutually exclusive but equivalent descriptors, they actually describe different things altogether. They don't have the same relationship as, for instance, continuous and discrete. And to complicate discussion, a signal that has never been digitized can be either analog or discrete. Of course, discrete is almost always digital, because it's so much easier to store and process if digitized. The bottom line is that most people use analog to mean continuous analog, and digital to mean quantized discrete sampled (usually PCM), so it just muddies to water to claim digital is analog. Yes, it's an analog of something, just not that of the original sound like what "analog audio" is.

I somehow feel less clean for going through that explanation. :p
Awesome post. Thanks.
 
Joined
Oct 13, 2022
Messages
50
Likes
98
I get your point, however, I feel mr Putzeys is not per se talking about the digital storage. I think he talks about digital behaving like analogue while designing eg. The Mola Mola dac, with thd below -140 dB, and almost inexistant jitter sensisivity, using NO off the shelve parts. It is a discrete dac …

one thing to ponder : if we consider the digital recording and playback chain as a whole, (forget for a moment about that one intermediate step/storage where the ones and zeroes exist), the difference with an analogue chain becomes really small. It is my understanding analogue tape has bit depth as well, and if you zoom in far enough, can even be considered a non-continuous storage medium. The question might be, when does analogue become digital ?

I don’t want to keep you from Cinderella …
 

Alexium

Active Member
Joined
Sep 22, 2019
Messages
223
Likes
144
Location
Ukraine
zero-level connecting the samples in the continuous domain
The digital signal is just the samples, there's nothing in between. How you connect the samples continuously is part of the reconstruction process and there are different ways to do it (all of them correct, but with different tradeoffs).
 

Alexium

Active Member
Joined
Sep 22, 2019
Messages
223
Likes
144
Location
Ukraine
analogue tape has bit depth as well,
Very true, and so does vinyl. It is at least limited by signal to noise ratio, and there are different stages in reading / writing the recording, each of which with its own distortion and SNR. There is a limit to how close two analog levels can be to remain distinguishable to the rest of the system.

In my opinion, it's only digital when it's losslessly processed / stored in the digital domain. As soon as you start putting a digital signal through the physical reality, even when trying to maintain its original digital form, you need to consider distortion and interference - the actual signal is analog, and you need to consider how you should encode and filter it to preserve the data without distortion. After successful filtering (e. g. on the other end of a transmission line / bus) you can again consider it truly digital, until the next time it escapes into the physical world.
And the insides of the microchip that performs storage and processing are also analog, of course, so the chip's designers need to take care in order to give you the robust abstraction of truly digital signal being stored or processed on the chip.
 
Last edited:

earlevel

Addicted to Fun and Learning
Joined
Nov 18, 2020
Messages
545
Likes
776
I get your point, however, I feel mr Putzeys is not per se talking about the digital storage. I think he talks about digital behaving like analogue while designing eg. The Mola Mola dac, with thd below -140 dB, and almost inexistant jitter sensisivity, using NO off the shelve parts. It is a discrete dac …

one thing to ponder : if we consider the digital recording and playback chain as a whole, (forget for a moment about that one intermediate step/storage where the ones and zeroes exist), the difference with an analogue chain becomes really small. It is my understanding analogue tape has bit depth as well, and if you zoom in far enough, can even be considered a non-continuous storage medium. The question might be, when does analogue become digital ?

I don’t want to keep you from Cinderella …
"When does analog become digital"...What you're really asking here is "when does continuous become discrete?" Fortunately, it happens lower than we can hear. And before it becomes discrete, it's already buried in analog noise. Details:

First, the analog noise part: There is a limit to how quiet you can make an electronic circuit, because there is a point where even a single resistor, no matter the quality, generates noise (Johnson-Nyquist noise). It depends on the resistance and temperature, but basically it's around -131 dB of typical maximum audio output levels. You can't make audio gear quieter and live to hear it (absolute zero doesn't make for a good listning environment). But if we ignore that and keep going down in level, we also get to a point where a signal is no longer continuous. But it's never truly continuous, because it's all about electron flow, and you can't flow a fraction of an electron. If you're flowing billions of electrons per second, one less will seems like a continuous change. But if you're down to three per second, there will be just three events in a second, and it's going to sound (if we could hear it) like a geiger counter (referred to as shot noise). But the same is true of most things in the world—water going out of a hose is continuous, for practical purposes, but at some point you'll get down to individual molecules over a noticeable amount of time—though surface tension will obscure that fact long before we get there.

So, if I were to pick a rough number of out the air, I might say continuous audio becomes discrete around -150 dB. The exact number isn't important, because we can't hear it anyway under practical audio listening conditions (for instance, we could amplify it, but that's not a practical listening condition because music has loud parts too, and it would be disastrous). And the signal would just descend into random noise anyway before we noticed.
 
Joined
Oct 13, 2022
Messages
50
Likes
98
Interesting, thanks. It (the quote) then makes sence, as the Mola Mola dac has better than -140 dB of thd plus noise (maybe significantly, as it is the limit of measurement systems). That is awfully close to the point where you picked a number where analogue might become discrete.
nice paradox !
 

earlevel

Addicted to Fun and Learning
Joined
Nov 18, 2020
Messages
545
Likes
776
The digital signal is just the samples, there's nothing in between. How you connect the samples continuously is part of the reconstruction process and there are different ways to do it (all of them correct, but with different tradeoffs).
Zeros in between, but we needn't store the zeros, because mathematically they are immaterial to any signal processing (see * for an intuitive explanation of why having the zero level and processing them gives the same results as skipping them). This was the whole reason we encoded via PCM—to make the information between sample instances zero.

PCM is Pulse Code Modulation, which is Pulse Amplitude Modulation, encoded into a numerical value (digitized, which further implies quantization).

I've had people disagree with me and say "we really don't know what's in between". Nope, do the math. No matter what continuous signal you start with, if you perform amplitude modulation with it and a unit pulse train (which is precisely what periodic sampling does), the result is a continuous signal with impulses scaled to the height of the original signal at the corresponding points, and a flat zero level in between. To digitize it, we just encode (the C in PCM) the impulse heights a ditch the zeros in between.

*But remember, we aren't required to digitize. We could stop at PAM, and do things like filtering, gain changes, and summing, using that continuous signal of impulses connected by zero-level voltage through a mixer, for isntance, then lowpass below half the impulse rate and get exactly what we'd expect, the same thing as if we had digitized and used the same mathematical steps.

(I've also had people tell me, "well, ok, but it's not really zero, it's some DC level we don't know". Completely immaterial, it's like saying that if your feet are grounded and you touch 100 VDC reference to ground, that it matters whether you argue that "ground" is really at another voltage level referenced to something else, therefore the actual voltage through you is unknown—nope, you're going to get 100 VDC through you.)

It's the very reason we upsample by integer ratios by inserting zeros then lowpass filtering. Because there are zeros between the samples, we just didn't need to store them. (But since we changed the sample rate by inserting more samples, we need to remove the resulting aliasing with the lowpass.)
 

Keith_W

Major Contributor
Joined
Jun 26, 2016
Messages
2,531
Likes
5,803
Location
Melbourne, Australia
These may already have been covered by earlier posts, but the reasons I can think of are:

1. More crosstalk on vinyl. Uli Bruggeman from Acourate has a white paper somewhere about digital emulation of crosstalk which he says produces a more relaxing sound which he has simulated via DSP with Acourate Flow in his convolver. I have tried Acourate Convolver with this feature and yes it does produce a different sound.

2. RIAA equalisation results in minimum phase behaviour. Pro audio people know that this results in a type of coloration that is actually desirable (the article says linear phase causes "precise, digital sound that to some engineers is too unnatural").

3. Loudness wars. Although this exists on newer vinyl where the signal is processed digitally (and the sound engineer applies dynamic compression), older pressings do not suffer from this problem.

The above would be consistent with what I hear with vinyl. I actually prefer this kind of coloration with better dynamics on well pressed vinyl. Since my system is full DSP I can emulate the first two via DSP, and I am looking for a way to exaggerate dynamics via DSP.
 

earlevel

Addicted to Fun and Learning
Joined
Nov 18, 2020
Messages
545
Likes
776
These may already have been covered by earlier posts, but the reasons I can think of are:

1. More crosstalk on vinyl. Uli Bruggeman from Acourate has a white paper somewhere about digital emulation of crosstalk which he says produces a more relaxing sound which he has simulated via DSP with Acourate Flow in his convolver. I have tried Acourate Convolver with this feature and yes it does produce a different sound.
Good points, worth talking about. I'm just taking some shots at advocating for the other side here:

Without this crosstalk, it's what the producers of the music intended. So, even if crosstalk improves it...should we advocate to music producers to add crosstalk? I don't think so—they produce the sound they want to convey in the studio, I think they'd want to maintain that authority. At best, if you're a person who like increased crosstalk, then I suppose it might be one reason you're attracted to it. (I wonder if there is a market for crosstalk boxes to add this feature to a digital listening environment, or amps/preamps or even DACs with this feature, in order to win back people from vinyl?)

2. RIAA equalisation results in minimum phase behaviour. Pro audio people know that this results in a type of coloration that is actually desirable (the article says linear phase causes "precise, digital sound that to some engineers is too unnatural").
RIAA equalization is spread through the audio range. In digital, it's a lowpass with a corner frequency at the outer bounds of the audio range. Two very different things. The digital filter can be minimum phase (my Topping DX7 Pro has minimum phase as a choice). That won't change how it was digitized, but if you listen to music sampled and processed at a 96k, then ultimately played back with a minimum phase filter that rolls off at 20k, you've effective got the benefit of minimum phase regardless of what was used to digitize it at the higher rate. Yet the world hasn't overwhelmingly discovered how that clearly sounds better.

Also, "that to some engineers is too unnatural", how many is "some"? Audio engineers work almost exclusively in digital audio, and I don't think many of them struggle with it due to sounding unnatural. Most talk about how incredibly good their premium converters sound. And if it's the difference between premium and typical, they both would be using linear phase most often, so I doubt it's "minimum phase" that's the secret.

3. Loudness wars. Although this exists on newer vinyl where the signal is processed digitally (and the sound engineer applies dynamic compression), older pressings do not suffer from this problem.
This is a separate thing. Older pressing don't suffer largely because massive compression wasn't in vogue back then. And in many cases it's not in vogue so much now, but the point is that vinyl isn't a cure for loudness wars, and music production that suffers with lack of dynamic range isn't cure by pressing to vinyl. Also, a lot of people still like to apply that compression with much of the same analog gear that was available before digital, so it's not digital processing, it's engineering practice.

Again, I don't mean this argumentatively. Just raising counter arguments.
 

Newman

Major Contributor
Joined
Jan 6, 2017
Messages
3,454
Likes
4,218
Are you “when does digital/analogue become analogue/digital” philosophers happy now? Good. Maybe we can return to meaningful things.
 

Sal1950

Grand Contributor
The Chicago Crusher
Forum Donor
Joined
Mar 1, 2016
Messages
14,073
Likes
16,609
Location
Central Fl
Funny, this love for vinyl. I love it too, when doing graphic work. Layout for LP is much more rewarding than a 14x12 cm box !

But knowing the recording side, and how extremely compromised contemporay vinyl mastering/pressing is, even low rate mp3 sounds better. (hang on to your vintage pressings !)

In the Eu turnaround for vinyl is 9 months, unless you are a really big client. Cd 10 work days…
in the Eu there are only a handful of facilities with direct metal mastering lathes, all the rest is in fact not suited for classical music.
in the USA there are apparently only two DMM cutters left, owned by the scientology church.

Speaking as a pianist, analogue is laughable. I never heard an analogue recording, played back on vinyl, which represents a grand piano with enough pitch stability to make it sound credible.

On a final note, I am amazed even on this forum there are those who think there are gaps of missing information between those 1 and 0 ? Seriously ? I would read up on the technology, it is literally getting old !

This myth once again proves that you hear what your brain thinks you must.
Funnily, according to some digital wizards (i guess it’s a quote by Bruno Putzeys) digital is in fact an analogue system.
Dynamite post, all totally right on point!
 

Keith_W

Major Contributor
Joined
Jun 26, 2016
Messages
2,531
Likes
5,803
Location
Melbourne, Australia
Without this crosstalk, it's what the producers of the music intended. So, even if crosstalk improves it...should we advocate to music producers to add crosstalk? I don't think so—they produce the sound they want to convey in the studio, I think they'd want to maintain that authority. At best, if you're a person who like increased crosstalk, then I suppose it might be one reason you're attracted to it. (I wonder if there is a market for crosstalk boxes to add this feature to a digital listening environment, or amps/preamps or even DACs with this feature, in order to win back people from vinyl?)

Actually there is a market for crosstalk DSP which are mostly available as VST plugins. If you have playback software capable of hosting VST plugins (e.g. JRiver) some of them are free to try. Most of these plugins are for headphone users, but they also have an effect on your home system. As for a DAC with this feature, there is the RME ADI2 FS. In a home system there is already crosstalk from the speakers and you can adjust the crosstalk by moving and angling the speakers. Adding crosstalk upstream to the speakers seems to have a subtle effect on my system but the result subjectively seems to focus the soundstage more towards the center and reduce the width.

RIAA equalization is spread through the audio range. In digital, it's a lowpass with a corner frequency at the outer bounds of the audio range. Two very different things. The digital filter can be minimum phase (my Topping DX7 Pro has minimum phase as a choice). That won't change how it was digitized, but if you listen to music sampled and processed at a 96k, then ultimately played back with a minimum phase filter that rolls off at 20k, you've effective got the benefit of minimum phase regardless of what was used to digitize it at the higher rate. Yet the world hasn't overwhelmingly discovered how that clearly sounds better.

Yes. I should have also mentioned that when the record is cut, they apply RIAA equalisation. When it is played back, it is corrected with HOPEFULLY an exact inverse of the filter used. In reality it is probably not an exact match because of the sheer number of cartridges / phono stages / and the near infinite permutations of these combinations along with impedance matching, etc. thereof.

Which brings me to another point I should have added to the post. Vinyl enthusiasts know that certain cartridges sound "brighter" or might have better dynamics, etc. so they mix and match cartridges and play with impedance settings until they get the sound they want. Not to mention adjust the VTA, tracking force, etc. I think that for some people, vinyl is in effect a way to tune their system to sound right to their ears - something which is not possible with digital unless you use DSP. They are effectively using it as an analog equaliser, a way to add or subtract frequency anomalies, correct for perceived flaws - all to achieve the sound that they want.

What most of us on ASR do is try to get all the electronics and mechanical components as linear as possible then treat the room as best as we can - but ultimately the speaker/room will mess it all up anyway so you have no recourse to correct the sound of your system unless you use DSP. A lot of people on ASR don't use DSP, yet there is a lot of sneering going on at people who use vinyl to tune the sound of their system. Well, at least they have a way to adjust the sound of their system, however imperfect, non-measurement based, and subjective it is. Remember, it's not as if changing anything in a vinyl system won't have an effect, and will not through fortuitous luck produce the result that you want. It's just that it's less predictable and expensive, but in my opinion that does not make it an invalid option for people who do not want to use DSP.

Also, "that to some engineers is too unnatural", how many is "some"? Audio engineers work almost exclusively in digital audio, and I don't think many of them struggle with it due to sounding unnatural. Most talk about how incredibly good their premium converters sound. And if it's the difference between premium and typical, they both would be using linear phase most often, so I doubt it's "minimum phase" that's the secret.[/reply]

Don't know. The experiment is easy enough to replicate provided you have access to an audio system like mine, where the crossovers are generated in a PC and hosted in a convolution engine. It would take me an hour to make a minimum phase filter instead of the linear phase that I am currently using. TBH I never did do that experiment to hear the difference between minimum and linear phase, I just read that linear is superior, implemented it, and that was that. I should go and set it up and hear for myself what difference there is. To be fair, linear phase does have its issues and not just with "unnatural digital sound". You need more computing power (which brings its own problems with expense, cooling, and fan noise), and you have pre-ringing.

This is a separate thing. Older pressing don't suffer largely because massive compression wasn't in vogue back then. And in many cases it's not in vogue so much now, but the point is that vinyl isn't a cure for loudness wars, and music production that suffers with lack of dynamic range isn't cure by pressing to vinyl. Also, a lot of people still like to apply that compression with much of the same analog gear that was available before digital, so it's not digital processing, it's engineering practice.

I agree. No argument there.
 
Status
Not open for further replies.
Top Bottom