We are a very long way off perfection in any dimension here. Every format recording and playback has imperfections. Perhaps you can be more specific?
My reason for posting this link was to stimulate discussion.
Has anyone been able to define well or measure differences between vinyl and digital?
It’s obvious right? They sound different, and I’m sure they measure differently. Well we know the dynamic range of cd’s is larger than vinyl.
But do we have an agreed description or agreed measurements of the differences between vinyl and digital?
I know this is a hot topic so I am asking not for trouble but for well reasoned and detailed replies, if possible. And courtesy among us. Please.
I’ve always wondered why vinyl sounds more open, airy and transparent in the mid range. And of cd’s and most digital sounds quieter and yet lifeless than compared with vinyl. YMMV of course, I am looking for the reasons, and appreciation of one another’s experience.
137 responses Add your response
@johnread57 , by posting that link, is that not an admission that all the differences are due to the recording and flaws in the format? |
Thank you @johnread57 . I don’t normally put mich stock in his videos but he does have access to the process. He seems to support my claim above that vinyl is noticeably different no matter how perfect the vinyl it is still changing the sound. |
First post ever here so I thought I'd jump into the deep end. To preface my comment, I'm a music-lover who likes quality audio equipment with limited tinkering. I'm not swapping out gear on a regular basis or cycling through various speaker setups, I have a mid-fi setup because that's what I can afford. I've cycled back and forth over the years w/ vinyl, reel-to-reel, CDs, HD digital, and streaming. One can run numbers all day about frequency response and bitrates, etc. and they're still numbers. At the end of the day though, for me it's about how the music sounds when I'm listening. Casual listening/background listening? Streaming is fine. I can "live" with the audio quality while I'm doing chores around the house. When I'm sitting down in the evening when the house is quiet, and I want to listen to particular albums I listen to the best available recording I have. That might be a CD or vinyl or HD digital download or HD streaming. People will go to the mat defending the warmth and audio purity of vinyl over digital but for my listening purposes it really is dependent upon when the master recording was made (pre-digital mastering) and what format the master recording was in (tape / digital). For example, my original copy of Beggars Banquet sounds like the late 60s piece of vinyl that it is. It's a "grungy" blues-rock album that sounds better on vinyl as it relates to warmth and the organic noise that is inherent to vinyl while the digital copy sounds a little too clean. Conversely, a new ambient / electronic album mastered digitally will sound great on CD / Streaming and marginally different on vinyl where the inherent flaws of vinyl imprint themselves on playback It's a different listening experience but in that respect, the vinyl isn't necessarily superior to the digital playback. And sometimes, digital /streaming sounds way better than the vinyl version. I had this happen on the new Calexico release. The vinyl doesn't sound nearly as good as the digital copy. It's too muddy. And that's an album with intentional "grit" in the recording to sound more vintage. The new vinyl pressing buries some instrumentation. I think if you're evaluating which recording format sounds "better" is very much subjective to your listening situation at that point in time and I also like to keep in mind that there's a lot of music out there that was intended to be played over the radio, loudly at concerts and on cheap consumer turntables, not overly analyzed for instrument separation and soundstage. |
We have internally used a blended metric similar to the Gm Gedlee metric for quite a few years. I know at least 2 of our competitors do as well. I agree on his basic premise that mechanical systems (speaker drivers) do not have as much higher order distortion, but on a practical basis, with multiple drivers, how distortion presents is more complex. One missing element is frequency weighting. In one of the links, 2nd order distortion was described as euphonic. That may be true at a high enough level. Using the Gedlee metric, at the levels indicated, it would inaudible or close enough. Almost any DAC today would have a Gm close to 0. Not just expensive ones, all of them. The vinyl vs CD article talks a lot about vinyl, not much about CD. One small, near useless section. I lost confidence in the author over a few items. Changing the load on an MC cartridge (135 ohms), from100pF to 200pF reduced the 3rd order distortion? I have done enough DIY electronics to know that smelled funny. That would change an RC filter from 12MHz to 6MHz. I say bad amp, bad switch setting, or multiple plays. Raises an issue with phase in filters on CD while ignoring that vinyl likely has phase shift too. Show digital samples that look bad, say it needs a reconstruction filter, then don’t show one? Bio attached to the article. Really impressive! He must be an expert? BS-Zoology, BS-Psychology, MS-Physiological Psychology, Ph.D.-Neuroscience. Welcome to 2022, everyone is an expert. |
Since this is also an exposé of human pleasure and pain, we should understand these systems too. Here’s a starter reference. |
Some references I found informative. Level I Popular level. Summarizes pros and cons, remaining in old paradigm, e.g. in regard to dynamic range estimation.
Level II Somewhat technical level. Parts 1-5 provide good overview of LP technology. Parts 6-9 describe distortions inherent in it.
Level III Very technical level. This explains why, despite obviously inherently high level of distortions, LPs sound just fine to many people. Executive summary: this is because LP distortions are mostly of "right" kind, which human hearing system do not register with the same intensity as "wrong" distortions. |
If you're talking about new prices, I don't think $10K can get you there. And you'll probably need to spend more than that for a phono stage. But I absolutely agree about the importance of a clean record - that makes all the difference in the world and is the reason I use an easy "one button" ultrasonic LP cleaning system.
Probably true. |
@cleeds , I did not mean you with respect to "digital confusion". I do think vinyl done well, excellent pressing (clean), good turntable, good cartridge, all properly setup sounds very good. I don't think you need to spend $50K either. $10K maybe. I also think if you listen to that side by side with CD, you will always be able to differentiate them, even if its the slightest tick. Anytime I have been in situations where they are compared side by side, they are always different. I won't claim the level matching was perfect. If the vinyl frequency response was not flat, I am not sure that is possible. I am open to it being just mastering, mastering and cross-talk, maybe my FR is not as flat as I think it is, etc. I think my only point, at this point, is that for people who have a vinyl preference, there are simple and probably obvious reasons we could find if we looked closely at their system or the music they listen to. You say you are an analog guy, but you have not commented on your preference or thoughts about why? |
This is a tricky question because "very small" is subjective and the extent to which differences are audible vary depending on the content. I have a digital recorder that can do 24/96, and the difference between that and 16/44.1 can be noticeable. A lot depends on what you're recording.
I'm not sure I agree. I know it annoys some of my fellow analogphiles when I say that the very best LP playback and the best CD playback sound very, very close. But that can't happen without a lot of effort and expense on the LP side.
I explained previously in this thread that the math behind digital audio - Fourier Transform and Shannon/Nyquist - is perfect and can be proven using math. Digital's flaws are elsewhere.
If we're talking high-end LP playback, the differences are much smaller than many believe. Many people are stunned when they hear first class LP playback for the first time. |
My posting issue was not vinyl versus digital except in as much as they are different, and can be use as comparators. Not better. The diversity of situations from recording to playback are very clearly well described in the article from thatspeakerdude that I cited above this morning. I agree that we are looking for progress in this conversation not walls or reverse cycles. In the playback domain, it is amusing cleeds, that audiophiles, me included work so hard to reduce distortions when each format has distortions of various impacts, some desirable in some situations and others perverse. From tubes to rooms, from vibration to EMI/RFI, the playground is full of hazardous pursuits. Again though to my OP, I was focused on the warmth, depth and richness as basic identifiable differences between the two formats. Not always, but generally. And we seem to have made some progress in elucidating the circumstances of these differences in the recording and playback chain. Flaws, like beauty, are in the eye (or the screen) of the beholder. |
@cleeds I included the weasel word little. I linked examples where analog sources were passed through a digital chain of CD quality and the listeners were not able to detect it. I could waste hours on the web finding examples of suitable implemented testing comparing CD quality audio to hi-res audio that would conclude no detectable difference. Can we agree that if there is a difference between CD quality done right and high resolution that the difference is very small, and hard or very hard to detect? Play any example of vinyl ever made and the closest CD and everyone will be able to tell them apart. Maybe you will find some obscure set where that is not true. Can we say 99.9% of them? Take the last two paragraphs and put them together. The difference between vinyl and CD is bigger, much bigger than CD and high res. I am working from the assumption that high resolution digital is good enough to be perfect. Two-four times the bandwidth of CD, 20db or more of added dynamic range and hard to tell the difference from CD. I would say it is near perfect.
Every vinyl versus digital argument seems to devolve into an attempt to find some mysterious flaw with digital that cannot be supported with math, engineering, nor experiment. Maybe there is some flaw at CD quality that we can possibly detect. If there is, it is very small. The differences between CD and vinyl are not small. Some progress in understanding would be nice. It is not going to happen by starting with an unjustified conclusion and working back. |
Read this morning: https://aestheticsforbirds.com/2021/04/07/an-audio-professionals-take-on-vinyl/ Very good summary of this discussion from a layman’s perspective. Covering the two formats and recording chains. I especially appreciate the descriptors used, warmth, richness, and depth. I can only recommend reading this paper. The academic research papers are a different kettle of fish. |
Sorry, no can do. Setting aside the problems and fallacies that are common to typical audio blind testing (this addresses some of them), there have been tests that show hi-res audibility. Here's an example via AES.
If your claim were true audiophiles wouldn't work so hard to reduce the noise and distortions to which analog is prone. On paper, those specs often aren't the equal of digital, but that's moot if they're near or below the level of audibility. |
I've seen accounts of experimental results seemingly proving that CD format is perceptually transparent. Also, accounts of experiments seemingly proving the opposite. Not satisfied with what I've seen in these papers, I conducted my own experiments over the past decade, where I controlled the whole recording and reproduction chain. The results, for me personally, were more definitive. I have to disclose at that point that I'm educated, certified, and worked in the areas of Physics and Neurophysiology, including stints in several national labs, in Europe and USA. So, my standard of inquiry is rather high. When the outcome of an experiment depends on accurate detection of one photon vs two arriving within a given microsecond, it has to be. Experiments in psychoacoustics are hard, mostly because music signals and the final "instrument" - human hearing system - are very variable. Thus, I put more faith in meta-studies rather than in individual experiments, including my own. Meta-studies, for instance like the one I already referred to in this thread:
Proponents of this or that point of view like focusing on one particular study, or a handful of them, proving their point. I don't believe that's the way to go. For instance, the oft-cited Boston Audio Society study (BAS study), which seemingly proved the perceptual transparency of the 16/44.1 digital loop, didn't conduct one of the mandatory steps any experimental science professional would do - calibration. If an experiment is to elucidate the importance of distortions, one absolutely has to take into consideration the nature and levels of distortions inherent in the gear involved in the experiment. What do we have in the case of the BAS study? https://www.bostonaudiosociety.org/explanation.htm "The Principal System The playback equipment in this system consisted of an Adcom GTP-450 preamp and a Carver M1.5t power amplifier." What is Carver M1.5t power amplifier?
SINAD reveals that the amp's distortions are high and rising with frequency. "... distortion-free range = 14 to 11 bits". "Distortions rises @ 66 watts". "Max power = 273 watts @ 44 dB SINAD". What the designers of the study should have done could include at least, back in mid-2000s: taking the Principal System into a certified anechoic chamber, and measuring its performance, including its distortion profile, using certified calibrated instruments. |
@johnread57, this is not at all about vinyl, but about analog tape recording from the perspective of a recording engineer. There are some great sound bites to listen to. Summary: Digital comes out just like the way you put it in. Analog recording does not sound like what you put in. It may sound better. It may sound worse. This guy has 9 albums nominated for Grammy’s. He is not a hack. Vinyl has a lot more distortion than digital and it rises with frequency. Add this to the qualities of vinyl that others has posted here. http://recordinghacks.com/2013/01/26/analog-tape-vs-digital/
https://aestheticsforbirds.com/2021/04/07/an-audio-professionals-take-on-vinyl/ There is MOFI, and this, http://drewdaniels.com/audible.pdf
Step 1, accept that properly implemented digital, even CD quality, has no sound or so little to be ignored. Step 2, accept that vinyl and other analog formats have a particular sound, or many sounds, and that we like it because it has those sounds that appeal to us, as living, breathing humans. Step 3, figure out what those sounds are and encourage audio companies to work on them and recording and mixing engineers to make better use of them. Step 4, accept that we are mostly a group of old buggers, and maybe young people don’t like the same thing. I tried to research more on this last item but I didn’t find a lot of work on it.
|
@johnread57 , I came across this NPR interview again, has some good comments on this topic. I am going to copy some and a link to the whole article. https://www.npr.org/2012/02/10/146697658/why-vinyl-sounds-better-than-cd-or-not
|
@thespeakerdude : Very curious from your username: what are your own speakers? Thanks you |
Let's make it more interesting. Shall we? You likely know who Bob Stuart is:
Despite his lifelong achievements and indisputable expert status, audiophile community met MQA format he co-created with significant controversy, e.g.
I propose reading a peer-reviewed paper he co-authored:
And then telling us what Tidal did wrong to further stir up the MQA controversy. And perhaps what Tidal and others can do in the future to improve the MQA situation. You may think it can't be improved, and this would be interesting to hear too. I hope the ensuing discussion would be very relevant to topic at hand, because Tidal and other "Ultra HD" music streaming services hasn't yet made a dent in LP sales, but I believe they may do so in the near future.
|
@johnread57 , I was out a few messages ago.
To quote @cleeds these are fanciful imaginings, not serious discussion points. Serious discussion points would come with serious analysis or serious links showing that analysis. |
A math theorem is a principle that can be proven with math. The Fourier Transform is a theorem. It has been proven by both math and in actual implementation - it's the basis of both digital and analog audio. Your alternative paradigms aren't consistent with Fourier. You're probably having fun with your fanciful imaginings but that's all they are. If you want to make actual progress you'll have to accept the math. Anything else is futile and more than a bit silly. |
Hi there. I’d like to ask @fair & @thespeakerdude to disengage from your conversation with each other. You have reached an impasse where no greater value is being added either to my original questions or each other’s position. Again I want say thanks to both for your energy and effort to address the technical aspects of these questions. But honestly, I learned somethings from each of you, and I respect you for maintaining civility. I ask that you and we need to move on. Please. Id like to hear more diversity of thoughts and experiences here. |
@wturkey while it is possible for Paul to provide an answer, I have nmot seen it happen for me so far. But a link is easier to click on.
Back to the reality of the thread..
Is there a synopsis of the measurements?
Yes… Do we have an idea of what to consider and what to look at? |
To be frank, I'm bored at this point. Not getting anything of value back from you. Clearly, you are not reading not only what I'm referencing, but also the parts of my posts that you are quoting. I'm telling you that the errors are signal dependent, you are quoting this, and then are asking me to tell you exactly how big the errors are? You know, serious people run long simulations to answer this question for specific digitization schemes and sets of representative signals. May return to answering your questions when I see your commitment to learning. |
| Post removed |
Not what I meant. I meant that advances in signal processing and understanding of how mammalian hearing system works were significant over the past three decades. Yet not all of the advances are reflected in engineering practice yet. CD format, as an example, was developed prior to that, and its designers couldn't take advantage of these advancements. Sadly, this format and its derivatives remain leading by volume for lossless online streaming. Certain common handbooks on DSP and auditory science haven't been updated yet either. Leave alone the mass of practitioners who still work in the paradigm expressed in these handbooks. I'll give you a couple of examples. First, a gradual understanding, over about quarter of a century, of the role of so-called Octopus Cells in the functioning of the hearing system: https://www.google.com/search?q=octopus+cells+hearing. Second, influence of new micro-surgery and robotic devices and techniques, which became widely available in the past two decades, on the hearing system research and medical practice: https://www.google.com/search?q=cochlea+microsurgery+hearing+research.
That's why I prefer to believe in results of meta-analysis. Some of the smaller-scale experiments support the hypothesis, some others don't. All are imprecise in one way or another. Yet with a larger array of data, statistical inference starts working with sufficient, quantifiable precision. From where I stand, this meta-analysis shows that the old paradigm, supporting the notion that 16/44.1 can encode perceptually transparently any music for any human listener, is experimentally disproven, and thus the paradigm itself is falsified by the evidence.
You are welcome. Winning a gratitude, however small, of a tough intellectual opponent, is one of the best rewards one can hope for in a discussion.
I lost interest in this particular discussion quite long ago, after I was able to replicate the "tube sound" inside a common DAW, using easily available software plugins. I occasionally listen to tube amps owned by others, including very expensive ones, but keep coming away underwhelmed. In a way, the proponents of the tube and other highly distorting amps are right that there are some properties of such amps that evoke hard to measure phenomena inherent in the human hearing system, yet these phenomena can be activated by other means too. For instance, the Missing Fundamental effect. It makes an amp sound more warmly-bassy to some listeners, which can be pleasant on some source material. Yet, this effect is routinely used in Bluetooth boomboxes, via quite simple DSP algorithms. The downside is of course masking of midrange, so it is beneficial only for some music genres. |
@cleeds
Fourier Analysis, as a theory, is a subset of Harmonic Analysis theory. It has dozens of theorems. If you are interested, pick one of these books - Harmonic Analysis on Amazon - and see for yourself. A handful of the theorems is routinely used by DSP practitioners. For instance, Parseval's Theorem can be used for quick estimation of THD contributed by a device, via comparing pre- and post- waveforms.
One of the problems, as I see it, is simplification, excessive popularization, and at times even vulgarization of science, which became widespread in the Western world during the several past decades. I guess I have to go that route as well, given the circumstances. So here it goes. A mathematical theorem is like a part of a legal contract: its words have precise meaning, often unexpectedly different from their everyday meaning; and it has small print. As it relates to the Sampling Theorem, the phrase "contains frequencies" means something quite different from what everyday common sense would make one assume. And the theorem itself is just a paragraph in a long contract, with lots of small print. Let's say you have a health insurance contract, and it covers your teenage son too. He rides electric bicycle. God forbid, he gets in an accident on the bicycle, hits his head, and requires expensive urgent care and rehabilitation. Naturally, you assume that the insurance will cover it all, and you feel safe in belief that you'll be only out for deductibles and copays. Suddenly, your insurance company sends you official letter saying that this event isn't going to be covered by them, because during the accident, according to a police report, your son was violating The Law. Perhaps he was riding without a helmet. Perhaps he was riding on a walkway in a town with an ordnance prohibiting that. And there is a small print in the actual insurance contract stipulating its provisions null and void if injuries were sustained in the process of violating The Law. As it relates to the Sampling Theorem, the signal has to fulfill very precise obligations, before the theorem can guarantee its accurate capture and reconstruction. The meaning of the words describing these obligations is precisely defined elsewhere in the theory, in other definitions, lemmas, and theorems. Nothing wrong with the theory per se, at all. Practical music doesn't fulfill these obligations, and thus the Sampling Theorem only works approximately. How well it works can be calculated too, using other parts of the theory, yet this is far more involved, and the answer is signal-dependent. The vulgarization, in the case of CD format marketing, was in omission of the facts I described in the previous paragraph. Was it done on purpose or through honest mistake? I don't know. |
Perhaps @fair , can enlighten with at least 2 or 3 of these research papers he claims are hard to find? A new paradigm with 3 decades of research that legitimately calls into question all current signal processing and hearing knowledge should have many available sources to reference.
Still I wait for this. A meta analysis of purely digital sources, some too old to be relevant due to hardware limitations and others with experimental flaws, does not support your hypothesis let alone suggest there is any new paradigm. I do appreciate the repartee as it demonstrates the vinyl argument.
This is just like the tube discussion. Even though there are significant, identifiable differences between typical tube amplifiers and SS amplifiers with good design practices, differences that are highly audible, every discussion devolves into a debate between those who point out those differences and those who believe in some unseen, unmeasurable property that "must" exist.
|
G’day MikeyDee
My wife and I had the opportunity to attend a small Mozart chamber orchestra in the castle in Salzburg Austria a few years ago. That’s where sitting close within 6-15ft the small audience had a wonderfully intimate experience of dynamic range. No Hi-Fi experience can reproduce that intimate experience in a double thick stone castle, huge rectangular room with pillars etc. |
No, Fourier Analysis is not a theory. It's a theorem; it can be shown to be perfect with math. Unless you can show a fault in the actual math, all of your hand waving and word salad is for naught. Again, I'm an analog guy. But if we seek better digital, it can't be done without understanding how digital works. I'm all for hi-res and everything it takes to get the best out of it. But let's not pretend that somehow the underlying premise of digital audio is somehow broken. It isn't. Its problems are elsewhere. |
Let's talk about this subject just from a musical standpoint. In classical music, when a composer writes a score, he or she notates dynamics (forte, pianissimo, etc) to denote how loud or soft a passage is to be played. These dynamic markings are as much a part of the musical score as the melody or harmony. By definition, vinyl is a compressed format and cannot completely express loud or soft the way digital can. This is not an issue in most rock music, because there are not the dynamic markers in rock charts. In jazz, very few artists put dynamics into their performances. Art Blakey was one who made his ensemble play with a large dynamic range (I was lucky enough to see his band live before he passed away). This is not an attempt to plug or criticize either vinyl or digital, but to what audio should strive for from a purely musical perspective. I hope anyone reading this gets a chance to sit front row to hear a great symphony orchestra, so you can really hear the dynamics I'm talking about. |
I’ve always wondered why vinyl sounds more open, airy and transparent in the mid range. And of cd’s and most digital sounds quieter and yet lifeless than compared with vinyl. Most studios record fully digital nowadays. When streamed lossless via Tidal or Qubuz you get this source straight to your DAC. Assuming it’s a capable DAC you get the best representation possible. To make a vinyl, there are countless extra steps: - audio compression and RIAA to limit the low frequency groove amplitude. - mechanical process of cutting the master disk - chemical / mechanical process of pressing a disk - wear and tear of the (master) disk - added wow and flutter of the turntable - mechanical / electrical process of pick up element - Reverse RIAA correction and pre amplification of pick up voltage And now all of a sudden it sounds more open, airy, and transparent?
|
Will do, referring to questions in the original post.
Yes and yes.
This is debatable.
No. Because there are multiple - at least two - paradigms, defining certain important characteristics like dynamic range differently.
Tried my best to abide.
In the first paradigm, CD is superior in sound quality to LP. Digitizing CDs and delivering their content via online streaming should have killed off LPs. In the second paradigm theoretical framework, LP occupies a middle ground between CD and such perceptually transparent digital formats as PCM 192/24 and DSD128. Correspondingly, the second paradigm maintains that CDs, physical or digitized, are not capable of superseding LPs. But perceptually transparent digital formats likely will. |
@johnread57
Very well. This 2016 review A Meta-Analysis of High Resolution Audio Perceptual Evaluation contains references to "18 published experiments for which sufficient data could be obtained ... providing a meta-analysis involving over 400 participants in over 12,500 trials" Conclusion is: "Results showed a small but statistically significant ability of test subjects to discriminate high resolution content, and this effect increased dramatically when test subjects received extensive training."
Pair of charts below illustrates my statement about the growing LPs popularity and vanishing CDs purchases. In a wider context: digital streaming appears to be decimating CD sales, yet LPs have not been affected by that (or maybe even helped?). |
@dsnyder0cnn
you are correct, I thought of that afterwards. I should have said "from the finished product" I can also add, the whole vinyl pressing process doesn't fix, alter, improve, up or down sample the grooves |
I did not present an opinion. I presented verifiable, researched, well understood, mathematical facts. Facts not disputed by those with the deepest understanding of the underlying math, and those able to adapt the math to practical implementation.
Below is an opinion. It misinterprets personal opinion, narrow market popularity, and different to "something". That something is only described in easily falsified claims, falsified with math, not an appeal to narrow market popularity. Perhaps @Fair, can enlighten with at least 2 or 3 of these research papers he claims are hard to find? A new paradigm with 3 decades of research that legitimately calls into question all current signal processing and hearing knowledge should have many available sources to reference.
|
This has been one of the deepest technical discussions I’ve read on Audiogon. Thanks to Fair and TheSpeakerDude for sharing their viewpoints here. I watched the video posted earlier to explain digital and that helped me to understand some of these recent expositions. @Fair can you summarize on this issue? |
If anyone cares my patience is officially at an end :-)
Nyquist is not a paradigm or a guess about how things work. It is not about engineering. It is a well understood, well researched, to this point not disproven mathematical theory. Nothing in the last 30 years has changed that. How it is applied is also well understood including translating real world limitations to accuracy. Those are not engineering principles, they are math principles. |
Call me skeptical. No that is not right. I flat out don’t believe you are telling the truth.
Well isn’t that convenient. Hate ta break it to ya but I learned all the theory in an applied math course. Because little of this has to do with engineering. It’s applied math.
That is totally irrelevant to what you replied to.
Wrong. Very wrong.
For processing not delivery just like I said
No such thing as a 32 bit ADC or DAC. Purely a data standard ... And a marketing ploy. Maybe you will find some rounding errors in those bottom 8-9 bits.
I am only storing that last paragraph for posterity.
Shhhh don’t tell anyone, but vinyl is terrible at <20hz and pretty awful in practice >20khz .... But I can’t hear it so I don’t care |
Thank you for referring this document. Oldie but goodie. Excellent for illustrating the older paradigm vs newer paradigm. Let’s look at the graph there marked with "Let us begin by examining a band limited square wave". That’s what I meant by saying that quickly changing signals start looking ragged when band limited. The document goes into a detailed explanation of why this is happening. For a briefer explanation, one can peruse a Wikipedia article about Gibbs Phenomenon. Note that what we see on a square wave is an extreme example. The underlying mechanism of the Gibbs Phenomenon is in action on any harmonic signal with changing magnitude - just to a lesser degree, depending on ratio between characteristic time of the harmonic components magnitude change and sampling interval. The concentrated difference between the older and the newer paradigm is this: Subscribers to the old paradigm believe that the wiggles we see on charts like that don’t ever affect perception of sound quality, as long as the signal to be band-limited is "music", and the upper boundary is set at 22 KHz. The new paradigm tells us that it depends. That certain wiggles may affect perceived sound quality of certain music signals band-limited under the conditions above, for certain listeners. By the way, 0.1% THD corresponds to a width of one pixel on a typical laptop display, if a graph like that is enlarged to fill the whole screen. Basically, we can hear a difference that we can barely see on a graph. If one sees any visual difference of a band-limited music signal compared to the original one, this should arise strong suspicion that such difference may be heard. |
We are getting somewhere.
This is precisely how it should feel, from the point of view of someone remaining in an old paradigm. New paradigm overturns some of the old paradigm's assumptions and conclusions, which is obviously "wrong" in the context of the old paradigm.
Mathematically, Fourier Analysis is a theory based on integral transforms with harmonic kernels. "Integral" means that there is an integral involved, calculated from low boundary of integration to high boundary of integration. Direct Fourier Transform takes bounds from time domain. Reverse Fourier transform takes bounds from frequency domain. Time domain and frequency domain can be, as classes of specific cases, continuous or discrete. This theory is beautiful in its simplicity. For instance, formulas for direct and reverse transforms, in their traditional formulation, only differ in one sign in one place. The simplicity affords efficient implementation of the transforms in computer code.
And here we move away from the theory and arrive to a paradigm. The "suitable bandwidth limitations" remove information contained in original analog air pressure variations over time at the point of recording. Central belief of the paradigm states that removal of frequency components beyond 20Hz and 20 KHz is perceptually benign, for all types of music, and all listeners. Technically, this is the crux of our disagreements. I do not subscribe to this belief.
I value your opinion. Couldn't asked for a better illustration of what I had to endure in my prior discussions at ASR. However, my professors, from leading European universities, and their teaching assistants, had other opinions, giving me straight As on all courses related to Fourier Analysis and DSP. The theory I'm using today contains the classic Fourier Analysis, and classic DSP based on it, as subsets. I absolutely do use them in domains of their applicability, when I believe they are going to provide accuracy sufficient for a task at hand. Yet there is more, which came mostly from research conducted by others over past three decades. Unfortunately, too much of it is still widely dispersed in numerous peer-reviewed papers, rather than concentrated in a few engineering handbooks.
When one writes enough code processing real-life music recorded with high enough fidelity, one absolutely starts believing that such music components do exist: going from zero to almost pain threshold in a matter of microseconds, and then rapidly decaying. One of the best examples of music genres rich in such components that I know of is Indonesian Gamelan. It is a curious genre: worshiped by its devotees in native land, and almost completely ignored by listeners outside the region. Even the best gamelan CD recordings of famous Indonesian ensembles sound to me like incoherent early practices. Live, classical gamelan compositions, played with passion by experienced musicians, sound heavenly to me.
This depends greatly on the nature of the band-limiting filter used. For analog filters this statement is generally true, with understanding that perfect brick wall filters don't exist, so there are still some smaller artifacts to be expected because of that. For digital filters applied to stream of oversampled values in some ADC devices, not so much.
See my comment above. Depends on the nature of bandwidth-limiting filter.
I don't believe so. I used more sophisticated understanding of those. Knowing, both from learning the theory and from practical experience, that absolute predicted perfection isn't practically achievable, and that one needs to very carefully look at what artifacts are produced by this or that digitization method, and whether the artifacts may be heard under certain conditions by certain listeners.
Research on masking of signal by noise is kind of 101 of psychoacoustics. What we already established is that I'm more interested in how practically encountered noise is masking practically encountered quiet music passages. I do realize that masking thresholds for specially constructed signals and noise patterns may be different.
Yes an no. Implemented with analog circuitry, yes. But, at some point inside a class-D amp analog signal is transformed into a sequence of discrete +V and -V segments, starting and ending at analog time boundaries. So, it is kind of a hybrid. Analog in time domain throughout, discrete at an intermediate stage in amplitude domain. Not what most people would call classically digital, yet not quite purely analog either.
I see it differently. The old paradigm is falsified by phenomena for which it gives invalid predictions. For instance, according to the old paradigm, LPs shall be long gone, the way of cassette tape recorders and VCR video tapes. Yet LPs persisted, and the classic paradigm produces no convincing explanation as to why. New paradigm not only explains why LPs persisted for so long, but also specifically predicts what they'll be replaced with. To repeat once again, to the best of my understanding, eventually they'll be replaced by digital recordings with information density equal to, or higher than, those of PCM 192/24 and DSD128 formats.
Yet another phenomenon not explained by the old paradigm. According to the old paradigm, 16/44.1 format shall be sufficient for capturing and delivering any type of music, yet in practice all those "pesky producers and sound engineers", for some mysterious reasons, want to use digital formats providing higher information density. The new paradigm not only qualitatively explains this phenomenon, but also accurately describes the numerical parameters of the formats that were found sufficient by trial and error by many highly qualified practitioners. The new paradigm also explains why gear providing even higher information densities, easily available these days (e.g. I own several nice 32/768 ADC/DACs), isn't as widely used at its highest settings in practical music production. |
