@nonoise
Ok, so we agree that there are preferences and those preferences may not correspond to the Harman curve, for instance, which is an aggregate measure of preference. In DACs and amps the notion that there is a preference for distortion artifacts is simply at odds with valuing accurate reproduction, however. I'm fine with that. Folks be free, always, but I still have no clear idea why anyone gets bent about basic accurate engineering that targets fidelity or make astonishing claims without evidence about cables or power conditioners?
We do have a disconnect with scientific reality that is itself hubristic because it strays from (repeating myself) epistemic humility by suggesting all these listening preferences supervene on basic measurable facts about music reproduction.
|
@mahgister I've read most of this literature, including the new (for me) Kunchur paper (his previous paper on cables is neither relevant nor well-designed). I'm not certain, however, as to how to parse your long digressions on these topics.
For instance, if we can measure noise and distortion in audio equipment, it is valuable to reduce or eliminate it in the reproduction chain regardless of whether there are potentially complex heterodyning/non-linear ultrasonic interactions or whether hearing capabilities can be shown to have greater sensitivity than assessable via Fourier analysis.
If there is an additional claim that perhaps cables and other tweaks that are measurably irrelevant to the signal reproduction can actually be heard, it is still in the best interest of general epistemic humility to remain skeptical until such measurements/ABX hearing results (with the LTM/STM refractory suggestions in Kunchur maybe) can be found.
So, in the meantime, we just get great measurements from ASR and can merely speculate that something might be missing, not that it demonstrably is missing.
|
@ricevs @mofojo
If there are all kinds of sonic properties that can't be measured according to these well-researched and standardized methodologies, how is it that engineers are able to formulate and use science/engineering to design the components in the first place? Do they just combine metals together and listen to them until they are happy? Do they just hope that they can get the right mix of tubes and opamps?
Your ideas don't seem very well thought through.
Every engineer is at some level applying science and engineering to the technologies they work on. In civil engineering they measure the soil properties to identify the correct foundation and footings. In ME they measure the elasticity of materials (like speaker cones). In EE we measure the transfer functions of systems and characterize noise and distortion properties and channel capacity and signal attentuation.
How is it that consumers of these engineered products have developed a mythic idea of how it all works? Measurements are core to engineering audio products. The opposite is faith mixed with marketing signals, which is fine for those who are so committed, but is not how the actual products get created.
|
@nonoise
OK, so the engineer/designer has used measurements and therefore understands the sonic characteristics of the device, but you are emphasizing that after it is complete some people have preferences about the different measured outputs? That's not the same as saying that there are things that aren't measurable, which many suggest above, or that there are secret measurements from special pixie dust that are some kind of hidden knowledge.
All it means is that people have...preferences, like enjoying the distortion signatures of tube amps because that is what they are used to. If, however, we value high fidelity to the original recorded music, then we know very well that removing noise and distortion (which are measurable) results in close approximation to the original.
It's really very odd to argue passionately and affirmatively about something that just amounts to preferences, hopes, pleasurable mystifications, deepenings, and so on. Just ignore ASR if it doesn't add value to your hobby. I value good engineering and high fidelity, but the glow of tube amps is really neat too. It doesn't bother me that some folks like to listen to high distorting amps or think cables make much of a difference, but there is a kind of gradualistic erosion of these myths that forums like ASR help to promote, which is good for me in helping avoid crap products!
|
@mofojo
Ummm, no, that is wildly inaccurate. Please read some ASR reviews and read/watch the primers to get an understanding of how measurements are performed. There is no pink noise, but there are various frequency response measurements, very similar to the way that DIRAC or other room calibration approaches work because that is exactly what they are emulating.
|
@audition__audio
I'm unclear on your points. You seem to begin by claiming that poor measurements might sound "more correct overall" which begs the question of what "more correct" means? I think you are smuggling in preference; what you like is somehow more correct than the preferences of others. I've less certitude beyond saying that accurate reproduction fidelity is mostly my preference.
What is in fact "completely false" about Amir's statements? Is it just that you claim, without evidence, that a majority of people disagree with accurate reproduction or flat frequency responses, etc.? That seems unlikely given the very high interest in ASR based on visit frequencies.
And, finally, no clear idea what sublimities you are fishing at in your final paragraph or why you consider his listening "mechanical." Is it because he uses different language to describe his listening outcomes or doesn't tarry sufficiently about some aspect of your preferences for listening adjectives?
|
@audition__audio
Well, he just provided listening test evidence above that suggests that experience, belief, and listening are not accurate or discriminatory. Therefore at least some of the group you claim to represent are actually wrong in their beliefs. The bundling of that is firmly tied up by known properties of human cognitive bias, everything from frequency effects to sighted bias and individual differences.
As far as characterizing the spending habits and wealth profile of ASR visitors, I wouldn't speculate. Speculation is an imaginative way to fail when not held at arm's length as a mere hypothesis that must be backed by facts and data.
|
@nonoise
I think you may have misunderstood the statement: audiophiles were in the less-trained cohorts (with reviewers as their proxy; trained listeners were trained by Harman), therefore it goes against the claims...
|
@audition__audio
There is actually a vast well of acknowledgement of the applicability of these measurements, though there of course remain deep discussions about aspects of their relevance to certain areas of audio science and listener experience.
The creators of audio equipment are engineers and use all the same principles in designing the gear that you enjoy. Noise and distortion are characterized and measured and how to manage them is part of the design process. For something like tube amps, there are deep technical issues in harmonic distortion components, and so forth. You can go to the primers at ASR on the relevance of measurements to understand more.
I disagree that it is a waste of "our" collective time. It instead provides depth and clarity, debunks long-held myths, and even improves the quality of components as a reflective market signal (see design changes by Schiit Audio, for instance) delivered by the community of users. That is a new and rare kind of social and capitalist phenomena.
|
@ricevs Well, it’s an interesting series of claims that I’ve read before but it’s not clear why that might be the case. How can we hear differences that aren’t somehow present in the audio signal and therefore measurable? It’s very incongruous and does not correspond with ordinary science and engineering principles, or with even everyday logic per se.
Do note that manufacturers’ specs can be false and also that a specific unit may be broken. Testing by a third party like ASR can help to ascertain the reasons for the differences, not always perfectly, but they would add additional support to these apparently tendentious ideas about these products.
|
@ricevs Of course it may be possible to hear differences between DACs. If one is poorly engineered and introduces significant noise and distortion while the other is state-of-the-art (SOTA) in terms of SINAD, then SQ may be distinct. Such differences will also be apparent when the two DACs are measured. You may be confusing this with the argument that two SOTA DACs without other specific flaws are not likely to have audible differences.
|
@deep_333 Sure, DSP can change the way things sound, but I'm not aware of any manufacturers who don't make such features switchable and therefore hide away their secret sauce. Pretty much all DAC makers strive towards accurate reproduction first and DSP features are add-ons (even if their algorithms are proprietary for this add-ons).
Seems like an odd and unlikely claim but I'd be interested to learn about specific instances!
|
@ricevs Actually, you are inverting the way science and logic works. We have theories like linear and non-linear distortions, as well as noise, are introduced into the transfer function of a system due to a range of thermodynamic and quantum interactions in stuff like semiconductors. The effects of these distortions and noise manifest as producing spectral spray, overtones, harmonics, etc. in the output signal. This spectrum can be measured with some accuracy using methods like Amir applies.
There is no evidence for the presence of other types of unmeasurable phenomena. We can never prove that there are no other distortions, noise, etc. We just have no evidence for them.
It is the positive task for a researcher who creates a theory that there are other measurable phenomena to prove that they exist.
I listen, but I doubt you know.
|
@ricevs
Well, the arguments presented above don't resemble a fantasy and flat-Earthers are the ones holding fast to irrational beliefs when presented with evidence to the contrary, so I'm both amused and confused by your presentation. You are also directly contradicted by Amir's discussion of cable testing, above, but I do encourage you to continue to research, learn, and perhaps someday ABX prove some of your ideas!
I personally have no particular ego investment in audio equipment but do like trying to understand online communities and how beliefs (and fantasies) develop.
|
@nonoise I do try to be careful in my use of conditional language...epistemic humility again about ideas and people, regardless of your characterization.
I'd love to be shown an anomalous result where there is in fact more than "meets the eye, and scope," but just haven't seen any evidence for such things yet. The Technics anecdote is curious but needs facts, data, measurements--proof of any kind. The anomalous results, when they arrive, are the great point of cognitive reorganization. There's a great beauty in that sense of impermanence and the arrival at a new plateau of understanding.
Sorry to wax poetic, but there is a sensual numinous feeling to good science for me. Bring me the novelty, but make sure it is fully baked!
|
@ricevs Cool, man! I'm just glad that all the engineers and scientists I know and have known use rigor, measurement, and scrutinize one another using mechanisms like peer review before reaching conclusions and are humble enough to admit when in error.
The known facts about cognitive and perceptual bias tell us to be a bit more careful about being "our own source of truth." Nothing sad about it! We are beautifully flawed.
|
@knock1
Not sure what you are getting at? Yes, we have theories explaining the existence of noise and distortion in audio reproduction. Those theories are used to develop engineered products that, in turn, reduce noise and distortion. That's how this science and engineering game works. Note that a "theory" in science is not just a speculation:
Scientific Theory
The critical language cloud includes ideas like "testable," "make predictions,"withstand rigorous scrutiny," etc.
|
@nonoise You are on to something there, even half in the bag! 🤪
Medical placebo effects do work surprisingly well with specific brain-mediated factors, like pain management, insomnia, stress-effects, etc. And auditory perception is definitely in that camp.
But, but, there is a great opportunity to design experiments that do some kind of preference testing/difference testing over longer exposure windows. I'm not aware of anything like that though it may exist...anyone?
Sounds like a very expensive experiment for a new PhD at JBL...
|
@kevn I think you are mistaken about several things in your post. And I am very aware of how electromagnetism works and the recent YouTube discussions of some of the more confounding aspects of electromagnetism. Like quantum mechanics, sometimes intuitions drawn from everyday experience seem a bit odd, but the equations typically serve as a bedrock for analysis, and we know that audio at low power and relatively low frequencies is fairly consistent in following relatively simple electrical law-like patterns.
First, Amir’s measurements of speakers are done by a microphone listening to the reproduced tone sweeps, so whatever special claims you make about magnetic flux are largely irrelevant to those measurements.
Second, measuring the signal at the front end of a cable and the back end and comparing them should show your suggested e-mag influences somehow...and they typically don’t (note careful addition of "typically").
Third, most folks don’t really know what "science" means in the modern world. There is a long trail of Philosophy of Science from Logical Positivism through to Popperian Falsification, then social models of science a la Kuhn and Feyerband, and then the actual praxis of science that is built up around institutions. Applying ideas derived from science to assess audio products is different from, say, figuring out how alleles affect phenotypic traits in molecular biology, but it’s also more similar to survey work in some of the social sciences. At best we can think of it as deriving from a commitment to deep scrutiny of empirical methods.
There’s nothing indoctrinating or brainwashing about that! It’s just a reification of being careful about everything and using tools with equal care.
|
@laoman You dropped the context! Unfair! 🤣 The end of that was a recommendation to do an ABX test after researching and learning about how to do so. Hardly condescending, I think, given the target of the response. I'll note also that I only have a few dozen posts at ASR and read it less than Roon (but certainly more than this bizarro world of Audiogon!), so not much of a minion.
I'm mainly interested in how online communities deal with science and pseudo-science, and how they evolve and change over time. The backlash over ASR is like a little experimental microcosm. It's kinda like the car communities where brand loyalty and electric/petrol divisions rule, but it has a strange aversion to measurements, unlike car-universe where 0-60, 1/4 mile, Nürburgring performance are all critical. It's like if a car community said that 1/4 mile didn't matter, just the driving experience and pedal feel.
Very interesting and odd!
|
@jtgofish I just crawled around Hi Fi Choice's online presence and searched for blind listening tests and they merely mention that things "really impressed the blind listening group" and similar such short statements. Is there somewhere where they archive their actual test methodologies, results, statistics?
|
@kevn Well, I try to be charitable in my critiques but there is a reaching quality to much of the posting here (combined with hostility, often). In this case it is an "appeal to possibility" and the standard worry over closed-mindedness.
I salute your interest in the idea that science is incomplete and there is much more to know, but ask what specific critique you have about what is currently known and applied in the case of ASR testing and how a better understanding of electromagnetism might improve upon that?
|
@deep_333 I've only very occasionally posted anywhere, actually, but do please ignore if you don't like my content. I've been an actual working engineer and scientist for most of my adult life, though not specifically in audio. Like I mention above, I only occasionally read ASR but am very interested in how ideas like science, quasi-science, and pseudo-science arise, are discussed, and debated in online communities.
I guarantee you that I am not Amir and don't have his depth of understanding of audio engineering issues.
|
@kevn
Right, I have no deep objection to most of that. I was curious what aspects of electromagnetism do you think are in fact contributing meaningfully to the high fidelity reproduction of music that are not contained in those measurements? You seem to have dropped that?
The rest of your rather long description is arguable but mostly not objectionable. Your characterization of how science works vs. Amir vs. technicians, for instance, appears to continue your general dislike of the presence of "science" in ASR.
There clearly are aspects of science in there up to and including his scientific survey paper on the performance of audio components. The measurements are part of the survey and the hypothesis is that there are not strong correlations between cost and measured performance. Now, you can argue that other sciences work in different ways than that but it follows the description that I previously provided. No scientist I know would object; they just do their science. But are you worried that the imprimatur might give new audio equipment "seekers" some kind of false belief that all they need is this particular kind of ASR science? I'm not too worried!
Your final paragraphs focus on the notion that listening in a specific space and matching components is in addition to the capabilities of the components. Of course! There are extensive and lively discussions on the science of room modes, the role of DSP, damping, integration, impedance matching, and many other topics at ASR. The measurements of the individual components and the listening by Amir to confirm aspects of those is not changed by the additional discussions or how any given audiophile will need to find the proper fit of the components to create their optimal experience.
I personally think much of it is overwrought and it is oldie audiophile snobbery that trends towards the negative stereotypes that are claimed (minions, brainwashed, etc.) for ASR folks in a reverse-golden-rule manner, but I hold out a kind of stochastic hope that given enough discussion and enough resources, searchers will begin to find and understand enough that some of the mythologizing that has been wielded for commercial gain will be abraded a bit.
Like I note, I'm curious how internet/social media change our engagement with exactly these kinds of topics that were once knowledge constrained by dealers, small-circulation magazines, audio shows, etc. I've never been to or subscribed to any of that stuff but it once was a contained marketing ecosystem that built stories as much as performant products. Times change!
|
Wow, such revelatory hostility! Dealers circling the boutique wagons in glamping configurations and boasting that their listening punditry is better than this stale, shrill modern stuff.
But we can do a bit of the kind of brainstorming that goes on in disruptive/dintermediating business plans. The boutique/high end audio equipment makers should produce mid-fi but excellent measuring components that become leads to the high-end components. They can reduce costs the standard way with manufacturing in Asia, use well-engineered chip amps, etc. It doesn't matter that those mid-market offerings are competing with many others. As long as they perform (measure) well enough they will get sales traction.
But, more to the point, they become brand ambassadors for the dealers and the higher-margin components. So what if they are mostly the same just with 60 pounds of heatsinks to allow for an additional 200 WPC into 8 Ohms. The high-end buyer may be trading up from the middle tier.
Brands that do this successfully include KEF (LSX -> Meta Blades), Chord even, Revel, etc. Now that sounds like a business model that can embrace the reality (and future impact) of ASR's measurements dashboards. The boutique shops just need better market planning and implementation rather than just being engineering (and measurement and, yes, listening) driven.
|
@audition__audio Excellent! I'm glad you are off to ponder proper and respectful communications!
|
@kevn Take your time! I just got back from the symphony and got caught up myself. I just do this for fun anyways. I have neither ego nor business nor status invested in this topic but do find it fascinating, like friscalating light through a dusty chandelier.
|
@kevn I'll just add one footnote to my previous post: we might actually be able to address the specific issues of heterodyning and nonlinear cochlea interactions in audio by using DSP to simply mute tones that interact in those areas of the hearing range. This would be like addressing a room mode but within the ear itself. Of course, we would be robbing the signal of its fidelity in so doing.
Still, in order to do this we could use experiments that first demonstrate it will improve human hearing. There is a great deal of literature on methods for overcoming hearing loss; there may be something in there concerning speech that points towards something useful for audio equipment design.
Let's get that gap filled in!
|
@kevn I previously addressed the issue of these hearing-excess-of-Fourier arguments as well as heterodyning and nonlinear effects within the ear. The problem isn't that there are interesting experimental results, it's that they don't demonstrate that there is anything that can be done to audio equipment to implement better solutions to whatever gaps may be present. For instance, if I am a DAC designer there are several different pathways to accurately reproduce a signal but there is no theory that says one approach will improve over another in matching the nonlinear merging properties of higher and lower frequencies in the cochlea.
Now, you can suggest that somehow listening on the part of the designer is allowing them to choose between design pathways but this is just speculation. It may be true, as I noted to @mahgister, but we don't know and neither does the designer.
So there is a certain faith built into all this speculation, just like god-of-the-gaps arguments in other online communities ("listening-in-the-gaps" arguments has a nice ring to it!). It's interesting but needs proof and a proper measurement methodology that shows a path forward for determining exactly how these phenomena impact equipment design and use.
Since you are a bit of a student of ideas in philosophy of science, one key one in contemporary thinking on the topic is lifted from Wittgenstein that we must remain silent on things we have no knowledge of and we have no knowledge of this. Until we develop it sufficiently we do have an AP and spectral sweeps.
|
@ricevs I think your metaphor may have gone awry there towards the end since we can measure the differences in performance between the Civic and Corvette?
|
@ricevs Well, it's a collection of nice sentiments, sure, but I don't see any reason not to include ASR rankings and reviews in one's journey of music and music equipment enjoyment...just another data point! You indicate some at ASR are mean but, hey, the world is filled with many types. I just ignore them and focus on the value-add.
|
@fleschler I guess I'm still baffled by this kind of sentiment. It doesn't impinge one iota on your preferences for others to state that X or Y has better objective performance than some choice you made. You b u!
If someone says something like, well, there is no evidence that cables make a difference, to me it's worth learning why they make that claim and what reasons and testing that they are citing to back up their claim. You can of course carry on thinking that cables do matter and you can hear a difference, but at least you have learned an alternative perspective.
I guess I side with @ricevs on a happier way of engaging with online communities, though I'm calmer in general and not quite so ecstatic! 😎
|
@nonoise Gradualistic repositioning of the audio Overton window 😂
|
@kevn Well, we certainly do seem to talk past each other. Your responses demonstrate that you are not quite internalizing the problem that I put forward. Sure, we have this great hearing capability and some experimental results that suggest some curious little edge phenomena but we have no way of actually telling whether it is useful to us in the context of music reproduction and listening.
We can't design systems that exploit it or, if we do, no one has demonstrated that they have done so. For instance, let's say that human hearing can detect aspects of music that are not distinguishable based on the spectrum and that the designer figures out a way to use that to improve the system, the way to establish that is through a consistently reliable ABX test that shows this remarkable new achievement while the signal measurements are otherwise unchanged. Or, the other way to approach the problem is to figure out how to measure the ability down to the level of granularity of the human ear, perhaps using approaches that are not applying FFTs for analysis. Then, voila!, we have a new measurement regime to use for designing equipment, etc. Measurements prevail again!
The other possibility as I discussed is to figure out a theory and model for how it improves on our ability to process music and then incorporate those findings into the design phase. Hence my suggestion that there may already be some insights in the hearing-loss literature...I'm not sure how you read my comments on that as somehow suggesting I wanted to improve age-related hearing loss! To be clear: "Still, in order to do this we could use experiments that first demonstrate it will improve human hearing." means that we need experimental results that show that my proposed method of DSP-ing to counteract heterodyning/nonlinear cochlear phenomena would actually improve on the listening experience, improve our hearing, rather than artificially scramble our natural interpretation of the sonic landscape, reduce our capabilities or just be neutral.
WRT the Wittgenstein quote, I was mentioning that it has been internalized as a way of expressing the epistemic humility of contemporary science crossed with the requirement for positive evidence of novel claims, but thanks for the clarifying context!
|
@kevn Well, fair enough, but you have not demonstrated that human hearing exceeds those measurements for music listening purposes! If we had just one great ABX test that showed me wrong, I would be thrilled because that would pave the way to something new. All those dynamics would dance again and the mad scientists who brought the systems to life would be celebrated for rightly finding a path towards a new audio Xanadu.
My focus on equipment design is a foil for, as the cognitive scientists say, requiring an "effective procedure." That is, if you can build it you demonstrate you understand it.
I'm certainly indoctrinated in the epistemic humility to be as careful as possible in assessing ideas, my own and those of others who hope but have not fully honed those hopes with the calm clarity of rationality.
Best to you, too!
|
@kevn I’ll note also that I think you may be misinterpreting the Fourier uncertainty principle in this particular context as I and Amir have mentioned to @mahgister in several contexts. The authors are showing that if you used Fourier analysis as a model for human hearing there are limits to its applicability because there is likely nonlinear bucketing that allows for discrimination of time/frequency in excess of what a linear system is capable of.
That in no way impacts the FFT of an audio spectrum which is an accurate frequency domain representation of the sounds up to its intrinsic limits, it’s just a showing that there are likely nonlinear ways we sort out sounds. The speculation is that the fine acuity is derived from evolutionary pressures and the mechanics of it are due to the shape of the cochlea (as I repeatedly discussed previously).
So when we focus on the specific topic of sound reproduction, the sounds so reproduced contain as much information as always (and with greater fidelity the more accurate the information is). It arrives at our ears and then we exceed the FUP. It has little to do with equipment design or assessing reproduction fidelity.
Still, it is an interesting little edge phenomena!
|
@audition__audio Well, we know that the experimenters were able to produce the test sounds and play them for subjects without difficulty using (likely) average quality audio reproduction equipment, so it's obvious that conventional reproduction systems are not masking the dynamics the subjects report, but I can change "edge" to "newly found" to remove any stigma the term invokes!
|
I like to summarize online discussions to try to tease out what the central points and realizations are. Often they amount to the collective desires of the participants to simply participate, which is understandable. I read the Roon community for specific details on new Roon Ready certifications, the fate of MQA, and technical details about using MUSE. I read ASR for reviews of new (and old) products and how they objectively perform, as well as new insights about the science and engineering of audio systems.
Here at Audiogon, however, I just check in after I get a Friday summary list and really don't see much new information at all. We have committed listeners who tell long narratives about how trusted friends told them about a product, or how everyone should try some new cables, and how listening convinced them of this or that. But what we don't get is any real or actionable information beyond "If you liked Nordblost's Mjolgurniator III, just wait until you can trade up to MRT's Fusionator 3000!"
I've yet to discover something novel beyond the brief deep-dive that I participated in above (and was prompted by the prolific and occasionally challenging @mahgister) concerning how exactly listening and measurements might have a divergence...at least currently.
Now I might be biased slightly towards novel and actionable information that has some depth to it based on my background and passions, but I am curious what other contributors get out of all this bashing and clashing, promoting and diminishing?
I'm curious why folks argue and contribute here besides the obvious commercial interests of dealers cultivating sales (a bit of a sad and fundamentally small market to be captive to, alas)? I'm developing a book on the topic, so any insights/confessions/realizations are of interest to me!
|
@kevn Well, very expansive but we actually can develop systems to do such measurements using precisely the same approach that nature uses: nonlinear systems. It's not terrifically mysterious; we do it all the time in optical systems that shift frequencies just like the heterodyning that is described in the paper. I will admit that the mathematics is quite challenging based on experience. Being nonlinear we sometimes have to use things like spectral analysis (wow, strangely familiar) to look at solution families.
But the problem is that no one has actually successfully applied any of this to designing audio equipment! Or at least no one has demonstrated that to be the case!
|
@kevn
Weird misunderstanding going deep!
(1) There is no ABX test that shows that a measurably transparent audio system is distinguishable from another based on additional properties related to nonlinear perception capabilities.
(2) I still think you are misunderstanding the thrust of the paper: they are arguing that a way of modeling human hearing in the past has been to isomorphically map it to a linear frequency breakdown of the signal that is spatially spread within the cochlea (see Kunchur, etc. for additional details). The difficulties of this are that tapering of the cochlea (at least) leads to nonlinear phenomena. The FUP is a mathematical idealization that limits the simultaneous resolution of time and frequency for a linear system and they use it to explain how nonlinear systems can overcome this resolution in this case.
(3) Per my previous points about simulation as the gold standard, here’s the skeleton of your argument:
(a) Human hearing surpasses the ability of linear Fourier systems to resolve micro-phenonema in timing/frequency.
(b) Since FFT measurements use linear Fourier systems they may present measurements of audio systems that do not include this micro-phenomena.
(c) Therefore, using human hearing to design audio systems may achieve improved results over the FFT results shown by measurements.
(d) And, in conclusion, show me any such realizations via ABX tests that demonstrate that a designer was successfully able to tune a system to, in fact, achieve (c).
(e) But, per my previous posts, there is the opportunity to also improve the measurement apparatus to account for the discrepancy or to develop an impactful theory about how nonlinear cochlear phenomena might add to the music listening experience. These do not exist and therefore there is no path yet.
Well, I think I’ve repeated myself 3-4 times!
|
Since summarizing has great power, I'll just update my previous efforts to build a syllogism:
(a) Human hearing surpasses the ability of linear Fourier systems to resolve micro-phenonema in timing/frequency.
(b) Since FFT measurements use linear Fourier systems they may present measurements of audio systems that do not show these micro-phenomena.
(c) Therefore, using human hearing to design audio systems may achieve improved results over the measurements or, conversely, if there are aspects of existing systems that exceed the measurements those may be heard by human listeners.
Now, (b) is contested by Amir along several lines of argumentation per above, but what is not yet contested is that:
(d) There are no experimental results that demonstrate there exist audio systems that show (c). The easy way to do demonstrate (c) is with an ABX test with proper controls.
Therefore, we don't know if (c) is true.
|
@cleeds No, I haven't, but have helped run human subjects in experiments, so I hear you! Achieving accuracy with human subjects is very challenging.
|
My initial read on the comment paper is that at sufficiently high sampling rates for the FFT, the effect goes away, though the authors are using a windowing Fourier x-form that I am not fully familiar with or the implications of the free gamma parameter that they set to the variance of the initial pulse.
There's a rich literature on using accumulation methods in image processing to overcome Fourier uncertainty and one implication might be that it's not so much a nonlinear effect but just stimuli accumulation in our cochlea and the neural systems that process the data.
I'll do a bit more digging but it looks increasingly like (a) might be untrue and (b) might also be untrue in my new syllogism.
|
Since I dove in, I have to deep dive! Not definitive, but an interesting data point:
|
@kevn It's always a journey to savor learning new information. I really haven't a clue why you or the crew here are mildly to ragingly hostile. As I have mentioned previously, I have no investment in these topics. They are interesting, is all, and I have the luxury to treat them as such with no commercial, ego, or other interests. I hold them up like a jar of fireflies and they are tulips of fire against the background nothingness.
I am, however, developing a series of micro-theses on how online communities shift over time, like enlarging the Overton window, and how certain media theorists apply framing theory or pumps/valve metaphors to the dynamics that shift and change online sensibilities. It's helpful to see it up close and rough-and-tumble rather than just in theory.
(takes an additional note)
Best, as always, and in your friendship.
|
@kevn Sure! That's cool... I will contemplate!
|
@kevn I see no such word games at all. I have been completely honest and my distilling down the discussion to the syllogism-like structure was an attempt to put you at ease since you like to write exceedingly long about issues like your distinction amongst fidelities and seemed ego-driven to dominate an animated and uncertain topic. You were also misunderstanding my points and I'm guessing I was some of yours.
My points were from the beginning exactly what ended up in the clarifying syllogism, though they may not have been sufficiently refined at that early stage. They were exceedingly relevant it turns out: there was no clear way to engineer a product to take advantage of the hypothesized phenomena; there was no established evidence that listening can exceed measurements for engineered audio products; there might be some additional insights in certain bodies of scientific literature but remains an undiscovered country; etc.
That's the gist: we don't know what is true until we discover it. We apply epistemic humility (sorry, I'm a champion of that phrase in this charged world). We sort through possibilities. I pretty much always operate in good faith and don't express certitude when it isn't warranted (attach your mildly uncomplimentary adjectives as you want...in good faith, I suspect you think! 🤣). I still say (a) might be untrue...the future is unknown.
In the end, though, here's what we have: Amir's measurements are currently effective for determining the relative quality of sound reproduction across a range of audio devices. The mysterious "deepenings" that were ginned up out of a few preliminary scientific findings didn't dethrone the value of ASR reviews. Those reviews continue to be high quality and valuable to me and to a large population of online readers.
|
@nonoise Wow, you put a lot of effort into that!
|
