objective vs. subjective rabbit hole


There are many on this site who advocate, reasonably enough, for pleasing one’s own taste, while there are others who emphasize various aspects of judgment that aspire to be "objective." This dialectic plays out in many ways, but perhaps the most obvious is the difference between appeals to subjective preference, which usually stress the importance of listening, vs. those who insist on measurements, by means of which a supposedly "objective" standard could, at least in principle, serve as arbiter between subjective opinions.

It seems to me, after several years of lurking on and contributing to this forum, that this is an essential crux. Do you fall on the side of the inviolability of subjective preference, or do you insist on objective facts in making your audio choices? Or is there some middle ground here that I’m failing to see?

Let me explain why this seems to me a crux here. Subjective preferences are, finally, incontestable. If I prefer blue, and you prefer green, no one can say either of us is "right." This attitude is generous, humane, democratic—and pointless in the context of the evaluation of purchase alternatives. I can’t have a pain in your tooth, and I can’t hear music the way you do (nor, probably, do I share your taste). Since this forum exists, I presume, as a source of advice from knowledgable and experienced "audiophiles" that less "sophisticated" participants can supposedly benefit from, there must be some kind of "objective" (or at least intersubjective) standard to which informed opinions aspire. But what could possibly serve better as such an "objective standard" than measurements—which, and for good reasons, are widely derided as beside the point by the majority of contributors to this forum?

To put the question succinctly: How can you hope to persuade me of any particular claim to audiophilic excellence without appealing to some "objective" criteria that, because they claim to be "objective," are more than just a subjective preference? What, in short, is the point of reading all these posts if not to come to some sort of conclusion about how to improve one’s system?

128x128snilf

Showing 21 responses by deludedaudiophile

@prof ,

Corollary to your post, just because we don't know everything about electricity, does not mean we don't know enough for audio :-)  I see that as a frequent justification in these forums for all kinds of what I consider ridiculousness.

I paid an acoustic engineer to design and setup the acoustics in my room. He took a lot of measurements and made adjustments to finalize the result. It sounded better than anything I had previously. For most music and most of my moods it is great. There are times, though, more my mood than anything, where I use the AV processor and one of the ambience surround modes. It sounds more alive and I enjoy it. Other times it just sounds fake.

My cheap headphones have never rocked. The moderately expensive ones, those rock. I recently purchased some IEMs on a whim that were measured for very very low distortion. If the measurement is accurate, the distortion is not that much higher than Sennheiser's stupidly expensive flagship. I have not heard that one, so I cannot make a comparison, but there is something sublime about the low distortion and perhaps it is the distortion without room reflections, but the proverbial "I am hearing things I never heard before" is happening in spades.

@mapman 


I think we can all relate to cars or even bicycles. I have 4 bicycles. They all do different things better.  One is objectively if the pavement is flat, much faster than all the others. It is not very comfortable for a leisure ride though. An F1 performs, but you would not want to commute in it. Some road cars excel at performance, but are difficult to drive at those limits, while others let average drivers excel. I like artists that make realistic painting, but not so realistic you feel you need to compare them to a photo.

Listening to audio is a personal thing. I am under no illusion that what I am listening to is similar to any real performance. It is manufactured. I don't feel I need to experience exactly as intended, because as intended is just manufactured by another human just like me with their own preferences and feelings. I did have a room designed and built that would eliminate what I can only call flaws as I don't think those ever help my subjective enjoyment. Now I have a canvas on which I can paint the music I want.

@hilde45 ,

I think that video needs to be paraphrased.

To me, here is how this subjective vs. objective boils down:

 

  1. There are objective audiophiles that believe that perfect fidelity, electrical and otherwise and adherence to a doctrinaire as the artist intended method, as the only true path. They are wrong. Why? Because all music is manufactured and the final result is determined by a fallible human.
  2. There are subjective audiophiles who believe there are as yet undiscovered physical properties of wires, electronics, and anything else they can come up that explain changes they perceive they hear. They are wrong. Why? Because while we don’t know everything about the physical world, our understanding of how thing work but even more, how things behave is extreme. Do you think we would be able to build nanometer scale semiconductors, nanometer scale battery materials, antennas at 10’s of GHz, etc. that behave just as our model predict if we didn’t?

 

One group says I trust the science without realizing that there is no science that guarantees the accuracy of the underlying fundamental product, the music. The other group says I trust my ears without understanding but more importantly accepting the basic limitations of us humans. If neither group is willing to accept the fundamental flaws in their approach and learn from the other, then neither will move forward.

Some like deludedaudiophile use the concept "accurate" and "noise" in a confused way

I am thinking one of us knows those terms much better and used them daily. I don’t think it is you. You appear to redefine terms as you see fit.

 

The method by which scientist can study the way introduction of noise can help non linear detection system is called: stochastic resonance method...

HOW non linear detection ears/brain structured system can use noise in a way an ordinary electrical detection instrument could not?


I had to Google about 10 minutes to form an appropriate response as there were some items I was unsure of.

Do you know a characteristic of a system where stochastic resonance will work? It must be bistable. Your leap of faith in another thread wrt non-linearity in hearing and DAC operation, missed that the researchers in the papers you linked highlighted the non-linearities in the physical nature of the cochlear. It would appear the bistable element in hearing would be neurons that relates to stochastic resonance. That would mean they have an element of quantization, making them digital in some fashion, not analog. Digital has fundamental non-linearity due to quantization too. Do you have anything that reveals limits of quantization of human hearing? If not, I have to assume it would relate to minimum hearing thresholds.

It appears that digital audio has been using this principle as well since its inception with dithering.

Which brings me back to how I should have started this thread and why I do not read your posts. Is there any point you are trying to make, because so far, you have not made one.

To give credit, you are attempting to relate experience to scientific principles even if, in my opinion, your attempts appear misguided.  Far too many posts come across as a call to magic.

 

Stochastic resonance requires bi-stable (or multi-stable), essentially it requires an analog to quantization, pun intended.

@thyname and @mrmb , do you not see the irony and hypocrisy that you are doing the exact same thing, but just selling a different form of religion?  That should be obvious to you.

I was not exaggerating. MRMB wrote 22 full paragraphs to accuse others of proselytizing. If that is not trying to sell a particular way of thinking I don't know what is. That tweak comment, to me, is very revealing.

I find it gross exaggeration that people who use measurements are only interested in measurements as a determination of quality. Obviously those people exist, but they are the exception. Most people I have met who rely on measurements in this hobby, especially when I was working through our custom room, use measurements as a tool to understand what is happening, or what is not happening. They use it as a guide to achieve a particular sound, as opposed to buy and hope, or playing mind games on themselves with questionable devices as @mitch2 identified. The term he used, "subjective rhetoric" was overly kind. I much more appropriate term comes to mind.

 

Thus, the goal of far too many objectivist posters, seems to be the need to save equipment buyers from themselves.

 

Are you familiar with the term hypocrite? What were the last 25 or so paragraphs you posted? Who were you trying to save.

 

Once, a really good frequency response level is attained, especially a superb, nuanced and detailed midrange, incremental improvements can found be in the intangibles and mainly unmeasurable areas, such as soundstage and imaging.  

 

You should have a chat with the acoustics engineer who designed and tuned my room. I believe the term he would use is poppycock. He does not think soundstage and imaging is at all intangible and many others I talked to do not either. Perhaps that comes from your lack of knowledge that others do not lack?

 

If you start with the assumption that other people could not possibly have the knowledge you lack then you are destined to repeat the mistakes they have long overcome.

 

There is enough animosity on both sides but not accepting the knowledge or experience of either makes little sense.

 

On a side note, the most revered headphones, very very expensive Sennheisers ($30K) have very very low distortion. They measure about as perfect as possible. Everyone who hears them raves. What can we learn from that?

 

 

 

 

Since we have not created full AI we have no real idea of how it will "think" and behave. Unfortunately or maybe fortunately we will find out.

What if they give you two glasses of the same wine and tell you they are different or put different labels on them. You then go on to describe how those two glasses are different when they are exactly the same.

Personally I think the subjective/objective debate is vastly over stated in audio. Most of it is marketing driven. When companies are trying to make you feel good or bad about your purchases based on what you believe and not the detailed proven merits of their products, then you know you are being played (or should know).

I see purely arrogance and ignorance at issue on both sides. In other words we are dealing with people and people don't like change or being wrong.

 

We get all these angry 'fact mongers' on forums, everywhere, on all forums..who don't understand this ..... and attack, demanding  their own self security be reflected by me, into them... a thing that.. which ultimately...cannot and does not exist.

Someone who both disagrees with you, has the ability to clearly communicate why, and back up their arguments with verifiable information is not a "fact monger". Facts are facts. They don't monger. They simply are. The only anger I perceive is from those presented with facts, not being able to refute them.

I will restate what I said previously. Anyone who tries to convince you to either feel good about your purchase or bad about your purchase due purely to holding a particular belief, and not by clearly communicating verifiable advantages, is trying to take advantage of you.

THIS, all of the above, is why, in the physics department and all of it’s subdivisions into science and training and academia.. that all of the professors, if pushed and asked, will tell you that there is no such thing as a fact, and all is theory.

Just how many university level physics departments have you been a member of, either as faculty, or a grad student? About the hardest thing to convince a fellow academic of is that their "theory" isn’t fact :-) ... but in more concrete terms, your statement is hollow and baseless. More accurately what they will tell you is that theories are for all intents and purposes factual, depending on the use, and that normally depend on the scale and/or how used. Newton’s second law at low velocities (wrt c), is accurate enough to be fact. Similarly Ohm’s law, and many other formulas used in electricity are accurate enough, where used, to be considered fact. They only break down when we approach very small scales (and theoretical discussions). A more modern model may be needed to properly model and develop a cell within a MOSFET, however, traditional models are many magnitudes more than sufficient to properly use that MOSFET in design and development of a Class-D amplifier. The same is also true of every other thing used in audio no matter the hand waving justifications given my people selling things.

 

More paragraphs. More ignoring the singular issue which is not even in the least bit complex. Either the two equal measuring components can be identified as different through listening alone or not. Stop complicating a ham sandwich. This is not a philosophical question.

Mahgister and teo_audio,

You put a lot of words down on this topic, but in my opinion they do nothing to advance the debate. To me they are nothing but special pleading sans any justification of said pleading offering no concrete relevance to the discussion.  There is a another more apt but colloquial term that I will not use.

It is already accepted that personal preference is real, and absolutely will deviate from whatever measure of accuracy one may choose, and that single components or certain groupings of components can cause these deviations resulting in their being preferred.

The only question that remains is can audiophile detect changes that measurements say do not exist. That is a yes/no question. It will be very difficult to prove no, since negatives are rarely provable. However, the yes only requires showing that someone can do in some experiment that is considered reasonable. Increase the number of people and number of experiment will increase the confidence in the result.

I think almost all here would agree that given the nature of the music creation process before it is in our hands, that objective accuracy is a false notion and even if true, that does not over-rule preference for enjoyment.

If you cannot answer the question of whether audiophiles can truly detect differences that modern test gear says is not there, then all those multitudes of paragraphs are quite meaningless.

I would argue it takes only average equipment in a good room. Before I had all my new equipment in my listening room, I had our old home theatre system in there. Not a cheap system but not what many here would call audiophile. Everything worked. Imaging was perfect. Sound was "balanced". When the main system went in the biggest difference was clarity at higher volumes and better bass from the new subs. At low-mid volumes they were surprisingly similar.

@mijostyn - pretty much a given that we will evolve to a human/machine hybrid. I don't see any other way for us to progress.

The human brain is the most astonishingly capable problem solving technology ever developed by evolution; why would we willingly give up exercising it? The machines will take over not by force, but because we've gradually ceded control over our own lives to "labor saving devices" that do supposedly trivial chores better.

 

You could argue that it opens up limited brain capability to more artful and altruistic or even scientific pursuits as opposed to drudgery, just like automation did with manual labour. I for one don't lament using a shovel or even a ox to plow my fields.

Of course gradually less and less people will be able to contribute in meaningful ways to society as AI becomes more capable. The obvious outcome is machine/human integration and an even greater divide between have and have nots.

@noske

 

Define capable. The prior paragraph does not support anything, it only asserts, or speculates. Then the prior words here may be examined

Since we are making things up, anything I says is as valid as anything else.  However, I will go with able to perform some function that someone else will pay for such that they may have a non purely government assisted life.

 

The obvious outcome is machine/human integration and an even greater divide between have and have nots.

A non sequiter, methinks, in context. [some words striked out for interference]

It is only a non sequiter to you, and perhaps others, but not to me since I can see a logical progression. In order to be of some productivity, to someone, i.e. in order to keep up, people will need to accept human/machine integration. At the bottom end of the economic pole, will be those who for economic reasons must accept the government supplied augmentations.  For those with the resources, they will be able to buy the best available. The divide between the haves and have nots will hence grow, from purely economic, to intelligence, effective life span, etc.

It will get ugly before it gets better.

 

 

 

A better corollary will be that just because we know everything that we can know about electricity, this does not means we know all there is to know about audio matter and experience :)

 

No, that is not a better corollary nor is it even accurate. The statement wrt electricity speaks to the ability to be able to model and measure all aspects of an electrical signal with enough detail to quantify the potential for change when converted to sound.

From my limited knowledge, those measurements are possible to much much greater accuracy and resolution than any identified or tested aspects of human hearing. Not just better, but much much better. I understand most of those tested limits are under controlled circumstances with ideal stimulus. With real music, the ability to detect is much much less. If we can test electrically to much much better than tested human limits and humans limits are much much worse with real music, then we can conclude with high confidence that our ability to test electrical signals is sufficient to indicate whether a difference will cause an audibly detected change.