SUMMER MUSINGS: "How Much Difference Does It Make?" - On Balance and Adequacy for audiophiles.

The other day, I was reading Jim Austin's "As We See It" post on Stereophile titled "How Does the Music Make You Feel?". A relatively sensible article putting forward his and the magazine's perspective around finding "balance" between the subjective and objective sides of the audiophile pursuit. I agree with the central thesis, ultimately it is a subjective decision how one wants to spend time engaged in this hobby based on the emotional effect of music (I hesitate a little to say "music" here because let's be honest, not all audiophiles are in it for the music - sometimes it's the "sound" that's being sought). Music is art and appreciation of art will always be subjective, perhaps even wildly idiosyncratically for each person. Nonetheless, it's not that simple is it if we "look at the forest instead of the trees"? Let's take a higher level perspective and consider the audiophile hobby as a whole, rather than the audiophile him/herself. Are there some principles we might want to keep in mind?

Remember that what I've called "hardware audiophiles" are actually the niche hobbyists Stereophile is speaking and advertising to. I don't think Stereophile is of much interest to the multitudes of "music lovers" who probably are quite happy already with decent bitrate MP3 streaming over cell phones. As discussed before, these two hobbies intertwine but IMO are not the same. In every Stereophile issue, while some pages are dedicated to albums, music reviews, or maybe interviews with artists, the contents mostly discuss hardware products used in reproduction of the music. Our emotional response to reproduced music is a combination of the artistry in the music and science of reproduction. The audio devices used have technical characteristics that are objective and can be measured. As an analogy, in the visual arts world, we might or might not subjectively like the Mona Lisa, but when we look at a reproduction (photograph or scan of it), we can certainly experience and measure the differences between high vs. low resolution copies, and we can comment on or measure other objective traits like color saturation, accuracy, brightness, contrast, etc. compared to the "real thing" under the lighting conditions at the Louvre.

In reading that article by Mr. Austin however, between the personal stories and opinions, I noticed that like in most editorials covering this topic over the years in magazines, it's side-stepping a very important dimension. Achieving "balance" is not just about the way it is done (subjective listening +/- objective measurements), but also about the depth of exploration from which we derive adequacy. In other words, it's not just about whether we subjectively look at a car/woman/man and be captivated by its/her/his beauty, or whether the thing/person is objectively faster/taller/shorter/skinnier, but rather, are these qualities "good enough" for our intentions?

Let's expand on this...

The balance between subjective and objective methods Mr. Austin spoke of appears to be only in the way we determine what is "good" versus "poor". A "subjectivist" hardware audiophile might hone in on his/her emotions and preferred sonic characteristics during playback while an "objectivist" might want to make sure the measurements for the device suggest a high level of performance (along with sounding good of course) when trying to determine the value of a product.

But what of discussions around the depth to which the sound quality achieves our intended goals? The adequacy of whether idealistic subjective / objective goals have been achieved was not touched upon in that article.

Over the years, while I have spoken of objective and subjective ways to determine quality (and here), I've tried to avoid rigid splitting between the extremes. As individuals, we have the right to choose anywhere along the spectrum between "pure subjectivism" and "pure objectivism". As you probably know, leaning "more objective" has been my preference. For the most part, I can respect most positions even if I do not agree. Where I do take issue is when certain individuals express very extreme views, appear unable to discuss rationale, or cannot take a stand with a reality-based perspective.

I have on a number of occasions used the word "neurotic" to describe the audiophile pursuit. I trust none of us want to be labeled as having neurotic tendencies, but let me define it here in a way that I hope nobody gets upset with the use of the word. Neurosis in my intended usage refers to the idea that we as audiophiles have an internal desire that is above and beyond the "normal" populace to achieve a higher level of sound quality with our playback systems (core pursuit of the "hardware audiophile hobby"). Often, it's like an "itch we can't scratch". How we contracted this "itch" is probably both a function of predisposition and formative experience. Complete satisfaction is difficult for us. For most of us, hopefully it's simply a "healthy obsession" that doesn't intrude negatively into careers, families, friendships, etc.

But what if the depth of the obsession is never satisfied by a reasonable sense of adequacy and intrudes negatively into daily function? This is when "neurosis" gets to the point of being unhealthy, doesn't it (eg, jobs, relationships suffer, etc...)? It becomes a "disorder" as in "obsessive compulsive disorder". Maybe for some it leads to unhealthy hoarding of music and albums. In the old-skool psychoanalytic literature, when mental imbalance gets to the point where the individual loses touch with reality, that propels him into the realm of "psychosis". I trust that for most of us, we can still laugh at ourselves and our neurotic obsessions around listening to and owning $$$$ gear, salivating at beautiful fascia, huge monoblocks, and gorgeous speakers weighing that of an unhealthy corpulent man. But I suspect the vast majority of us would have problems if a fellow audiophile started spewing psychotic material about conspiracy theories or fantastical pseudoscience in the face of obvious financial gain if some of these claims were to be believed and products sold. Holding on to beliefs such as those promoted by Machina Dynamica (I seriously hope this is just a spoof of the craziness that audiophilia can get rather than them taking anyone's money!) would be a great example of psychotic "wish fulfillment" lacking in common sense reality testing. Lesser forms might be better categorized as "snake oil", and the most shameless purveyors may be more on the "antisocial" spectrum financially taking advantage of the unwise/vulnerable audiophile rather than being truly psychotic themselves.

[I certainly do not want to make light of something serious here... Anyone with OCD, hoarding, or challenges with thought disorder, please consult your local mental health professional.]

I suspect it is because audiophiles can become so obsessed with the hobby that sometimes they will forego strict reality testing and allow "snake oil" claims to infiltrate significantly into the products sold. I don't think this is a good thing. Of course an audiophile press, perhaps encouraged by the profit motive, unable to call out such companies, or even show adequate reality-testing themselves doesn't help matters. To maintain balance, there is always a need for some mechanism to self-regulate and arrive at truths; IMO, this is a role the educated free press was expected to play, and seems largely missing in the audiophile mainstream these days.

How then does a reasonable audiophile reign in our neurotic tendencies? I think the answer is already found above - be mindful of balance and adequacy.

... Sense of Balance between objectivity and subjectivity

I personally believe that all of us should find our place of balance in the continuum between the extreme forms of subjective and objective analysis. Audiophiles IMO should be educated in the 'skill' of listening so that they can experience what "neutral" and "transparent" sounds like. Sometimes, I've seen reviewers call what are regarded as objectively excellently-measuring devices as sounding "clinical" or "hi-fi" as if this is a "bad" thing (a subjective judgment of course). What if some of these guys just like to hear music with "rose-tinted" EQ, higher noise, or "euphonic distortion"? I'm not sure if these kinds of reviews tell us more about the devices or the man.

My belief is that one should start with trying to achieve objective transparency, and from there, we can decide for ourselves whether EQ or intentionally adding distortions might subjectively "sweeten" the sound quality. I suspect local audio dealers can play an important role in putting together relatively neutral systems and good sounding rooms to demonstrate what high-fidelity sounds like. Beyond listening, I think serious audiophiles should also know how to interpret objective test results and have some general awareness of what "good" results look like. This is how we can each judge the engineering capabilities of the devices we might be interested in purchasing.

Remember that at its core, the hardware audiophile hobby is a technological pursuit. High quality storage, retrieval, amplification, and transduction of electrical to sound waves are based on science. Engineering is the method by which these physical and electrical principles are realized in the products we own. As far as I can tell, there is no such thing as an audiophile device that changes the sound significantly without at least some measurable effect using today's test gear. Think interconnects and power cables make a "huge" difference? Show us how this manifests in the measurements! When a company proclaims that their products are capable of massive sonic improvement but yet cannot demonstrate an effect, this tells us whether the company employs good engineering or whether the company is operating in the realm of the "snake oil" salesman and/or pursues psychotic wish fulfillment.

... Musings on The Adequacy of Things

As for adequacy, remember that both subjectivists and objectivists can go way overboard! I think it's more difficult for pure subjectivists to say when they've explored deep enough and the system is at a "good enough" point because human desires are hard to contain. Given available resources, who doesn't want a faster car, bigger house, even heavier and overdesigned amplifier :-)? I think this is why pure subjectivists are often criticized because regardless of what is said to be experienced in the mind, there is still one objective "measurement" we can all judge by real-world standards - the price tag! As I have spoken of before, within the human ego, there are things we all want whether to impress ourselves or others... These are the "non-utilitarian" benefits of luxury products. Don't forget - even if one truly has been blessed with "golden ears", you're still human and psychologically affected by the countless multisensory and social appetites.

On the other extreme, I also do not advocate being "extreme objectivists" - or "measureabators" as some might colloquially call this. All things being equal, when feeding power into reasonably sensitive speakers in normal-sized listening rooms with reasonably low ambient noise, to claim that an amplifier that achieves 0.005% THD+N (SINAD 86dB) into say 1W across the audible spectrum is somehow very much less desirable than one achieving 0.0005% THD+N (SINAD 106dB) would be rather extreme I think. Of course there is correlation between sound quality and objective results up to some point, but after which, we're just chasing even more "zeros". For example, I can tell you with no uncertainty that this $500/W amp doesn't sound better than my less expensive Emotiva both based on measurements and listening. But objective measurements ultimately will bring us to the point of diminished returns. As technology gets better, the point whereby "more than good enough" becomes easier to reach at less expense. Obviously this truth will not be appreciated by purveyors of luxury goods who insist that their products are priced based on the merits of "sound quality". Remember folks, eventually we all will run into the limits of human hearing whether we like it or not. One might even be surprised how quickly or easily his/her limits are hit. Some will not want to be reminded of this which I think is a reason why certain individuals feel compelled to speak against controlled listening and blind tests.

... In summary

First, find your balance when it comes to how you want to evaluate gear. I've found great satisfaction in embracing both objectivism and subjectivism. But it is the objective side which holds the most truth to me and I think is of more value to the readership than my idiosyncratic preferences when I'm writing this blog. One should know the lingo and values of both emotion and intellect. I think I've spoken enough about this over the years.

Then know what is "good enough". This is where I believe the understanding of objective principles clearly provides practical insights. When we can appreciate the levels at which the noise floor is low enough, distortions are minimal, frequency extension is beyond the needs of our sound rooms and our hearing, etc... and then listening for ourselves to good performing devices to figure out what we subjectively enjoy, we can then perhaps start casting off some of that "itch" we feel a need to scratch. Not only will we know about the devices we might lust after, but even better, we get to know ourselves in the process. We might even get to a point where we appreciate when transparency is achieved for our own needs; as we approach this point, we can credibly relieve ourselves of unnecessary neurotic baggage. In doing so we can also develop a healthy skepticism about questionable claims and certainly keep ourselves firmly away from antisocial "snake oil" peddlers and the psychotic beliefs of some.

In the big picture, I think it's worth asking if some in the Industry might not want audiophiles as a group to find a place of balance, mindfulness, and serenity.

Remember that I'm talking about higher level abstractions. Each audiophile is free to choose his/her philosophical leanings. Over the years, my personal audiophilia nervosa affliction has changed significantly, I think I'm "cured" for the most part :-). Some might never be fully "cured" of this affliction, nor want to be. This is fine of course! Even if intellectually one recognizes that there's no need to spend more on a DAC, or seek yet more powerful amps with lower distortions, that doesn't mean one is not free to still buy "better", "bigger", "sweeter" sounding designs. I certainly recognize that as humans we are free to be irrational.

Just be insightful enough to laugh about our own insecurities and neuroses once awhile. Especially as one is about to enter the gladiatorial ring of combat in the on-line forums :-).

---------------------------------

To end off these "summer musings", I came across this post on Audiophile Style - "How much difference does it make?". The topic is certainly relevant considering the discussions above. I was particularly intrigued by this response by PeterSt with his perspective on this question and how he splits up the relative contributions to sound quality for each component listed based on a percentage.

For "fun", let me try this as an exercise as well. Remember, this is for fun! I have the right to change my mind on this at any time. ;-)

First let's talk about context when doing things like this. Context is everything.

Obviously any single poorly performing component would be a major setback for the overall sound quality! A system is only as good as its weakest component and there could be synergistic benefits where components may complement each other although ideally it would be nice if each piece simply does the job without depending on something else to maintain neutrality. The numbers below of course will assume that the devices "work". Secondly, let's assume that we're playing good recordings. After all, what's the point of using multi-thousand dollar equipment with poor source recordings barely having any dynamic range, limited frequency spectrum, and is of low resolution?

The "%" I'm listing is referring to the amount of sonic variance each class of component contributes as it affects the final hypothetical "100% sound" we hear with the intent being technically accurate, transparent, high-fidelity audio reproduction of the source in the listening room. Therefore, one would expect that there would be greater value in taking the time to research when choosing higher percentage components. These components may also need to be tweaked (eg. speaker placements, room treatments). Also, within the same price range, there's likely greater variation in sound quality than similarly priced low-percentage components.

I'll use similar categories as PeterSt but will include speakers for obvious reason! For simplicity, let's just focus on a modern digital system. Here we go...

Room: 40% - major importance! This obviously implies that judicious room design, furniture, and room treatments can provide big returns on investment.

Speakers: 40% - major importance! Full range vs. small speakers. Sub? Placement? If you're a headphone listener, those should be of major priority along with the amp since you don't have to worry about room contributions.

The link between room and speakers is likely the most important factor when we talk about system synergism.

Integrated amp: 12% - gimme flat frequency response, decently low distortion, low noise, low crosstalk for whatever target power you need. Remember that most of the sound quality is in the "first watt" which is where I believe measurements should focus on while also making sure that the manufacturer is honest with peak output ratings. For component systems, we can break it down further to something like:
     Power-amp: 8%
     Pre-amp: 4%

Resolution-wise, it is more difficult to find power amplifiers that maintain high fidelity than pre-amps.

DAC: 5%. Gimme a device capable of >16-bit performance with low noise, low distortion and low jitter - quite easily achieved these days with standard DAC chips without needing to go to exotic designs. Remember the results of the blind test recently.

Source hardware (PC, streamer): 0.5% - With a decent DAC, whether a streamer is a PC or Raspberry Pi makes no real difference as demonstrated. I'll throw in 0.5% here just to remind folks to be reasonable in placement and to use fanless devices. Again, not tough to get at low prices these days given technological advancements in speed and lower power demands.

Give me a bitperfect and asynchronous digital interface like asynchronous USB to the DAC and of course ethernet. WiFi would be fine so long as it's fast enough and maintains a robust connection. While often not as ideal from a fidelity perspective, S/PDIF like coaxial and TosLink are fine though generally more jittery with most DACs while providing better backward compatibility and ease of hookup (no special handshaking, typically "it just works!"). It's more about function and features than "sound quality" for computers and digital streamers.

Source software (ie. playback software) - 0.5% - Again, give me bitperfect and use decent drivers. Free foobar with appropriate ASIO/WASAPI driver sound as good as anything else. Features like a good library function, tagging support, metadata benefits (like Roon), ease of use are all important and worth commercial product consideration. Obviously DSP processing can make a huge difference. Nothing else to say really because the vast majority of this isn't related to sound quality at all for modern devices.

Analogue cables - 1% - make sure speaker cables decent like 12AWG and short to keep resistance low. Make sure RCA interconnects shielded, not too long (like maybe 10' max), and not exposed to areas of strong interference. If possible, use balanced cabling for better common mode rejection (in most set-ups, the noise level between XLR and RCA cables are practically inaudible).

Digital cables - 0.1% - just make sure the damn thing works ie. no digital errors. Remember, digital errors sound obvious like this - not the nonsense some cable companies claim such as a change in "soundstage" or deviation in frequency response! I'll throw in a 0.1% here because one might find (likely inaudible) jitter at times depending on the cable chosen and digital interface selected.

Yes, bits are essentially just bits. Despite some claims to the contrary (see discussion on Ted Smith's claims here), there has been no demonstration that temporal qualities like jitter are audible with reasonable devices these days.

Add it all up and the pie looks like this:

Notice that there's 0.9% left over for "miscellaneous" things. You know... Maybe sometimes there is a need for better grounding or power conditioner which might improve sound quality and worth investing in especially in certain parts of the world.

Like I said, this breakdown is mainly for your (and my) amusement. However there is some meaning in this exercise because it suggests to us how we could prioritize the hardware we buy. Put more energy into optimizing the big percentage (and likely "big ticket") items and be wise about not spending excessively on the low-percentage components.

Up to now I have avoided talking about price and how this might correlate with those percentages. Yeah, I think there should be some correlation. For example, I think it's important to spend money on the room and speakers mainly, then the amp, then the DAC. A streamer like an inexpensive Raspberry Pi is totally fine but feel free to upgrade once you're reasonably content with the higher priority pieces.

BTW, a good example would be the battery-powered US$35 Yeeco amp I discussed last week. Remember I said my system with >$10,000 speakers in the sound room still sounded OK despite the inexpensive streamer, DAC, and amp. Now imagine if I had inverted the cost structure of that system where I'm now listening to say a $2000 streamer, $3000 DAC, and $6000 preamp/amp system but $200 small speakers in a squarish 10'x10'x8' room. Would that sound just as OK? I don't think so... And I think it would be obviously ridiculous for anyone to suggest electronics of any price would be able to compensate for such a limited speaker/room combination!

Notice I allocated only 1.1% of the pie to analogue and digital cables! [Blasphemy eh?!] Suppose for a moment we literally translated this percentage to the price of a system. Considering that my main sound system costs ~US$20,000 including speakers, preamp, amp, sub, the computer, storage, network, and software, 1.1% is still $220 which easily covers all the generic 12AWG 99.9% OFC speaker cables, banana plugs, double-shielded RCA, XLR, USB 2.0, ethernet Cat 6 cables I might need to hook everything up. Yeah, I know cable companies and retailers will suggest something like 10% on cables ($2000 for a $20,000 system) - of course they would given the little actual R&D necessary and the high profit margins on these things! Alas, I'm not neurotic enough to care how thick my cables are, the brand name or how pretty they look these days.

One last reminder. The listening room is important and obviously a good one costs good money! Given the importance of the room in contributing to sound quality, don't forget the potential benefits of digital room correction. Getting this done right will require some investment in time and expense which I consider as part of improving that 40%.

As this blog post goes to "press", I'll be somewhere in the Mediterranean between Gibraltar and Rome. Time for some R&R with the family and divert attention elsewhere. As such, I'll be away for a few weeks so I hope these "summer musings" might provide some food for thought over the season...

All the best. Hope you're all enjoying the music!