Saturday 21 November 2020

On "Measurements, Listening, and What Matters in Audio" by Robert E. Greene, with unfortunate Robert Harley "counterpoint".

It has been a busy week, so alas, I didn't get a chance to finish some recent measurements in time for the weekend. Will aim for next week!

The other day, I read Robert E. Greene's editorial on The Absolute Sound's webpage titled "Measurements, Listening, and What Matters in Audio". Nice, I didn't think I would ever read such coherent introspection in those (virtual) pages. Probably also one of the first times I have seen the mainstream audio press willing to consider the "beginning of audio wisdom" (hat tip to Proverbs). Of course, seeking wisdom plus achieving a more rational basis in this hobby are overarching themes in many of the articles on this blog. So too, Greene's reference to concentrating on the "fundamentals", the foundation of acoustics whether it be in the production chain (ie. "microphones") or the reproduction/perception system (ie. especially "speakers and their room interactions") are not unfamiliar to readers here - we covered some of this earlier this year. Absolutely agree, Dr. Greene, including the part about "I think that almost everything in audio can be explained by measurements" is a fair and in these days, a very safe statement to make.

Wishing you good health.

Now as for Robert Harley's "counterpoint". Oh my... Surely Harley's ramblings cannot be the last words on this topic because clearly there are issues!

Where do we begin? Let's go thru a few items...

1. Let's start with his conclusion: "Because music speaks to our humanity, a piece of test equipment, no matter how sophisticated, can never replace the experience of sitting down between a pair of loudspeakers". Isn't this obvious? When has anyone advocating objective measurements ever suggested that one should listen to music with "a piece of test equipment"? Yes, music is for human consumption until the day perhaps when some AI might find it enjoyable. What's the point of such as statement? No objective-leaning audiophile I know of would suggest that music doesn't incorporate an emotional, subjective component which gives it value!

2. Regarding two aftermarket power cables: "These differences in the shapes of the musical waveforms are far too small to see or measure with even the most sophisticated technology, yet we as listeners not only routinely discriminate such differences, we sometimes find musical meaning in these differences." Nonsense. Just because people claim to "routinely discriminate" differences doesn't mean it's true or they're right. Apparently many have witnessed UFOs but that doesn't mean they actually saw extraterrestrial visitors, does it? Some have seen/heard a deity speaking to them "routinely"; does that imply that they are surely communing with an unseen/unmeasurable spiritual force(s)? Can we not put a little more effort into confirmatory reality-testing first when "the most sophisticated technology" can find nothing in 2020? (Of course, speaker cables can measure differently as per herehere, even if not necessarily audible in many cases by the time we connect amp to speaker.)

3. Regarding jitter: "Yet reconstruct an analog waveform from digital samples with a clock whose timing precision varies by just a hundred picoseconds (0.0000000001 seconds, or one one-tenth of a billionth of a second, the time it takes light to travel about an inch) and we hear the change in the analog waveform’s shape as a reduction in spaciousness, a hardening of timbre, a “glassy” character on high-frequency transients, a softening of the bass, and an overall reduction in listener involvement."

Nonsense. Permit me to be blunt. What is it with typically old(er) audiophiles insisting that they can hear one-tenth-of-a-billionth-of-a-second time anomalies when they're not running out of the room screaming when confronted with thousandths-of-a-second abnormalities as one would find with turntable playback (see here and here, compare here)!? Where in all these decades of research has there been any reproducible evidence for such a fraction-of-a-billionth-of-a-second anomaly being detectable by human listeners, much less having meaningful effects on musical enjoyment?

As readers know, I encourage you to have a listen to simulated jitter for yourself and tell me if even 875ps is such a big deal (forget 100ps!). So, do you hear the gross "glassy" anomalies and bass softening? BTW, how in the world does picosecond jitter "soften" bass given the long wavelengths and the fact that it's still bit-perfect and amplitude levels remain unchanged?! How does one even propose that for woofers and subwoofers, the cone is even able to start/stop with billionth-of-a-second precision assuming this level of accuracy even matters after the crossover!

Seriously, how can anyone write this stuff? Unless there is some evidence, isn't this just shameless "fake news" repeated now over years with no basis in reality? Apart from doing what one can to debunk these types of beliefs, there's really no debate to be had because the argument lacks any merit.

4. Please, stop name dropping and bringing up irrelevant quotations as if this adds to your case. If Richard Heyser were alive today (he died in 1987), do you think he would agree that aftermarket cables can sound quite different while remaining unmeasurable "even (with) the most sophisticated technology" and that 100ps timing anomalies would be of serious concern to him. Would he feel honoured to be quoted and used as supportive evidence in an article that makes such claims? By the way, would many of the members of the Audio Engineering Society agree with such claims? (Here's an actual AES seminar on audiophile myths which is probably more reflective of the mindset of AES members.)

As for Heyser's quote, yes, of course the brain is very sophisticated and uses numerous pathways to decipher sensory information including speech/language decoding typically lateralized in the left hemisphere. We are indeed built to discriminate language from childhood (and anomalies in development like autistic spectrum disorder leads to issues with language acquisition), but it doesn't follow that this means "Humans seem to be hardwired to discriminate very small differences between similar things". There is some serious misunderstanding and lack of logical connection here; being hardwired for language is not the same as "very small differences"! These days, we understand much more about the neuroscience of language including the various pathways that result in phonological and semantic processing, the disruption of which would lead to various forms of aphasia. Furthermore, Heyser would not have live at a time when sophisticated machines are now routinely capable of natural language processing to listen to sounds and pick out words - "Hey Google!"

Another quote comes from Bob Stuart in the context of how "some of us are absolutely obsessed with tiny variations". Hasn't Stuart said enough questionable things over the last few years (see here) such that he might simply be wrong?

By definition, obsessive people perseverate over "small differences", so what? That doesn't mean the minutiae are actually meaningful (nor wise)! In fact, it could be pathological obsessionality resulting in nothing of value. For example, over the last decade, audiophiles have obsessed over DAC impulse responses yet it hasn't proven to be all that important (see here for an example), nor obsessing over "Hi-Res Audio" would necessarily have achieved better sound. I believe that for many products (like DACs, orthodox amplifiers, certainly wires), we are at a point of technological maturity where if you follow Greene's thesis to "concentrate on fundamentals", much of the little things have already been resolved by folks who know the science in a very competitive marketplace developing products over many generations.

5. "Music is different from other forms of communication in that the meaning and expression are embodied in the physical sound itself. The vibrating air molecules striking our eardrums are not a representation of the music, but the music itself."

Hang on a second. Surely he must not be saying that "music" (which in itself is a higher level mental construct apart from just "sound") can only be represented by an absolute type of "vibrating air molecules". Is the same piece of music played through a mono AM radio not just as potentially "musical" in many ways as the $1M high-end system, yet the qualities of those "vibrating air molecules" very different? Relative fidelity compared to the source signal would be very different between an AM radio and a good hi-fi however.

Then he says - "You might not hear a subtle dynamic inflection, miss a crucial rhythmic interplay, or be oblivious to the way tone colors combine that would otherwise create an ineffable flood of emotion. The sound contains the meaning; it is not a representation of the meaning that can be divorced from the physical phenomenon conveying it." So he's just saying that it's good to have high-fidelity reproduction so we can catch the subtleties in the music which might evoke emotions and at times make the music even more impactful, right?

Is this any different than saying that a high resolution 4K/1080P video is better than 480P in allowing us to catch facial expressions and subtle body language when watching a drama, perhaps evoking greater emotional investment in the characters and stories? Do videophiles go through such tortuous use of language for presumably a simple concept?

Bottom line, dear audiophiles...

Measurement equipment allows us to determine the accuracy of audio reproduction and if a product is living up to claims of high-fidelity. We've been through this before and there's nothing new Harley is adding as far as I can tell beyond loose associations and other irrelevant/tangential content. I believe his thinking lacks clarity and he is confounding different ideas at times - for example:

"All these observations point to the fallacy that technical measurement can replace the discrimination ability and auditory-processing power of our ear/brain system." (emphasis mine)

"Technical measurement(s)" can already easily detect anomalies at resolutions beyond human hearing as shown time and again, so indeed, test equipment has high "discrimination ability" between different sounds.

"Auditory-processing power" however - which I interpret includes complex cognitive stuff like how a listener attends to the music, finds meanings, processes lyrics, evokes emotions - is not what testing hardware is used for. Simply put, measurements evaluating hardware fidelity/accuracy is different from "music appreciation". To put these things together in a single sentence and suggest measurements can "replace" such disparate functions of the "ear/brain" is simply ridiculous! By not being careful, Harley creates an argument that doesn't even apply to anyone's belief system.

As audiophiles, practically, objective measurements allow us to determine 3 important things IMO:

1. It tells us if something even has an effect or not. This helps us be mindful of snake oilCable measurements might be a good example where digital cables can be deemed to be bit-perfect and certain "audible" claims about them must be considered questionable.

2. It helps us confirm or disconfirm whether "differences" should actually be considered "beneficial" for fidelity. MQA might be such an example - sure, it could sound different but please don't use terms like "lossless" and "de-blurring" because it doesn't qualify nor achieves the implied goals.

3. It gives us guidance on how best to use the gear and whether there are gross anomalies/pitfalls. Speaker measurements for example can show us frequency response limitations and give guidance on directivity. Amplifier measurements can verify power limits and distortion levels before we buy. A DAC's measurements can demonstrate if it's truly capable of >16-bit playback. Measurements with different digital frontends confirm bit-perfect capabilities and whether jitter is a problem with a certain DAC. Bugs in software (remember JPlay?) can be demonstrated. Etc...

Knowing the above simply allows us to be educated consumers. We can then go listen for ourselves to create a final impression including personal preferences, recognizing that test gear was never meant to be a stand-in for human emotions/experience!

Robert E. Greene's editorial is good and IMO should have been left to stand on its own so that readers can take time to appreciate this thoughtful introspective essay and discuss. Mr. Harley's long apologetic intro was absolutely unnecessary and his "counterpoint" is an unfortunate intrusion that in my opinion was poorly conceptualized.

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

Happy Thanksgiving to the American friends this coming week!

I suspect and hope there won't be much Black Friday door busting this year (perhaps we can see this as a blessing). Here in Canada, Black Friday wasn't really a thing until about maybe 10-15 years ago. I continue to check out deals online for small speakers on Amazon and I see that the walnut KEF Q150 is currently about 40% off from last week, Elac Debut 2.0 B6.2 about 25% off and the well regarded NHT SuperOne 2.1 is about 25% off also - all very tempting :-). Presumably similar kinds of sales in other countries.

Hope you're staying well and enjoying the music.

16 comments:

  1. right on archimago - it is so refreshing to hear your counterpoints - its nice to have someone have their bs meter working at full strength

    ReplyDelete
    Replies
    1. Thanks for the note Stewart,
      It truly is unbelievable the amount of BS out there in the audiophile world. Quite sad I think for the hobby, especially to newcomers who pick up the magazine genuinely interested in hi-fi and being fed stuff that I believe ultimately leads to disappointment more often than not...

      Perhaps eventually the audiophile comes to his senses and recognizes that much of what is spoken at places like TAS was always attempts at selling fantasy. In some circumstances, this might be after blowing a few thousand bucks on USB cables, bizarre tweaks, and the like. :-(

      One can see in the structure of that article just how defensive Harley was in literally "framing" Greene's words... Sandwiched between his own intro and "counterpoint" in order to somehow not burst the fantasy bubble for a certain type of readership!

      For the record, I don't think Harley is an unintelligent or delusional man reading some of his early material years ago but yet seems to have to play one as the TAS editor. Reading his prose, I see that he is choosing specific examples, conveniently leaving out details, and selecting words carefully. I don't actually believe what he says at face value and IMO, he knows what he's saying is highly questionable and indefensible, but must maintain the façade for a certain type of "culture" and for the Industry the magazine caters to.

      Delete
    2. about 1996 or so mr harley wrote this incredible review of the so called "digital lens" https://www.stereophile.com/digitalprocessors/824/index.html - and of course based on that review - i was so hot to purchase one for $1,800 - it turns out i was so dissapointed in the piece - luckily i was able to sell it 3 weeks later.

      Paul Mcgowan was the owner of genesis technologies - the "lens" left such a bad taste in my mouth that i never bought a mcgowan product again - and the harley review left me wondering if i could ever trust his reviews again - he never heard a piece of gear that he didnt slober over. to this day i have some doubt about his glorifications of products he reviews

      Delete
    3. Oh yes Stewart,
      The good old Digital Lens by Paul McGowan. Notice that back in those days Harley was doing the measurements as well but this has disappears in the ensuing years while our ability to access ways of measuring, and share over the Internet have improved.

      This is why I have doubts about Harley's perspective as noted above. I believe he can do better than this and IMO can be more honest because it simply doesn't make sense that the man can be capable of a more technical and objective perspective and just abandons this for the sake of hand-waving extreme subjectivism unless it really is to side with companies rather than being honest with consumers about the facts.

      Delete
  2. You can read Dr. David Rich's review of Bob Harley's audio book on page 82 at:
    https://www.pearl-hifi.com/06_Lit_Archive/02_PEARL_Arch/Vol_16/Sec_53/Audio_Critic/1997_Audio_Critic_No_24.pdf

    Peter regularly quoted some 'harley howlers' in most issues of The Audio Critic.

    ReplyDelete
    Replies
    1. Brilliant... Hmmm. This really links in with something I read this week ;-). Might need to be part of the next commentary! Thanks gnickers.

      Again as for Harley, it's a shame that he appears unable to change over the years given all the feedback. Maybe this was just meant to be his niche in life.

      Delete
  3. As an electronic tech (now retired !), I have been using quite fancy measuring equipment in my career: I have no doubt that today's measuring equipment is to say the less, as sensitive as the human ears and I would rather say the opposite, i.e. it can measure precisely things we can't hear, such as low level of distortion, low phase shift, etc.

    It is also very clear to me that if we can perceive a change in what we are hearing, it means that the audio spectrum has changed accordingly, being the amplitude of some the various signals, the harmonics content or the phase: all this can be measured.

    I may be oversimplifying the 'two sides' audio community, but to me, in this old debate, I would say that most of the time, "subjectivists" are those who don't have serious technical knowledges and for which science behind audio reproduction is just too complicated for them to understand and therefore, they trusts only what they know about, their personal believing or the one of their preferred audio magazine reviewers. "Objectivists" are those who learned and understands the science behind audio reproduction and trust that accurate measurements reveals what's going on with the signal the equipment under test is processing. And when you have those electronics and physics knowledge well 'soldered" in your mind, it is absolutely impossible to figure out, as an example, how a 2 meters passive AC supply wire can improve the 'air' between instruments and provide a much a deeper soundstage....

    ReplyDelete
    Replies
    1. Indeed, an old debate but perennial "favourite", DColby in audiophilia :-).

      I tend to think that while there are 2 extremes, there is actually a "centrist" side which I hope to put myself in that category even though I readily admit that I'm "more objective" mainly because the audiophile world has become so biased towards the subjective corner all these years - at least when I started writing in 2013.

      IMO, over the last few years, I believe there has been a gradual but perceptible shift. At least I'm seeing more pushback of questionable claims and products which I think is simply a healthy sign that the consumers "get it" and are empowered to speak out. Audiophiles are not as air-headed or devoid of some understanding about the technology as the magazine writers think.

      This is perhaps "difficult" for some magazines and certain subjective-only reviewers as their words are examined and dissected. I see nothing wrong with this and it's about time that audiophiles become more rational and clear up a number of falsehoods peddled by certain "high priests" and their false idols backed by nothing but mere words and opinions.

      Delete
  4. I would agree with this to a degree, but there is a caveat. I don't have an electronics or physics background, or indeed an in-depth knowledge of these subjects. I do, however, have a knowledge of how science works and therefore, having read much objective/subjective discussions over time, I choose to believe those with a scientific/technical background as to the efficacy (or otherwise) of audio engineering. I am aware though that this could still be construed as an article of faith as I do not understand everything that is written. I trust the objective measurements, allied with the blind tests that Archimago and others make, as they are open and reproducible, such that anyone with the requisite knowledge and equipment could disprove them if they so choose.

    I've built a few amplifiers, headphone amplifiers and preamplifiers (discrete and IC based) and truth be told, I cannot usually hear any difference between them when played through the same speakers using the same source (and this includes linear power supplies and SMPS).

    I have no doubt at all that some people have very discerning hearing and can hear things I cannot (53 years old, some tinnitus), but I am yet to be convinced that what they can hear cannot be measured or indeed if they are actually hearing differences.



    ReplyDelete
    Replies
    1. Absolutely Neal,
      I believe it is wise to take that middle ground and recognize that indeed just like any trait, there will be a "normal" distribution of abilities. Some will hear better than others for sure. And within a population, there will indeed be "Golden Ears" with especially good acuity for the softest sounds, pitch precision, fantastic discriminations of time and rhythm.

      The question of course is whether this specific person, that is, someone like Mr. Harley is capable of discerning what he described. Are "we" truly about to hear 0.0000000001 second anomalies? Who knows...

      The problem is that he wants us to believe that the above is true and makes it seem like a substantial number of audiophiles are capable of such a feat; that it's somehow a known fact that jitter is audible, leads to glassiness, changes soundstage, and reduces bass. With no evidence of such a thing, this is simply nonsensical! That's hogwash of course but some will have faith in such a thing.

      I believe this is all just selling a fantasy. It's a "just so" attribution theory for claimed subjective perception that in itself might be erroneous - honest mistake or otherwise. And since people like this refuse to accept usual scientific methodology, will also never subject themselves to blind testing and will take opportunities to simply complain rather than doing any work themselves to prove their point - like this from Harley:
      http://www.enjoythemusic.com/tas/183/editorial.htm

      The fact that you've told us your age of 53 and humble enough to admit that there is tinnitus thrown in the mix plus the fact that you have built your own gear IMO already makes your opinion likely more worthwhile than Harley's claims...

      BTW, how old is the man?

      Delete
  5. I would add another point for measurements:
    4. Render questionable the pronouncements of the golden-eared priesthood of audio reviewers, whose subjective evaluations rarely actually agree with one another. (They do generally recognize one objective criterion, however, in that they usually find a correlation between the price of an object and its value.)
    Meanwhile, Stereophile posted another desperate defense of subjectivism today: https://www.stereophile.com/content/somethings-coming. They are fighting for their lives here. I can't help but think of recent conflicts south of the border between evidence-based professionals (and jurists) and those shrilly insisting on the existence of things that can't be proven.
    Just a thought.
    Best as always
    Phi

    ReplyDelete
    Replies
    1. Bingo Phil,
      Truth? Evidence? Objectivity? Reliable sources? Fake vs. fact? Critical thinking? Educated, honest, noble, prosocial citizenry?

      Clearly talking about this stuff in the microcosm of the audiophile world is relatively harmless; I don't expect the "audiophile mafia" to do any "hits" on critical objectivists. :-|

      In the broader world, the repercussions of course could be grave when biases and imbalances go on for much too long...

      Delete
    2. ...and Mr. Reichert @Stereophile is a rambling idiot IMO...

      Delete
  6. Archimago,

    Thanks for the reality check.

    Here is Benchmark's stance on the relationship between listening and measuring:

    https://benchmarkmedia.com/blogs/application_notes/listening-vs-measurements

    It is the latest in my series of application notes. It represents my stance on this issue as well as the stance of Benchmark Media Systems, Inc. We manufacturer hi-fi and pro-audio equipment.

    John Siau
    VP, Director of Engineering
    Benchmark Media Systems, Inc.
    https://benchmarkmedia.com

    ReplyDelete
    Replies
    1. Thank you John for dropping a note.

      Fantastic work on the Benchmark gear and I've appreciated your no-nonsense perspective over the years in the audio press and also in Mark Waldrep's articles. Love the line:

      "An APx555 test set has far more resolution than human hearing, but it has no intelligence."

      Well said!

      Delete
  7. Just purchased, as a Christmas present to myself, a new tonearm to replace my former arm which have given up for good. The new purchase is a fabulous arm, a Audiomod Series VI, and after having some distortion issues, I have finally made it sing, or rather, made the other equipment sing.

    What a relief after about a year with no possibility to play my beloved records, yay!

    And it sounds awesome too. I am aware that records should be inferior quality-wise to digital files, but there is some inexplicable joy and pleasure with the playing of LP's.

    Man, what a lovely Xmas present I just gave myself :)

    Sorry if this is the wrong place for this, but I am so happy...

    Merry Christmas to you all

    ReplyDelete