Article: Realism vs Accuracy For Audiophiles | Part 2: The Real Sounds Of Live Music

[bluesman]

View full article

[bluesman]

I hope it's helpful to our little community!

 

There are two links that are a bit wonky on an iPad but work fine on my computers, and I can't fingure out why.  So if the music files in the last few paragraphs don't open properly, here are the URLs:

 

• Cannonball Adderly on Autumn Leaves (classic big, full alto tone)
  • https://youtu.be/rsz6TE6t7-A
• Paul Desmond on Autumn Leaves (the sweetest alto sax tone in history)
  • https://youtu.be/Gsz3mrnIBd0
• Art Pepper on Autumn Leaves (edgier and less round than Cannonball's)
  • https://youtu.be/Kwlcgkhuxv4

 

 

• fretted Fender Precision Bass (the timeless classic Fender bass model heard on countless 1000s of hits in all genres)
  • https://youtu.be/YBqmnkVh-nY
• fretted Fender Jazz Bass (the "other" Fender bass, introduced in 1960 and also heard on countlesss 1000s of recordings)
  • https://youtu.be/YBqmnkVh-nY?t=134
• fretless electric bass (a similar bass but with a smooth fingerboard like that on  a cello or double bass)
  • https://youtu.be/HPAHQXms6_A?t=58

 

 

[bluesman]

29 minutes ago, CG said:

Anyway, I got thinking about video.  I wonder what the percentage is of people who have the color and contrast controls on their TV sets turned to 11 so that everything is more everything than it could possibly be?  I bet it's very, very high.  This is with video, where our memories are better and you can actually do real time side-by-side comparisons, too.

You may not have noticed, but a growing number of TV shows are shot with HDR - and the degree of processing seems to be increasing every season.  No color seen on the new Magnum or Hawaii 5-0 exists in nature, including the red of the Ferrari.  Hawaii's definitely green, and its flowers are quite colorful - but it looks black and white compared to those processed TV images.

 

I'm working now on #3 in this series.  It's a discussion of how recordings are engineered for specific effects.  I'm making a series of demos showing how simple mic placement changes what you think you're hearing, using my instruments (Yamaha grand and multiple guitars - acoustic, electric, archtop, flattop, resonator etc).  And I'm assembling a bunch of stuff you can do yourself with a simple audio editor like Audacity, to change image width, instrumental positioning in 3 dimensions, apparent instrumental size etc.  Once you discover how easily and extensively even a simple, single instrument recording can be changed, and you accept that it's been done to most of the recordings you own, you lose some of that razor focus on the emperor's new sonic image.  This frees you to enjoy and to listen more intensively to the music.

 

Once you start doing that, you learn about the music you love at an amazing rate.  And after learning how your system treats the many musical elements in your favorite kinds of music, you can hear the evolution of the players and their instruments in any genre.  For example, if you like the tenor saxophone in jazz, you need to hear and appreciate how tone evolved from the swing era to today.  Ben Webster had a tone that was unlike any other tenor player.  It was big and full, with a vibrato so intense that it intermodulated deeply and richly with what he was playing - he played like Domingo and Pavarotti sang.  And it ran from sweet as sugar on ballads to growling and biting on up tempo tunes.  Follow the tone train from Webster through Coleman Hawkins and Dexter Gordon to Stan Getz and John Coltrane.  Both Webster and Gordon had huge tone, but Webster usually growled and Gordon generally purred.

 

This is especially true for those who stop tweaking and changing components every time they read or her about something "better".  I strongly recommend having at least one system that you do not change at all for a year at a time, for a consistent perspective on your listening.

[bluesman]

1 hour ago, CG said:

I prefer to watch older stuff from before I was born

If I did that, I couldn't watch TV 🤪

 

50 minutes ago, CG said:

I pretty much build all my own electronics - DAC excepted - so I can take a more incremental approach to system evolution.

I've built a lot of mine too.  That's one reason why I suggest having at least one decent system for long term listening.  Once you learn how your music sounds on it (especially if you have a chance to hear at least some of your favorite perfomers live from time to time), you have a steady platform for enjoying and learning more about your music.  Then you can build away to your heart's content, using your reference system as a true baseline as well as for listening pleasure.  So you can compare what you're building to that reference, to see if it equals or betters your daily driver.

 

"Reference" in this case doesn't mean state of the art - it just means a stable baseline with SQ acceptable to you.  Most of my daily listening is from ROCK on a NUC, rendered through a Raspberry Pi driving an iFi Nano DSD into a pair of JBL 305s supplemented by a Yamaha powered sub.  I recently set up a MC system in the same room by adding more 305s and driving them with the 8 channel DAC / renderer I built from a Raspberry Pi and described in this prior article.  

[bluesman]

31 minutes ago, CG said:

 

I was specifically referring to movies, but I feel your pain...

Pain? No pain here!  To be honest, I’d rather be me and over 70 than anyone else and under 50.  I’m lovin’ the ride and I’m staying on it until the end of the line.  The older you get, the fewer knobs you need 😎

[bluesman]

44 minutes ago, Spanna said:

“If the phase relationships of the fundamental and harmonics are altered, everything from perceived pitch to timbre can change.”
 

“Along with their phase relationships, variations in the strength and frequency of harmonics can affect the perceived fundamental pitch.”

 

Would you agree that we could use the converse of this and listen to the perceived pitch to give us an indication of how accurately the fundamental and its harmonics relate to one another, with the most pitch accurate components also being the most musically accurate components?

The problem with that postulate is that perfect pitch and intonation accuracy are rare.  So it’s highly unlikely that any two musicians playing “the same note” will actually be playing the exact same note.  Open strings on tuned instruments like pianos will vary by a few cents with time, temperature, humidity etc.  Pitch accuracy in wind instruments is dependent on the player’s embouchure, breathing, etc.  The pitch of bowed stringed instruments is set by the player’s finger with no frets to guide accuracy.  And the pitch of notes on fretted instruments is altered within a few cents by finger pressure on the strings (which are suspended between the raised frets). So how would you know if a pitch inaccuracy of a few cents is caused by harmonic inaccuracy or because it’s being played that way?

We don’t hear random pitch variations in music if the instruments are well enough tuned, because (with almost no exceptions) humans who are not trained can’t discern pitch differences of 5 cents (which is how musicologists refer to percent of pitch shifts) or less [D.B. Loeffler, "Instrument Timbres and Pitch Estimation in Polyphonic Music Archived 2007-12-18 at the Wayback Machine.". Department of Electrical and Computer Engineering, Georgia Tech. April (2006)].  So music sounds fine despite a cent or two of pitch variation among instruments.

 

Remember that very slight differences in fundamental frequency between instruments playing (or trying to play) the exact same note will result in "beat freqeuencies" from natural intermodulation.  If 4 violins are all trying to play a middle C (256 Hz at concert pitch), one may be playing 255, one 256, one 257, and one 258 - and all will sound in tune to us.  But there will be intermodulation among all the harmonics of all those notes, which becomes part of the rich harmonic content that gives a violin section its sound.  I suspect that the best ones (e.g. the Philadephia Orchestra) are closer to perfection in both tuning of the instruments and intonation of the fingered notes, and that the reinforcement of even order natural harmonics is part of the reason such sections sound so lush and smooth.  
 

The bottom line is that even though though we don’t recognize 2 or 3 cents of variation in the fundamental pitch of a note,  we do perceive shifts in the phasing of harmonics as slight variations in the timbre / character / etc of an instrument even though it doesn’t sound out of tune.  And the reason for this is that (as I described in the article) we only hear the fundamental pitch of multiple simultaneous tones that are within the harmonic structure of that frequency.

[bluesman]

48 minutes ago, Spanna said:

You appear to be talking about musicians being out of tune with each other, whereas I’m wondering if listening to the effect HiFi components have on pitch reproduction can be used to determine which component is more musically accurate.

That’s not what I’m saying.  Instruments are ideally tuned to the same standard, which in much of the world today is that A above middle C is 440 Hz.  Being off by a few Hz is both inevitable and inaudible for most.  I haven’t seen or conducted a randomized measurement trial, but I suspect that the distribution around A 440 for all the instruments in an orchestra or band is probably a very tight Gaussian function with a mu of 0 +/- 1 Hz and a very low sigma squared.  The 5th and 95th %iles are probably somewhere around 437 and 443.

 

Not having any way to know where a given instrument is tuned within the “in tune” window, you couldn’t assess the accuracy of reproduction because you wouldn’t know if the minor deviation was in the instrument, the system, or a combination.

 

There might be some merit to using an electronically generated or sampled tone with known harmonics corrected to an exact fundamental pitch.  I’ve never heard of this as a test, but it’d be interesting to compare FFTs on it at the line level input of an amplifier with microphonic capture from the speakers.  Of course, this would require perfect signal chains (the proverbial straight wires with gain) in both the reproduction and output capture systems.

 

But knowing that Rachmaninoff’s prelude is in C# minor won’t assure you that every C# minor played in the performance was at exactly the same frequency, and most probably aren’t. So you couldn’t use recorded music as such a test. 

[bluesman]

2 hours ago, hopkins said:

There's no difficulty, however, in identifying Ben Webster even coming out of a cheap bluetooth portable speaker.

There’s no difficulty identifying him even when playing a cheap student model sax, although his tone might not be quite as rich and enjoyable as it was from the Selmer Balanced Action tenor he favored.  Then again, Charlie Parker played a plastic saxophone for the last year+ of his short 34 years and no one complained about it in his hands.  Just before a 1953 concert at a Toronto venue called Massey Hall, he’d pawned whatever good horn he had (allegedly for drug money, and not an isolated occurrence - he even pawned instruments he’d borrowed).  So he had a concert to give but but no sax to play.  And this was a serious gig, with Dizzy, Bud Powell, Max Roach and Charles Mingus rounding out the quintet.
 

The rep for the Grafton sax company heard about it and lent him one - but Graftons were made of injection molded acrylic. You can hear how he sounds on the recording of that date (called Jazz at Massey Hall and readily available both as captured and in much improved remasters).  It’s far from a good recording, but you can easily tell that it’s Bird.  And he got a pretty fine tone from that Grafton, even though many others who tried it thought it was harsh and strident.  In fact, most pros hated it.  Ornette Coleman even used it as a “special effect”.

 

Bird was playing a Grafton on the wonderful live recording of the 1953 “One Night in Washington” concert with a fine big band (although none of the bebop legends was in it).  I only have the vinyl, but there are digital files of this beautiful and historic night of great jazz, and every jazz lover should own it.  Parker rarely played with a big band. Yet he showed up, hit the stand, and nailed every tune without rehearsals or printed music.  The recording is surprisingly well done, especially as it was made not in a concert hall or studio but in a club.

I’m pretty sure he started playing Graftons regularly, and I know they made his to his spec.  The key action on a Grafton used coil springs rather than the wire torsion springs used on standard saxes and clarinets.  This made the action so slow and spongy that it limited speed and accuracy of playing. Parker was so technically facile for much of his career that he was apparently not affected by these mechanical issues, even though most others were (no matter how well they played). 

 

This is exactly the kind of thinking and listening that I’ve been encouraging.  We’re discussing how different players and instruments can sound from each other, and what that adds to music.  Ben Webster’s Selmer was part of his sound, just as Stan Getz’s Selmer was part of his. No one who knows how they sounded would confuse the two.  But I hope you can see how a preference for one sound over the other could easily bias an innocent but unknowing audiophile toward systems that add the preferred flavor.

 

For me, seasoning is best reserved for food.

[bluesman]

23 minutes ago, hopkins said:

Thanks for the recommendation - will listen to that concert ("One Night..."). Charlie Parker grew up in the "big band" era (not a term I like), played with several (including a year in Chicago with Earl Hines), so not surprising he fit in well.

Parker had a less than stellar early career.  Documentation is a bit inconsistent, but I think he first joined Jay McShann’s band in 1938.  He left for New York in ‘39, but worked several jobs (including dishwasher, which gives him something in common with Little Richard) because he was unable to survive as a musician.  So he returned to KC and McShann in ‘40 when his father died, playing for McShann until moving to the Hines band in ‘42. He then joined Billy Eckstine’s new band for some months before moving on to a life in bebop.

[bluesman]

1 hour ago, hopkins said:

Thanks for the recommendation - will listen to that concert ("One Night..."). Charlie Parker grew up in the "big band" era (not a term I like), played with several (including a year in Chicago with Earl Hines), so not surprising he fit in well.

Unfortunately, there was an AFM ban on recording in 1942-44, so we don’t have much to document and enjoy what was a very important period in jazz.  The Hines band in 1942 was really the first bebop band.  With Bird, Dizzy, and Sarah Vaughan, it must have been wonderful.  Then Parker joined Billy Eckstine’s new big band early in 1944 before moving on to being a star and playing in small groups.

 

Most people don’t realize that Parker played tenor sax too.  The story is that Hines needed a tenor but Parker only had an alto. So Hines bought him a tenor.  He recorded on tenor with Miles on Davis’s first Savoy sessions under the name Charlie Chan (Chan was his wife’s first name). There are some YouTube videos of early Miles with Bird on tenor.

 

This is yet another great example of the importance of accuracy in reproduction.  My bet is that many audiophile jazz fans have heard recordings of Bird on tenor and didn’t hear or notice a difference.  If you can’t tell a tenor from an alto on a recording good enough to differentiate between the two (which is almost all commercial jazz recordings), there’s something missing somewhere between the source file and your higher cortical centers 😗

[bluesman]

44 minutes ago, hopkins said:

In that particular case, I would say the system is certainly not at play - I would assume that most people who don't take note of the difference may simply not be interested in that aspect

True for some.  But the note ranges of tenor and alto sax overlap by an octave and a half.  The alto’s range is from concert Db 3 (ie the one below middle C) up 2 1/2 octaves to Ab 5.  The tenor’s range is from concert Ab 2 up 2 1/2 octaves to E 5.  And most playing is within that overlap.  So even a slight midrange emphasis could make a thinner tenor sound more like a fat alto.  Throw in the many variants among mouthpieces, reeds, embouchures, and playing styles and you can find yourself loving an auditory mosaic that simply doesn’t sound like the player you think you’re hearing, if your system adds coloration similar to that of a thicker reed or a smaller instrument.

 

This level of distinction is far more obvious than the effects of many system mods and substitutions discussed on AS and elsewhere, and used by many audiophiles to “improve” the SQ of their systems.  If differences of this magnitude don’t matter or are inaudible to someone, I don’t understand how he or she could possibly hear or justify the effort and cost of “audiophile grade” stuff.

 

If there are audiophiles who don’t care enough about the sounds of music and of the instruments and musicians playing it to want to be able to enjoy the distinctly beautiful nature of whatever music they like (eg the very different sounds of Paul Desmond and Charlie Parker on the “same” song and instrument or the vastly different sounds of baroque and modern orchestras and instruments playing the same music), about what do they care?

[bluesman]

3 minutes ago, CG said:

 

Based on the relative small percentage of loudspeakers available on the market that approach that, it may be that many people either aren't sensitive to this, or just don't care.

These effects are not huge, and because they’re randomly distributed among all the components in a system (even drivers, crossovers, cabinets, baffles etc), some probably mitigate or counter the effects of others.  We all agree that even the best systems don’t have SQ indistinguishable from each other and that all fall short of total accuracy and perfect realism.  The effects under discussion are certainly one reason for this.  And the costly perpetual search for better SQ is ample evidence to me that people do care.

 

I just hope my input helps them understand a bit more, listen a bit more carefully, and learn to better distinguish among and appreciate the many potential sources of sonic differences.  Sommeliers, baristas, and security experts are trained to identify thousands of important (and sometimes blatant) factors that the rest of us don’t even know enough to look for, let alone care about.  But once educated and experienced, we learn.  Audiophiles are not immune to this effect.

[bluesman]

1 hour ago, CG said:

To some degree, it's a lot like buying and trading expensive watches.  Most are pretty close to getting the time right, so the desire for more or different comes from someplace else.

I agre with that.  But watch aficianados collect watches for reasons that audiophiles deny considering, like appearance, prestige, and bragging rights.  As you correctly observe, keeping time is not the reason people buy Pateks, Ulysse Nardins, et al.  If you just want a timepiece, you can get equal or better accuracy for a lot less money from an inexpensive Swatch or G-Shock. And as the old saying goes, a man with one watch usually knows the time.  But a man with many watches rarely knows the time.  People buy multiple Pateks, Langes, VCs etc because of their artistry, their beauty, their rarity, their alleged investment value (which is usually no more than an excuse given to one's spouse and/or one's self), etc.  But many high end watch owners have them only because the watch makes them feel successful (or at least in their minds, makes others think they are). The basic raison d'être for all watches is (allegedly) to display the time - so they all fulfill their "main" function.

 

I'm assuming that the audio system's raison d'être is reproducing music well.  Most AS participants have systems that do a fairly good job of this, and no two of them agree on the "best" system or component.  So following your logic (with which I agree), audiophiles who can't live with the same system for very long are probably seeking something other than better SQ.  As you suggest, "different" is probably a common goal.  But it's an elusive one that brings only temporary satiation.  And audiophiles who seek change for change's sake may not be the best sources of critical analysis of recordings and equipment.

 

When we retired and downsized, we gave our kids their choices from our watch collections (which included our parents' and grandparents' watches plus a few dozen of our own acquisitions) and sold most of the rest (keeping only our few favorites).  Like most true watch collectors, we love to look at them as art, wear them as jewelry, and enjoy them as mechanical marvels.  Most of my good watches were almost as accurate as my Apple Watch, which is definitely more accurate as a timekeeper than my audio systems are at reproducing music.

[bluesman]

50 minutes ago, Spanna said:

we could use the converse of this and listen to the perceived pitch of notes (as generated by various HiFi components) to give us an indication of how musically accurate each component is? That is, the more tuneful a component appears to reproduce music, the more the fundamental and its harmonics relate to one another (harmonically) and therefore, the more linear and accurate the component is.

I just don’t think that’s possible.  Our perception of pitch is not perfect, and I don’t know how to define, qualify, or quantify tunefulness.  And any component in the entire signal chain from mics to master could also alter harmonic structure.  In addition, you may judge what you hear to be tuneful even though it’s not an accurate reproduction of the source waveform.  
 

We’re talking about very subtle differences audible only as minor changes in timbre or tonal characteristics.

 

[bluesman]

43 minutes ago, CG said:

Today the same watch is worth roughly 30 times what she paid.  I know - had it appraised somewhat recently when the stupidly expensive maintenance came up.  The resale value has gone up far more than it should have in a sane world based on reality.

Right - and my 71 Rolex cost me $210 new.  When we downsized, we sold our better watches for similar paper gains.  But reality is that the total cost of ownership far exceeds purchase price.  This used to apply to audio electronics too.  I sold a Marantz 7 and 8 for far more than they cost new.  But serious maintenance and mods reduced “profit” to break even.  Tubes alone cost more than the electronics over the years I owned them.
 

I don’t know what watch you have, but service on a Royal Oak is a minimum of about $2k and on a Patek with any complications, it can run well over $3k.  Throw in routine service every decade plus agreed value insurance for 35 years and you haven’t really made a 30 fold gain.

 

These are concepts we use to make ourselves feel better, except for some spectacular and rare exceptions like a Nautilus or a rare and coveted amplifier.

 

FWIW, appraisals usually exceed real world selling prices.  I just love my watches, instruments, audio systems etc for what they do for me.  Anything beyond that is gravy.

[bluesman]

1 hour ago, Jud said:

humidity had penetrated tiny holes in the shrink wrap of nearly all the records and caused the cardboard inside to swell against the plastic, bending the cardboard and severely warping the records inside to the extent that they certainly must have been unplayable.  But they were supposedly worth quite a bit of money as collectible "recordings."

Apropos of this theme in music, the market value of original (ie unmodified, unrepaired) vintage musical instruments is highest if they have the original strings, reeds, pads etc. A Fender Nocaster (the original model that became the Telecaster and Broadcaster) is worth the most if, in addition to being all original, it has its verifiably original strings.  It will be unplayable with 70 year old strings, which will probably break if plucked. It will sound terrible even if the strings don’t break and the pickups are functional.  They’ll be so oxidized and the insulation on the windings will be so broken down that there will be no output. Yet a collector will pay top dollar for the instrument, despite the fact that it’s unplayable.

 

Likewise, a vintage Selmer Mk 6 or cigar cutter will have hard pads that leak so badly the horn won’t play a clear note.  The springs will probably break the first time a key is pressed, after sitting for decades under a bed or in a closet.  And entirely original tube electronics from 1950 will bring big prices despite being useless.

 

Vintage wines trade at exorbitant prices but will probably never be drunk.  The buyer of a rare old bottle has no assurance that it’s not salad dressing - yet auction prices keep rising.

 

Sadly, the value proposition for such stuff is severely warped.

[bluesman]

1 hour ago, ARQuint said:

Any interest?

You can bet your buns there is, Andy!  I’d love that.  And thanks so much for your input and kind words.  Along with all others in the audiophile community, I’ve learned a lot from your contributions.

 

The real question for me is how to help people find more to enjoy in their music.  I suspect that a lot of this can be learned, just as we can learn to identify and describe the subtle characteristics of wine, tailoring, antiques, and portrait painting.  In 41 years as an academic surgeon teaching on the professorial faculties of the 2 largest & highest ranked academic medical centers in our region, I and my colleagues have helped thousands of bright, educated, and experienced students, residents, fellows, and practicing surgeons achieve a fairly uniform fund of knowledge, keep it current, and master newly developed knowledge, technology, and skills (eg microsurgery).  
 

I approach audiophiles the same way (actually, I approach most things this way - it’s how I expand my own horizons).  I’m quite confident that we could teach many, and probably most, to hear and understand at least some of the phenomena we’re discussing in this series.  And I know it would enhance their joy of listening.  If I were 30 years younger, I’d be developing and hosting an online program to do just that.

[bluesman]

1 minute ago, CG said:

I'm not certain that this understanding will help everybody get more out of the home audio systems.

That’s probably true for some.  But for me, the focus is on helping everybody get more enjoyment out of listening to music (which seems to me to be the main enjoyment to be gotten out of a home audio system).  The more you hear, the more joy you get - and you don’t have to be know why you’re hearing it to enjoy it.

 

2 minutes ago, CG said:

Some aspects are plainly genetically common to the species - both people and other animals respond to loud noises and the sound of bad weather approaching.

Here’s another common example of our letting assumptions overshadow deeper truths.  Humans and animals sense and respond to many weather-related changes. The most prominent is barometric pressure, to which we and animals respond in our brains, joints, neural networks, inner ears, skin etc.  We respond to changes in humidity.  We smell weather related changes in the air, and we feel sudden shifts in both the flow of air around us and its temperature.  We often also see the approach of bad weather before we hear it, since light travels 186,000 miles in same second during which sound travels about 5 miles.  So in many ways, we sense the approach of bad weather long before we hear it.  
 

Occam’s razor applies to theories, not to facts.  Human audition is well understood, as is the science behind the creation and perception of sound and music.  One of my main points in this series is that many audiophiles oversimplify their approach to listening and over-complicate their approach to equipment.

[bluesman]

18 minutes ago, CG said:

Well, then!  I guess I'd better pack up all my physics and engineering textbooks, test gear, soldering iron, computer simulation tools, and so forth and put it on sale while I still can!  

Better yet, leave it all where it is, put on your favorite music, fill a vessel with your favorite beverage, sit in your favorite listening spot with your favorite person, close your eyes, and savor the multisensory input.  There are few greater pleasures for us!

[bluesman]

4 hours ago, CG said:

I suppose that if these are all solved problems, then there really is no point of pursuing any of it.

I'm not saying that they're solved.  I'm just suggesting that technology is, at most, no more important than the music itself.  I believe that we're far more likely to benefit from critical listening and learning more about our music than we are from tweaking our systems.  I'm pretty confident that almost every AS participant has a system with more than passable SQ.  And I'm equally confident that our AS friends and colleagues will hear many of the things I've written about, once they start listening for them.  

 

Physics and engineering are both great knowledge sets and paths to further improvement.  But you can't engineer a solution until you've defined the problem and validated your measurement system.  The music holds the keys to defining new goals, and science has the tools to buld the paths to them.  I, for one, believe that many audiophiles cling so hard to the physics and engineering of audio that they leave music (the reason they have audio systems) relatively unexplored.  We need both.

[bluesman]

Just now, Jud said:

I'll let it go for a bit to see if anyone else comes up with it. If not, I'll happily let the cat out of the bag in a while.

So I should go make the pasta dough, huh?

 

Miles threw so many ironic twists into the strange brew in that album (which he called Get Up With It, for those who want to own it) that you need a scorecard to keep track of them.  FWIW, I never liked it enough to buy it.  Oh well - off to the kitchen for a while.  We’re having home made spaghetti and my famous seafood sausage (made this time with salmon and tilapia).  I’ll be back in a while.

[bluesman]

2 hours ago, Jud said:

"Red China Blues" is an example of what I was talking about with Brubeck, big changes that can easily pass unnoticed.  How many beats to the bar?  Listen, and you'll notice it's 6.  A blues that is very nearly in waltz time.  (Waltzes are 3 beats to the bar.)  Miles is saying "You think *you're* cool?  I'm playing a blues in g*ddamn *6* and making it *cook*, motherf**ker!"

A blues in 6/8 is the mildest of Miles’ middle fingers in that tune and on that album.  And it’s not really that far out. Blues in 3/4 or 6/8 time broke no new ground. There’s one on Kind of Blue called All Blues.  
 

The first big poke in the eye was his use of an electronic effect on his trumpet.  Miles’ signature sound was from use of a Harmon mute, which is what many call a wah wah mute. But making the wah sound with a Harmon is done by using the fingers of the right hand to cover and uncover the cup at the end of the center tube, as seen in the video Hopkins embedded.  Miles played without the center tube.  He made the wah wah sound with a guitarist’s wah wah pedal. Contrary to popular belief (and Wikipedia), he still played his standard trumpet -  but he mic’ed and amplified it, using guitar effects like the wah pedal.  And, apropos of the article that started this thread, you can and should be able to tell the difference between a Harmon mute and a wah wah pedal on this tune (and all of Miles’ electric albums).  They sound quite different.

 

Finger #3:  Miles played the organ on that album, although not on Red China Blues. For reasons known only to him, he played several keyboards on his electric albums, perhaps trying to tell his many keyboard players over the years that he really did know better.  No, he was not as good on keys as he was on trumpet.  Once he discovered electric keyboards, starting with the Fender Rhodes, he actually forced some of his pianists to use them even though they hated them.  For example, Keith Jarrett and Herbie Hancock were both initially quite resistant to playing the Rhodes for Miles.

 

Finger #4: He used musicians on that album who were not respected in the jazz community, probably again to poke his finger in the eye of a world he believed was against him.  The bass player who stirred the most controversy was Michael Henderson, who was about 18 when Davis hired him away from Stevie Wonder (which was perhaps finger #5).  Henderson was with Miles for at least 5 or 6 years, but he was simply not a jazz musician.  He was a rock solid funk player, and Miles made a huge statement by having him in the band.  Many of his fans were horrified when he went electric, which was apparently one reason he did it.

 

There’s a lot more, but the idea is clear. Miles set out to tell the world what he thought of it by upsetting as many people as he could and rejecting everything that made him what he’d been.  Remember that he didn’t play at all for about 5 years because he hated pretty much everything and everybody.  In another “social statement”, he was badly injured when he tried to exit West River Drive at about 80 and totaled his Lamborghini.  The guy was seriously screwy, antisocial, passive aggressive, and a generally horrible human for much of his life.

 

 

[bluesman]

1 hour ago, Jud said:

But loudness variation (even an average of 1dB) creates a big problem.

I don’t think level matching is necessary or desirable here, Jud.  I suspect you’re basing your concern on a legendary study showing that listeners (whose knowledge, skill, and experience at listening to music were neither described nor standardized in cohorts) consistently preferred the “SQ” of the louder of paired exposures to the same program through the same equipment (to which they were blinded) at levels averaging 1 dB difference between them.  
 

Even if there is such a study, and I’ve never been able to find it if it exists, it’s not relevant to the task of identifying which of two alternatives is being heard.  The characteristics described by the listener who identified 12 of 12 clips correctly are all heard (by those who can both hear and recognize them, which is not many) in live performance, and the dynamics of live performance are quite variable even for the same music played on the same instrument(s) by the same musician(s) in the same location.  Conductors have their own interpretations of the “correct” loudness, tempo, dynamics, balance among instruments, etc, which adds even more variability to an already broad mix of factors.

 

Even a solo violin piece played twice by the same person on the same instrument in the same location will vary by a few dB - perfect level matching could happen but would be a rare and random event, given the variance and distributions of physical exertion, temperature, humidity, propagating occult stress risers and nonlinearities in the various components of the instrument, etc.  And then there are the emotions of the performer in the moment.  Some nights, you’re “on” and some nights you’re not.......but the show must go on and we hear it as it is.

 

I’m looking forward to Andy’s trial with great interest, and I’m lovin’ this discussion!  Keep the faith, everybody.

 

David

[bluesman]

15 minutes ago, Jud said:

The interesting thing is that 1dB wasn't enough to make it obvious that a change in loudness had taken place, and when I listened, that version of the file impressed me subjectively as more "open" and "clear," not "louder."  I chose it as the best sounding, along with the overwhelming majority of other listeners here on CA/AS.

Different “study”, same problem - a quality assessment was based on a level difference.  In trying to distinguish a Strad from a Guarneri (or a plastic sax vs a metal one, a cornet vs a trumpet etc), perceived quality is irrelevant and to be ignored.  And all that matters in the determination is being able to tell that there are two alternatives and to correctly identify which is being heard in each trial.  Knowing which is which and preferring one to the other are both irrelevant and are not being questioned.

 

It’s certainly true that physical transducers of other forms of energy into sound are nonlinear, and that many speakers and musical instruments sound better somewhere between the ends of their volume spectra than they do when turning the knob to 11.  But we’re talking about variance of one or two dB smack dab in the middle of their sweet spots. And unlike the preferred speakers of many audiophiles, the sweet spots of great Cremonese violins have a very low Q.

[bluesman]

29 minutes ago, Jud said:

 

It will be interesting to see how well listeners will be able to separate those conscious or subconscious impressions from the given task of distinguishing the alternatives.

It would be interesting to see how well listeners can separate impressions of that kind from any single parameter.  I humbly suggest that being able to identify instrument A from instrument B in a forced choice trial is easier than it is to identify power cable A from power cable B.  Any audiophile who hears a galactic and life altering transformation in SQ after slipping 4 rubber washers under an amplifier should be able to beat random chance in Andy’s test.

 

To be statistically sound, the listening trial should be repeated multiple times with the order of presentation randomly altered.  And as a control, at least one trial should be conducted for each listener using 12 presentations of the same instrument.