Geoff and other engineers - what do you make of this - substance or foo?

http://www.stereophile.com/interviews/597kondo/#b4Rj4oSowybTvOxX.97

http://www.tnt-audio.com/intervis/audionote_e.html

I have just stumbled across these articles in which Hiroyasu Kondo—founder of Audio Note Japan (now deceased) speaks about his experience of designing audio equipment and the difference in 'sound quality' between different capacitors, different tubes, silver vs. copper, annealed vs. not. I am perhaps more an objectivist or try to be and started wincing and smirking with distrust when I read this. He also talks about the importance of good measurements but also intimates that not everything can be measured. He used to work for Sony and worked on analogue and emerging digital technologies but got frustrated by the volume production and what he saw were extreme compromises to 'sound quality'. He was well respected by many engineers across the world and is still spoken off highly.

I recall being at medical school and picking up a stethoscope for the first time - I could barely hear heart sounds let alone a heart murmur. The cardiologist who was teaching, impressed on us that it takes many years of careful listening before achieving competency. There was technique: close your eyes, slowing down your breathing, time the pulse in your finger with the heart sounds arriving in your years and tune in. Over a few months and listing to over a hundred patients' heart sounds, I started 'tuning in' and got better at differentiating regurgitation from stenosis. This was a psycho-acoustic skill I had to learn. The senior doctors taking their examines practised this day and day out before their clinical exams where they would almost certainly be asked to place a stethoscope on a chest wall and diagnose the heart murmur there and then on the spot by listening alone. Heart murmurs can also be diagnosed using echocardiograms where the Doppler flow and movement of the valves can be visualised on a screen with quantifiable data on flow, turbulence, distorted flow etc...In this case most of us can and never will be able to hear heart murmurs and differentiate them but with time and careful training one can start to heart slight turbulences generated by a slightly malfunctioning valve. A few will get even better at it and some won't. https://en.wikipedia.org/wiki/Heart_murmur (if you're interested)

Geoff and other engineers what do you make of this?

Have you heard differences in components such as capacitors, wires, valves, DAC chips, different circuit board designs, yourself?

Where does one draw the line in objective vs. subjective experience over years?

Is there credence to the argument that not everything can be measured (or not at the moment) until we devise a way to visualise air molecules moving/vibrating in space like we can with an echocardiogram?

Some out there are blatantly trying to cheat the buyer and come up with all sorts of expensive audiophoolery but some chaps in the audio engineering world remain dedicated and focused to this more subjective than objective approach and swear that these things are just as if not more important. Is this an auditory illusion? lifelong delusion? an as yet undiscovered country?

http://nwavguy.blogspot.co.uk/2011/05/subjective-vs-objective-debate.html

 

seethroughyou:

Have you heard differences in components such as capacitors, wires, valves, DAC chips, different circuit board designs, yourself?

Everything in the signal path contributes (positively and negatively) to the final resulting sound character. That is the simple answer.

seethroughyou:

Is there credence to the argument that not everything can be measured (or not at the moment) until we devise a way to visualise air molecules moving/vibrating in space like we can with an echocardiogram?

Measuring sound vectors is nothing new. People that work in NVH labs around the world have been doing this for decades by taking vast numbers of measurements with a single microphone. Now we have Siemens/LMS to thank for the brilliant sound brush which makes this process much faster. If you ask an engineer, he will tell you air is a fluid which can be analytically modeled quite easily and efficiently. Measuring large noise fields/sound sources can be accomplished with a microphone array and a SCADAS. 

When I was a student I worked in the university's Acoustic & Dynamics Lab and we had a 2-axis CNC driven 64 microphone array that could sweep the entire section of our anechoic chambers.

The honest, hard working and passionate engineers want to make quality components within their given constraints (usually cost, space and power consumption). The 'blatant cheaters' probably care about quality sound, but they do not have measurements to support their arguments and prove their technology.

I am an engineer with 25+ years experience in testing and inspection. Just because you can measure something doesn't mean it's important or even applicable.

In welding inspection, there's an old saying: "pretty ain't in the code." Just because a weld doesn't look nice from a lay person's perspective, doesn't mean it doesn't meet specs for size, strength, etc.

In audio there essentially is no "code" to test or inspect against. Sure there are electrical safety requirements and other standards, but what makes a great piece of audio or video gear? Put 10 random "authorities" in a room and you'll have at least 4 if not 10 sets of criteria.

Personally I think you can achieve great sound for about the cost of 3 or 4 car (monthly) payments. Others think an "entry level" turntable of acceptable quality will cost more than that.

Some believe the most important element of the system is the speakers. Others insist you have to invest the most on the input device.

What's the point? It only matters what you think.

Hi,

In my opinion, this issue can be broken down into two discussions: the specific and the general.

 

Specifically, there has been much discussion and analysis over the years regarding the specific performance parameters (typically focusing on non-linear behaviour) of specific components (such as capacitors) in a specific role (i.e. filtering) in an analogue audio circuit. Knowledgeable persons who have these discussions know that you can't make general statements (i.e. "Electrolytic capacitors are bad") - you have to be very specific about the intersection of the application, the implementation, the intention, and the expectation in order to make a valid statement about validity... (or something like that...) For someone interested in the specifics of this, a good place to start is this book, however, there are plenty of other valid analyses and scientific papers. I will say that any time I see a statement like "You should never use a Philbert Flange in an audio circuit." Or "You should always use a Grapple Grommet in an audio circuit.", I am reminded of Wendell Johnson's advice: "'Always' and 'Never' are two words you should always remember never to use."

 

The general discussion revolves around the measurable vs. the audible. There should be no doubt that there are some differences in audio quality or performance (whichever word you prefer) that are measurable but not audible. The problem is that, if you don't disagree with this, then you have to admit that the logical conclusion is that an audible difference MUST be measurable. However, there is no question that some audible differences will not show up in the usual set of measurements. If you do a frequency response measurement using an MLS-based system, you are (at least in theory) only measuring the linear behaviour of the system - so you will not detect the non-linear (distortion) effects. This is one of the main reasons I have the job that I do at B&O. My job is, after the usual measurements have been done, to listen to a loudspeaker and go back to the engineers to discuss what other measurements we need to do (or possibly even invent) to investigate an audible problem that has not been detected.

In other words, if you can hear something but you can't measure it, you're not doing the right measurement.

 

Finally, we have to be careful about the use of "objective" and "subjective". "Objective" means something that is repeatably measurable. However, the measurement tool could easily be a listening panel doing a perceptual evaluation. If you have two loudspeakers that are in every respect identical except that one has 12 dB more bass than the other, an objective perceptual evaluation from a well-designed listening test will show that one sounds like it has more "bass" than the other. An objective physical measurement can tell you exactly how much more level you have at a given frequency. However, a "subjective" evaluation will only tell you which one the listener prefers. You like more bass, I like less bass. We will both agree that "Loudpeaker A" has more bass and "Loudspeaker B" has less (the objective perceptual evaluation) but we will disagree which is better (the subjective evaluation).

 

As for sellers of snake oil... Many emperors have been known to buy nice new clothes, and many people do good business selling to them... Start by following Duke Ellington's advice: "If it sounds good, it IS good." - and make sure that it's not someone else telling you that it sounds good by following Brian's advice: "You've got to think for yourselves."

 

Cheers

- geoff

 

Fabulous post, Geoff!

Im looking forward to hearing the BL 90's on Tues in NYC!

Barry

This might be slightly off-topic but.........

I had been using my Olive One for some time with its analogue output connected to a 1970's BM2000 and early S45 speakers.

A couple of weeks ago I decided to effectively cut-out the BM2000 and connect the Olive's Class D amplifier outputs direct to the S45’s. At first I thought that the results were fine, but not quite as good as using the BM2000. However, after a couple of days use the Olive output really seemed to improve, and I would say the results were every bit as good as the BM2000.

I am using the same long-worn-in speakers, so what has happened? Have my ears adapted in some way, or is it possible that Class D amplifiers can actually change and improve their output after a few hours use?

Graham

Of course on the stethoscope question, get a Littmann 4100 - sound by B&O!!