Analysis of Paper on Measurements of RCA Cables by Kunchur (Video)

[Cars-N-Cans]

No one who understands electronics will be worried by Ethan's adjustment of volume to tune the deepest null. Do people realize he is adjusting something at the level of one millionth of a volt?

Who says we are worried? I’m not, but I understand, as do you. But per my last post audiophiles either don’t, or choose not to either due to ideolical reasons, or because they have a business that market such products as audiophile cables and have a vested interest. If they can split hairs, they will and the further you can shut the door, the better. And anyone who has used measurement equipment knows he is down in the uV region. However I agree that the AP equipment is very much a luxury few have. But most people don’t have a grasp of the fact that dB is a logarithmic unit and the ranges we are talking about are enormous. Forget the 110 dB absolute limits used as a CYA by Amir, most people won’t be able to hear more than 45 dB down in a recording, and that’s using headphones. I’d imagine most speaker setups will be some 5-10 dB worse. Just about anything other than solitary naked tubes and MOSFETs without feedback will be mostly, if not fully transparent during casual listening.

 

[Cars-N-Cans]

No one who understands electronics will be worried by Ethan's adjustment of volume to tune the deepest null. Do people realize he is adjusting something at the level of one millionth of a volt?

Even digitally you'll not get deeper nulls. As I recall Ethan is boosting levels up to 80 db to just barely hear low level hiss. Using Deltawave and good ADC's gives about the same performance as Ethan's analog device. You cannot get to complete silence due to thermal noise. You can get results in the -100 to -110 db level. With better ADCs you can do better. Amir has gotten down around -115 db and near -120 db, but he has the APx 555 to work with. You get the same results whether comparing cable A to itself 5 minutes later or cable B to cable A. Also it might interest people to know the left and right channels of even very expensive DACs don't null this deep.

And just so we don't have to argue over banalities, the FFT gain will make Amir's measurements look nicer. Additionally I appreciate Ethan's honesty in showing us everything even if he will be slightly worse noise-wise, but then again, they will just use your honesty against you. In some ways I feel these tests only add fuel to the fire as it gives the impression that we are potentially unsure and need the measurements as a possible "crutch." Quite to the contrary, I'm sure. Really damn sure and don't need to do any measurements at all, but then if we don't do such things, those who are not subjectivists, but not technically as knowledgeable wont have the benefit of actually seeing actual numbers put to statements. Additionally its ammunition if you want to take said audiophiles to task.

 

[Blumlein 88]

And just so we don't have to argue over banalities, the FFT gain will make Amir's measurements look nicer. Additionally I appreciate Ethan's honesty in showing us everything even if he will be slightly worse noise-wise, but then again, they will just use your honesty against you. In some ways I feel these tests only add fuel to the fire as it gives the impression that we are potentially unsure and need the measurements as a possible "crutch." Quite to the contrary, I'm sure. Really damn sure and don't need to do any measurements at all, but then if we don't do such things, those who are not subjectivists, but not technically as knowledgeable wont have the benefit of actually seeing actual numbers put to statements. Additionally its ammunition if you want to take said audiophiles to task.

I was referring to Amir's cable tests where he did a null more so than his measurements which as you say have FFT gain. Otherwise I don't disagree with what you have posted.

 

[Cars-N-Cans]

I was referring to Amir's cable tests where he did a null more so than his measurements which as you say have FFT gain. Otherwise I don't disagree with what you have posted.

Oh ok. My bad I misconstrued part of your earlier post.

 

[krabapple]

Double blind peer review means that the reviewers are not known to the author and the author is not known to the reviewers. It's not an uncommon term.

I'm fully aware of what the term means.

It's the syntax that got me wondering. 'Peer reviewed journal" is fine, "double-blind peer review" is fine, but "double blind journal" would be strange. "Double-blind peer reviewed journal" is acceptable as long as 'double-blind' is read as a modifier of 'peer review', not 'journal'.

It would be easy to avoid the ambiguity by using better writing, which makes me wonder where such journals come from (geographically).

 

[RigorDude]

Let’s just hope he’s more knowledgeable with superconductor physics than he is with anything at all associated with audio. His papers literally read like they were written by some random audiophile after someone gave him a lab and an oscilloscope to play with. An example would be trying to test the ”timing resolution” by literally soldering random capacitors and resistors to a relay, cause digitally synthesizing it would cause too much aliasing and filter lag, or something which would clearly bias the experiment. So… we’re just gonna whip out the soldering iron from AliExpress and go to town! The pinnacle of true experimental control excellence.

Edit: The experiment is clever in that it uses a low pass filter and a square wave to alter the "attack" of the waveform by essentially altering the harmonic content out at supersonic frequencies that are, as the paper shows via its own testing, inaudible. In principal, any discernable differences should be due to hearing the "envelope" of the waveform and thus showing high resolution in the time domain. But what he fails to do is have any sort of control (both in terms of actual controls and also as an experimental control to screen out that obvious result) that can maintain the same fundamental energy as the various filters are switched in and out of the signal path, and in Table 2 he even shows this that the amplitude clearly changes, which is what the test subjects are really reacting to. They handwave that its below the JND and can be "neglected" but show me the actual measurements from the amplifier output! Essentially he is just establishing a dB threshold above which changes become audible to the test subjects since to the ears that is just a sine wave at the fundamental instead of a square wave. On top of this there could even be resonances in the headphones themselves, but the "measurements" seem to indicate otherwise. And god knows what other issues there could be. Be interesting to have someone who is knowledgeable in Kuncher's main field of work peer review it, but I would not want to waste their time with such trivial things.

I make no claims for the merits of his cable paper. On the other hand, dismissing his work for not conforming to standard audio engineering practice strikes me as questionable.

 

[Cars-N-Cans]

I make no claims for the merits of his cable paper. On the other hand, dismissing his work for not conforming to standard audio engineering practice strikes me as questionable.

In the paper I quote, it was not questionable practices, it was that it was just wrong. He didn't bother to even investigate the obvious alternate explanations of not only the dB changes induced by the crude setup, but that it was even doing what he claimed it was doing, which is only providing the harmonics that are in a square-wave. Edit: Going back and re-reading what I originally wrote, I assume you are referring to his work in superconductors. After his poor performance in the audio realm, I couldn't help but wonder. I did have a look at them, and they are quite different. Only a few pages with the data and a brief conclusion, as opposed to the audio ones which are more verbose. It could just be that he is so incompetent, but overly enthusiastic, about audio that he does not he how fundamental his errors are. Who knows.

 

[Bruce Morgen]

...and here we go -- again....

 

[DonR]

...and here we go -- again....

ooh my... that is some serious words in there. Calling out @amirm directly as dishonest. I had a chuckle at using "cult" to describe ASR.

 

[BDWoody]

...and here we go -- again....

So, still no blind test results?

 

[Astoneroad]

ooh my... that is some serious words in there. Calling out @amirm directly as dishonest. I had a chuckle at using "cult" to describe ASR.

 

[amirm]

...and here we go -- again....

Thanks. Take away the personal insults and commentary and there is zero substance there. Amazing that he thought he should bother posting something like this.

 

[BDWoody]

ooh my... that is some serious words in there. Calling out @amirm directly as dishonest. I had a chuckle at using "cult" to describe ASR.

I think someone needs to send him this:

 

[Bruce Morgen]

Take away the personal insults and commentary and there is zero substance there.

Can we quote you?

 

[voodooless]

Can we quote you?

You just did

 

[Peluvius]

The academic community needs to stand against this type of fraud (I hear crickets)?

I never went down that path but man there is serious integrity issues here if this is the sort of tripe that passes for academic rigour these days.

How do you introduce summary and conclusion content that was not included as part of the subject document? Even I remember from my Uni days (lost a few marks there ). Surely even a relaxed peer review process should pick that up? That's aside from all the other tripe in his document (and yes, I have subsequently looked into a few of the issues Amir raised as I was finding it hard to believe this gents paper could be so off course). I can therefore confirm beyond a shadow a doubt that his statement "And his followers appear unwilling to do their own fact checking", is incorrect

 

[voodooless]

...and here we go -- again....

Okay, let's do this then..

A spectrum analyzer focuses on only the first and is not best suited for studying impulse response and transients4 , which are influential in defining instrumental timbre

Citation is needed for the last part.

Humans can detect a time difference between left and right ears of 10 microseconds at a frequency of 900 Hz. This time is 100 times shorter than 1/f ~1 millisecond. This has been known for over half a century5,6 .

Once again with the inter-aural measurements... This has been thoroughly debunked right here. This is different from a phase difference on a single mono channel at different frequencies. Secondly, the test was done with headphones, not regular speakers. That makes a massive difference. 10 uSec is about a 3.4mm difference. Getting that done on a speaker pair and a listener is very difficult because the distances toward the sources should be lower than during the whole test. Good luck without a vice, let alone what this says about listening to audio in practice. Not to mention what this actually means for real music. And then there is the matter of why this matters for a cable? What cable introduces a 10-microsecond difference anywhere in the audible spectrum? This is just a massive strawman.

Mr. Majidimehr seems to have a nonsensical notion that digital temporal resolution equals ~T/2n-1 (T=sample period, n=bit depth). That corresponds to the shortest shift in a waveform’s edge that can be detected. It does not represent the fineness of features in a waveform that can be resolved, which is in fact limited by approximately the sampling period T (not T/2n-1 !). For further explanation, see the review paper7

So reference 7 is "7 J. D. Reiss, “A Meta-Analysis of High Resolution Audio Perceptual Evaluation,” J. Audio Eng. Soc., vol. 64, no. 6, pp. 364–379 (2016 Jun.). http://dx.doi.org/10.17743/jaes.2016.0015.". The paper says nothing about waveforms or how they are limited to the sampling period T. From the paper's conclusion:

Furthermore, though the causes are still unknown, this perceived effect can be confirmed with a variety of statistical approaches and it can be greatly improved through training.

You really thought people would not read your references and see through your Bullshit?

And that doesn't get us started about the paper itself, because there is more than enough to talk about right there...

From his statement (“4 nanoseconds is 250 megahertz…why would I care about a 4 nanosecond pulse to go through anything”), Mr. Majidimehr seems to not understand one of the most fundamental concepts in electromagnetism, which is the connection between low-frequency properties such as the dielectric constant and signal propagation speed. This is analogous to relating the density and pressure (low-frequency properties) of a gas to its speed of sound8 .

So you're referencing how helium affects your voice's pitch? It's just a formula, that is not an actual reference. The helium just makes your vocal cords swing faster because the air is lighter. Why is that relevant to cables? This is just another strawman.
"Signal propagation speed"? How is that relevant to audio frequencies through a copper (or equivalent) conductor?

 

[Cars-N-Cans]

"Unfortunately, instead of pursuing the tedious route of formal science, some people join the cult of a self proclaimed guru."

I don't even know what to say to that other than it being such a profound failure of self-awareness. Its just a case study in why self-critique is so important, because handing one's own ass to oneself is much easier given it is already attached in some fashion. Much harder to endure when someone else has to tear it off and hand it back to you.

 

[voodooless]

You really thought people would not read your references and see through your Bullshit?

Funny how that works, I missed this piece of absolutely profound lack of self-awareness:

read papers in their entirety including footnotes, references, and other details

Yeah, that is exactly the problem! We did!

 

[Sokel]

I'm surprised by the very existence of the response.
I thought multiple MSc/BSc/PHd's followed the unwritten law to not answer critique but by their peers.
That's REALLY strange.