Milind Kunchur

[SIY]

First line of abstract references an unpublished self reference as proof, why go any further?

There’s a pony in there somewhere.

The next references were to van Maanen, whose work demonstrated that he doesn’t understand sampling or Nyquist any better than Kuncher.

Langmuir nods.

 

[Blumlein 88]

That still does not explain the blinding methods. Was this double-blind, i.e. was the person who flipped the switch not only not in the room when flipping, but also not in the room during the listening? You'd be astonished at what a subject will unconsciously use for distinctions.

Also, to what extent was the through-amplifier gain tested in both positions. This may have been answered but I'm at work and kind of swamped.

Matching is best if to .1dB or better. .2dB gain difference is more than enough to cause extremely strong preferences.

He did match both inputs at the speaker leads to less than .1 db.

 

[dc655321]

First line of abstract references an unpublished self reference as proof, why go any further?

To be fair, the reference is "in print" (JAES).
FWIW, I did not look at the paper you are referring to, only the one linked here.

 

[SIY]

To be fair, the reference is "in print" (JAES).
FWIW, I did not look at the paper you are referring to, only the one linked here.

Footnote 1 is wrong. And intellectually dishonest. I read no further.

 

[j_j]

He did match both inputs at the speaker leads to less than .1 db.

Now we're down to "how did he do that, and when and where was the switch flipped, and by whom", coupled with "and how about order effects".

It is getting somewhat better. I will say, however, that the question of "what cables was he using" comes to mind, I read some of the measurements, which are a touch, well, odd. I will assume for the minute his measurements are correct.

 

[UKPI]

"Cable pathways between audio components can affect perceived sound quality“ – Journal of the Audio Engineering Society, in press (2021) Milind N. Kunchur, University of South Carolina

Since the paper doesn't explain the method and equipment used for level matching and double blind procedures in detail, we can only speculate.

Of the 18 virgin trials, 8 were B/A and 10 were A/B, which scored 8 out of 8, and 5 out of 10, respectively.

Maybe something went wrong with the B/A setup?

 

[Blumlein 88]

The other obvious thing to wonder is using xlr vs rca. The noise he talks about in the listening setup is mhz level noise, and is much higher in the rca cable not surprisingly. The Spectral being of such wide bandwidth maybe even in some way is effected though it is hard to think it would end up below 20 khz at a level one can hear. Also now that I think about it how did he match levels to +/- .0045db unless something provided gain or attenuation to match the two signals.

Plus as J_J mentioned how was the blinding with the person switching inputs? Of course there are so many reasons to wonder about the result we need replication.

 

[Frank Dernie]

Maybe the amp they used has marginal stability with cables without filters?
This from the dealer years ago:

"Whereas other Spectral Audio components (i.e. their preamplifiers and digital source components) may be used with any other brand of equipment, Spectral mandates that their non-Universal amplifiers be driven by a Spectral preamplifier to ensure reliable operation. Spectral Universal amplifiers however may be driven directly by any brand of preamp, D-A, or surround sound processor. In addition, Spectral amplifiers need to be used in conjunction with the proper Spectral-approved MIT interconnects and speaker cables. These usage guidelines are now a condition of Spectral's warranty on their amplifiers. There are several reasons for these guidelines:

Because of their wide bandwidth, Spectral amplifiers can amplify high frequency signals in the MHz range at massive power levels. Therefore radio frequency noise must be kept out of the amplifier. The MIT interconnects contain RF protection, and Spectral preamplifiers are stable at these high frequencies.
Unlike most all solid state preamps, Spectral preamplifiers are true discrete designs, are DC coupled, and can output the high current that the (non-Universal) Spectral power amps are designed to be driven by. Other brands of preamplifiers often use integrated circuits or other devices which will probably not be able to drive the amplifier properly.
As part of their stability design, Spectral amplifiers use an output network found in the proper MIT speaker cables. (Spectral has found that by placing this network close to the load, i.e. the speaker, they can achieve superior results.)
What will happen if you don't follow these guidelines? Well, it's a probability thing. You may not encounter any problems—or you may encounter premature failure of some components in the amplifier. Or (worst case) your amplifier may suffer severe RF problems, and you will have to pay to have all the output devices replaced (which is expensive). Therefore, we highly recommend that you follow Spectral's warranty and operational guidelines."

I don't know what a Spectral Universal amplifier is but when I had my DMA-180 I was told to use cables with filters (my dealer recommended Goldmund but he was both the Goldmund Europe distributor and the European importer of Spectral).

I wonder if, like the older Naim amps, the cable for use on Spectral amp needed a narrower than usual set of L, C and R characteristic for the amp to remain stable. I notice he had filter networks at the speaker end of his biwired cables, different for woofer and tweeter.

 

[Cbdb2]

So the switching is done with the amp inputs and the signals go thru 2 different input circuits, one balanced the other unbalanced and no one has a problem with this?

 

[Kal Rubinson]

So the switching is done with the amp inputs and the signals go thru 2 different input circuits, one balanced the other unbalanced and no one has a problem with this?

Of course. It is an obvious issue with the author tries to wash away.

 

[NTK]

So the switching is done with the amp inputs and the signals go thru 2 different input circuits, one balanced the other unbalanced and no one has a problem with this?

It isn't true that no one has a problem with it. But it seems to be true that no one on none of the JAES review committee reviewers of this particular submission has a problem with it

[Edited per JJ's comments]

 

[Cbdb2]

It isn't true that no one has a problem with it. But it seems to be true that no one on the JAES review committee has a problem with it

Yea thats what I meant. And wasn't the balanced line driven by a balanced output? So the signals actualy go thru 2 different circuits each. And then he adjusts the gain after the amp whit every switch. How? With a resistor? And then he avoids using a A/B switch in case it alters the sound. LOL. Really starting to have doubts about the AES, the E seems to be disappearing.

 

[Wes]

The research uses special signals (e.g. 7 Khz square wave) to detect certain thresholds. ...

It might be interesting to send an ai system roaming thru a music catalog to detect how often any musical waveforms change as quickly as the leading or trailing 'edge' of a 7 kHz square wave.

 

[j_j]

It isn't true that no one has a problem with it. But it seems to be true that no one on the JAES review committee has a problem with it

AES reviews are not always done by the same people. For instance, they didn't send this one to me to review, but that's ok, I've had two run-ins with Kunchur already.

 

[krabapple]

I'd be curious to see the reviewers' comments. (I presume that JAES requires more than one reviewer?)

Also maybe a moderator could correct the spelling of ML's first name in the thread title? It's MILIND.

 

[JustJones]

If this shouldn't be posted here then please delete it. I found the paper out on the internet in an "audiophile" forum. I understood very little of it, one thing that jumped out to me is the LCR of the cables were very different.


http://boson.physics.sc.edu/~kunchur/papers/Audibility-of-cable-pathways--Kunchur.pdf

 

[SIY]

If this shouldn't be posted here then please delete it. I found the paper out on the internet in an "audiophile" forum. I understood very little of it...

That's OK, neither did the author.

 

[danadam]

If this shouldn't be posted here then please delete it. I found the paper out on the internet in an "audiophile" forum.

It was already posted in this thread :

"Cable pathways between audio components can affect perceived sound quality“ – Journal of the Audio Engineering Society, in press (2021) Milind N. Kunchur, University of South Carolina

 

[audio2design]

AES reviews are not always done by the same people. For instance, they didn't send this one to me to review, but that's ok, I've had two run-ins with Kunchur already.

I see we agree on some things

 

[audio2design]

Table II: Measured electrical parameters: resistance (R), inductance (L), and capacitance (C) and calculated quantities: characteristic impedance Z= (L/C)1/2

Perhaps someone could inform the not to good Dr. Kunchur that characteristic impedance at high frequencies is not the same as at audio frequencies rendering this value in his table worthless.