• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Milind Kunchur Blind Cable Test

SimpleTheater

Addicted to Fun and Learning
Forum Donor
Joined
Jun 6, 2019
Messages
927
Likes
1,789
Location
Woodstock, NY
I know that people have stated, with links, that Kuncher made many mistakes in this paper:

Thread showing inaccurate measurements:

However, I have not seen why his blind test was invalid. Can someone point to an issue where he may have allowed a sighted A/B comparison, or where the dB level was off?

We are often told to “do a blind test”, and I’m assuming no one is asking for detailed mathematical formulas, just the method of conducting the test and the results.

Thanks in advance.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
I think the disputed paper was not about the cable test but about ultrasonics.
 

Sgt. Ear Ache

Major Contributor
Joined
Jun 18, 2019
Messages
1,894
Likes
4,150
Location
Winnipeg Canada
I skimmed through the first paper there. It's pretty involved and I can't really specifically point to anything that might be "wrong" about it or the conclusions. But it seems to me that a lot of effort was employed to make sure that IF there was any audible difference, that it would be heard. There was an indication early on referring to wanting to achieve some sort of "audiophile" standard - meaning the test was trying to live up to the way audiophiles assess things. To me, that's ludicrous. Audiophiles aren't doing strenuous things to make accurate comparisons. They are connecting stuff, and listening to it. They plug a new cable in, listen to some music, and then pronounce the new cable to be awesome compared to the old cable which is no longer connected and never will be again...they aren't making a back to back comparison of any sort going back and forth and immediately comparing.

I'd be interested in seeing how many people could go into a room...listen to a system connected by cheap cables, then leave the room and go back an hour later and listen again and determine by ear if better cables were now being used. I mean that's the actual way audiophiles draw their conclusions.
 

RobL

Addicted to Fun and Learning
Forum Donor
Joined
Mar 4, 2021
Messages
925
Likes
1,509
Correct me if I’m wrong, but from what I’ve read he found that in a certain passage of music listeners were able to distinguish noise floor between a $1000 xlr cable plugged into the balanced output from a $25 RCA cable plugged into the single ended output of the same device. Not really a stunning discovery as of course the balanced output rejects noise better but perhaps he was actually probing the limits of human hearing’s ability to detect noise?
 

Sgt. Ear Ache

Major Contributor
Joined
Jun 18, 2019
Messages
1,894
Likes
4,150
Location
Winnipeg Canada
Correct me if I’m wrong, but from what I’ve read he found that in a certain passage of music listeners were able to distinguish noise floor between a $1000 xlr cable plugged into the balanced output from a $25 RCA cable plugged into the single ended output of the same device. Not really a stunning discovery as of course the balanced output rejects noise better but perhaps he was actually probing the limits of human hearing’s ability to detect noise?

yeah that was the other issue I had - it's comparing expensive XLR to cheap RCA. I mean I think most of us would be somewhat more open to the notion that there might be some audible difference between rca and xlr (whether true in any sort of normal listening circumstance or not). The more interesting test would be cheap RCA vs. expensive RCA, or cheap XLR vs expensive XLR...
 

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,003
Location
Albuquerque, NM USA
Interesting paper with seemingly sound methodology, apart, of course, from using RCA on one hand, and XLR on the other. It strikes me as somewhat obvious, a second set of cables with the opposite terminations should have followed. If preference followed the cable from one to the other, I would have been impressed. If instead it remained with the XLR input, case closed. And yes I understand the reason they chose to do it this way, but you don't play the World Series at one park.

Ultimately, it still gets a big yawn. For 950.00 I could upgrade the Proac's or buy a pair of capable subs--something that would have a true impact, and didn't require 5:40 seconds x 5 of my life to discern. I am absolutely certain that free software could replicate the effect, and even exaggerate if it so preferred.
 

Cbdb2

Major Contributor
Joined
Sep 8, 2019
Messages
1,530
Likes
1,485
Location
Vancouver
I guess logic is not needed to become a Phd. He was comparing a balanced system to an unbalanced one, not just the cables. Only 3 samples per person and most got 2 out of 3 guesses correct. Not much proof. I knew there would be some fudging going on when I read this.

"As explained in the “Introduction” section, HEA level components tend to already be extremely good in specifications and subjective (sighted) sonic performance. As a result, blind tests proving audible distinguishability between components (even major components like levelmatched amplifiers and CD players, not just cables) tend to fail (these negative results don’t get published in journals)."

His lack of logic= If people can't hear a difference in a blind test between amps than theres something wrong with the test. So let's design a test that proves our point or we will never get this published.
 

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,003
Location
Albuquerque, NM USA
I guess logic is not needed to become a Phd. He was comparing a balanced system to an unbalanced one, not just the cables. Only 3 samples per person and most got 2 out of 3 guesses correct. Not much proof. I knew there would be some fudging going on when I read this.

"As explained in the “Introduction” section, HEA level components tend to already be extremely good in specifications and subjective (sighted) sonic performance. As a result, blind tests proving audible distinguishability between components (even major components like levelmatched amplifiers and CD players, not just cables) tend to fail (these negative results don’t get published in journals)."

His lack of logic= If people can't hear a difference in a blind test between amps than theres something wrong with the test. So let's design a test that proves our point or we will never get this published.
Frankly, I am surprised it was published--I thought the AES is all about engineering (and by extension science). Am I wrong in thinking this?
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
I really don't see much that is "wrong" with this result. It is a surprising result. Kunchur always tries to champion the audiophile way of things which is suspicious as his earlier papers always portrayed tests with obvious problems he glossed over. You wonder if something was left out about this one. In any case, it appears there was a difference in RCA vs XLR which from the gear and measurements provided appears to have no reason. Obvious test #2 would be to change the fancy high end XLR for one of the cheaper, but well made XLR's and see if a difference is still observed. Or for someone else to replicate it and see if the result is robust.

A simple thing he could do looking for differences is replicate his test setup, get a high performing ADC (even say an RME) and record the output of the two systems and compare it with something like pkane's Deltawave software. While the RCA has more noise it didn't appear to be to the level of creating an audible difference.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
yeah that was the other issue I had - it's comparing expensive XLR to cheap RCA.
It is actually more that that. He compared the RCA connector, to RCA cable, to RCA connector. For the other, he used balanced output of one gear, XLR cable, balanced input of the other gear. In other words, both the cable and signal paths were changed. It is entire possible that there are differences in those two interfaces outside of cabling. I measured a Schiit DAC for example that had different response in balanced vs unbalanced:

index.php


He should have made this absolutely clear in the paper but failed to do so. I commented on this on AES forum but the moderator deleted all three of my comments. :( He said I should discuss it with him in a live chat they were having later!

Also, while results had significant statistical power, in no way did it back people who claim they can instantly tell one cable from another. For that, they would have had to get 100% right which they did not.
 

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,003
Location
Albuquerque, NM USA
It is actually more that that. He compared the RCA connector, to RCA cable, to RCA connector. For the other, he used balanced output of one gear, XLR cable, balanced input of the other gear. In other words, both the cable and signal paths were changed. It is entire possible that there are differences in those two interfaces outside of cabling. I measured a Schiit DAC for example that had different response in balanced vs unbalanced:

index.php


He should have made this absolutely clear in the paper but failed to do so. I commented on this on AES forum but the moderator deleted all three of my comments. :( He said I should discuss it with him in a live chat they were having later!

Also, while results had significant statistical power, in no way did it back people who claim they can instantly tell one cable from another. For that, they would have had to get 100% right which they did not.
I'm not sure I agree on that issue, but do agree with the shortcomings in methodology. The question posed seemed more a matter of whether audible differences could be detected--if you favor A twice as often as B, then that means something, especially when the evaluations were sustained over a period of several minutes, and involved more than one listener, and who all were presumably well acquainted with the program material (and yes, I know that musicians are mot known for being "golden eared" any more than a guy off the street, once trained to listen.). So I found it interesting, but as I posted above, such a minute difference if real and not simply balanced v unbalanced, would not justify the price were one willing to introduce true EQ into the system and spend hours upon hours of making minute adjustments. Personally, I have neither the patience nor the motivation to commit to such a painstaking enterprise. Again it would have been simple enough to swap the connections--presumably someone there still knows how to solder these types of connectors, and can afford the good stuff. :rolleyes:
 

Sgt. Ear Ache

Major Contributor
Joined
Jun 18, 2019
Messages
1,894
Likes
4,150
Location
Winnipeg Canada
I can't imagine any circumstance where EQing would (or should anyway) take "hours and hours of minute adjustment." As far as I'm concerned, if you've spent 15 minutes making EQ adjustments and it hasn't resolved whatever issue you're concerned with, then the issue probably isn't resolvable with EQ. It's an extension of audiophile mythology really - the idea that tiny, minute little EQ changes are audible. You can make some pretty significant EQ alterations and then switch back and forth between EQ on and EQ off and have a real hard time telling the difference. For me that's another "canary in a coal mine" example (similar to the difficulty of hearing differences between high bitrate lossy and lossless) that informs my notions about just how goofy so many of the classic audiophile tropes relating to identifying differences between things that measure exactly the same actually are.

I don't know anything about the fellow who wrote the paper, but my initial quick read definitely left me with the feeling that he's basically an audiophile interested in finding scientific proof of audiophile claims. In the end, it all comes off as almost a big strawman - when these debates about boutique cables arise, it's never about cheap rca vs expensive xlr or cheap speaker wire vs expensive power cord. It's about cheap rca vs expensive rca etc etc. It's like saying "oh, you think you can't hear the difference between cheap and expensive amps if they measure the same? OK well check out my blind test where I compare this cheap dac to this expensive amp and 2 out of 3 people could hear a difference." Whether there should be an audible difference between those 2 things or not, why would that be the way you would formulate the test? You're just introducing a variable that muddies the conclusion. Does the paper (if it proves anything at all) prove there's an audible difference between cheap and expensive cables or does it just prove that XLR is audibly better than RCA?
 
Last edited:

FrantzM

Major Contributor
Forum Donor
Joined
Mar 12, 2016
Messages
4,337
Likes
7,720
To me that is the HEA reacting to the fact that [people are becoming more aware of the advances of measurements and of Science in audio reproduction evaluation and design. Let's face it people, if your $15,000 DAC fares no better than a $9.oo dongle, you're in trouble; even your allies , S'Phile and TAS are now alluding that Topping less than $1000 preamplifier, is, could be, perhaps, you-never-know, well-maybe- as good as ... or at least challenging ... The HEA industry knows, it is in trouble. What better way than an "academic" paper to show off? To build some "cables are different" momentum? to delay the inevitable?
 

Cbdb2

Major Contributor
Joined
Sep 8, 2019
Messages
1,530
Likes
1,485
Location
Vancouver
yeah that was the other issue I had - it's comparing expensive XLR to cheap RCA. I mean I think most of us would be somewhat more open to the notion that there might be some audible difference between rca and xlr (whether true in any sort of normal listening circumstance or not). The more interesting test would be cheap RCA vs. expensive RCA, or cheap XLR vs expensive XLR...
They may have tried that and got no results.
Frankly, I am surprised it was published--I thought the AES is all about engineering (and by extension science). Am I wrong in thinking this?

Looks like the AES is changing. Science, for the general population is becoming personal mythology and corrupted by woke. Proof is not necessary, just a strong belief. For example, genetics don't define male/female anymore. We've all seen the proof on many audiophile sites. Look at the difference in Scientific America in 40years.


 

syn08

Senior Member
Joined
Aug 9, 2019
Messages
376
Likes
461
Location
Toronto, Canada
Right. Regarding the paper quoted by the OP, I gave up reading after I got this:

"Audiophiles sometimes view cables as “tone controls”, thinking that their effect on timbre is the result of fine changes in frequency response. This notion is wrong. First of all, at the level of HEA, time-domain effects rather than spectral alterations are more influential on sound quality [6]–[14]"

Which is totally hogwash and balderdash. I could not believe my eyes: a professor with no grasp of Fourier transform, maintaining that frequency domain and time domain are independent. And quoting in the references 3 times van Maanen (a well known lunatic, that got what he deserves even in the now defunct Linear Audio (not a peer reviewed journal)), continuing with papers irrelevant to his statement, and ending with the Reiss infamous "A Meta-Analysis of High Resolution Audio Perceptual Evaluation".

P.S. HEA = High End Audio. Of course, HEA is driven by an OPP (Own Private Physics)
 

Cbdb2

Major Contributor
Joined
Sep 8, 2019
Messages
1,530
Likes
1,485
Location
Vancouver
From his conclusions: ". This work shows that two system configurations differing ONLY by the interconnect pathway are audibly discernable, even by average listeners".

So he doesn't consider the change from balanced out/in electronic circuits to unbalanced between the 2 tests as a difference in system configuration? Talk about amature science. His work shows nothing.
 
Top Bottom