My six-year-old daughter flawlessly passed a blind test between a silver-plated wire and a copper one
Nov 30, 2023 at 9:58 AM Post #196 of 476
I will repeat the Christmas test for two reasons:

- The first because it turns out to be something fun and as someone said, it connects 3 generations around a common hobby.

- The second because I don't want to be left in doubt and there is plenty of room to improve the test and make it more rigorous.

As you can read there is no third reason called "to enlighten head-fi users and challenge established scientific criteria", neither is nor will be my intention and I hope nobody takes it this way.

P.S: Since I opened the thread I have been doing some research (not too much) on blind testing of cables and there is something I don't like at all, there is always an intermediate component to be able to quickly change sound samples. To my taste, and with my limited experience, that component completely alters and contaminates the test. You cannot compare the sound of two cables using a splitter or an internal separation system with cables of other material inside. The soldering already worries me, as to add cables of another material, lol.
As I mentioned a few posts ago, I will be happy to read the results. And, I add, the opinion of those who - very authoritatively - will also comment on them here. I hope for a reasoned opinion that does not simply say: the procedure was not respected, or vice versa. then it's true: the protocol for a barely reliable comparison between cables is not trivial. And with terror, but I want to say this even if I am shot, an instantaneous switching is not enough to grasp the greater or lesser feeling, but, as for tasting a good wine, it is necessary to savor the differences for a while. This applies to us, certainly to me. Not for the young ears of our daughters, as long as there are these objective differences. (subjectively: no doubt. The mind is fooled)
 
Nov 30, 2023 at 10:40 AM Post #198 of 476
I thought it might be worth expanding on this point, particularly for the OP and others not so well versed in the subject:
The headphone, while not being as extreme as some IEMs, has more potential for sound change with a reasonable increase in impedance than most non-portable headphones(high impedance of dead flat planar stuff). I say that because of the low, and apparently non-flat impedance curve I saw online. I don’t know anything else. So without more information about the cables and gear, I don't know why the possibility of audible change simply gets rejected?
I’m seem to be one of those rejecting “the possibility of audible change” and yet I “Liked” castle’s post, aren’t I disagreeing with castle and contradicting my “Like”? The answer is “No” and here’s why:

I‘m not actually rejecting “the possibility of audible change” due to different cables! There are various conditions under which different cables will produce an audible change, that I’ve already mentioned; If a cable is broken, if it’s the wrong cable for the job or due to user error. That all sounds very obvious of course but there are in fact some potential subtleties in the last two points that aren’t so obvious, in certain specific scenarios. First a little history: In the beginning the cable industry started due to the telegraph, relatively little was known and literally millions were wasted trying to get it right by trial and error. The first transatlantic telegraph cable cost millions to make and lay, worked intermittently for about a week then never worked again! :) Then in the 1860’s experiments with electromagnetic telephones advanced, which required a greater bandwidth than telegraph and therefore specialised cabling (Meucci), although telegraph cabling was used in early demonstrations by Alexander Graham Bell in the late 1870’s and his developments which allowed telephones to be commercially viable. In the 1880’s the Bell Telephone Company had researchers dedicated to working through Heaviside’s writings, to develop cables which avoid issues that would otherwise have seriously hampered the expansion of telephone usage; skin effect, reflections and power loss in transmission lines, etc. Jump forward to the 1920’s and the introduction of electrical recording: The recording industry adopted what had already been discovered and employed by the telecoms industry, EG. Impedance matching. The input and output of the audio equipment in the chain, including the cabling/patch cables was matched to a nominal impedance (600 Ohms, same as telecoms). However, with the introduction of tape recording and then multitrack tape recording, matching impedance was not very optimal, it was still employed by telecoms because it was great at minimising power loss in transmission lines but the audio recording industry didn’t use transmission lines, so power loss, skin effect and other issues didn’t apply or were insignificant, fidelity within the audio band was all that mattered so the paradigm of impedance matching gradually died out in favour of a low output, high input impedance paradigm. IE. The input of any audio device should have relatively high impedance and the output should have relatively low impedance. This is still the case today and the only examples of audio cables making an audible difference is when this paradigm has not been followed. For example, in some guitar rigs the output impedance of certain equipment in the chain can be high, some turntables can have high output impedance and more recently, some IEMs or a few HPs can have very low input impedance (in certain regions of the spectrum). BTW, one could argue that such very low input impedance is a defective design. But if this is the case, it can be relatively easy to use the wrong cable for the job, EG. One with say an insufficient gauge for the task, resulting in an audible difference even though the cable is listed as say a headphone cable (or guitar cable, etc.). This is effectively user error, although the headphone or other maker may not publish comprehensive enough specifications, thereby making it difficult for the user to avoid error in these rare cases when using cables other than those supplied. The general exception is speakers, which typically have somewhat low input impedance, 8 Ohms is typical and 4 or even 2 Ohms isn’t unheard of, so care sometimes has to be taken with gauge choice, especially as speaker cable lengths vary significantly and are typically much longer than interconnects or HP/IEM cables. In the vast majority of cases though, such “wrong cable for the job” or “user error” isn’t a concern, providing it’s the right basic cable type for the job it won’t make any audible difference and even a completely wrong cable for the job often won’t, the old lamp cord or coat hanger for speaker cable being obvious examples.

Going back near the top of this already overly long post. I’m actually agreeing with castleofargh, I’m not completely rejecting the possibility of a different cable causing an audible difference, it is possible, though quite unlikely compared to the numerous alternative potential causes that were not eliminated. However, if this does in fact turn out to be the case, it will be due to a significant difference in gauge or some factor other than that stated in the thread title (silver plate vs copper). The differences over that cable length caused by silver plate vs copper are so minuscule they almost certainly can’t even be resolved into sound, let alone be audible.

G
 
Nov 30, 2023 at 10:53 AM Post #199 of 476
Let’s see, there are also many variables that can affect the accuracy of the test. For instance, my daughter might feel pressure to perform well and get nervous. Whatever the case, I will publish the results that come out.
Go! I will be a prophet. Your daughter will feel what she has already felt after you comply with all the necessary procedures. I'll bet you a symbolic Martini that I will enjoy online.
 
Last edited:
Nov 30, 2023 at 10:55 AM Post #200 of 476
The first transatlantic telegraph cable cost millions to make and lay, worked intermittently for about a week then never worked again!

I googled this to find out why it stopped working after only a few weeks:

'When the Atlantic Telegraph Company made its second attempt to lay the line in the summer of 1858, it used the same cable—which had deteriorated while sitting out for nearly a year, unprotected from seasonal temperature changes.
The workers who handled the cable “noticed it was all completely gone to crap, and so they cut lots of bits out and had to splice ends together'.

'Now that the cable was laid, the Atlantic Telegraph Company’s chief engineer in Britain, Edward Orange Wildman Whitehouse, was ready to send a message through it. Believing that a high voltage was necessary to send the message successfully, Whitehouse pumped up to 2,000 volts into the cable, writes Allison Marsh, a history professor at the University of South Carolina.
This level of voltage was unnecessary, and damaged the already-damaged transatlantic cable'.
Becky Little. 2021.
 
Nov 30, 2023 at 11:00 AM Post #201 of 476
I googled this to find out why it stopped working after only a few weeks:

'When the Atlantic Telegraph Company made its second attempt to lay the line in the summer of 1858, it used the same cable—which had deteriorated while sitting out for nearly a year, unprotected from seasonal temperature changes.
The workers who handled the cable “noticed it was all completely gone to crap, and so they cut lots of bits out and had to splice ends together'.

'Now that the cable was laid, the Atlantic Telegraph Company’s chief engineer in Britain, Edward Orange Wildman Whitehouse, was ready to send a message through it. Believing that a high voltage was necessary to send the message successfully, Whitehouse pumped up to 2,000 volts into the cable, writes Allison Marsh, a history professor at the University of South Carolina.
This level of voltage was unnecessary, and damaged the already-damaged transatlantic cable'.
Becky Little. 2021.
Interesting
 
Nov 30, 2023 at 11:03 AM Post #202 of 476
The speed of switching is very important. Audio memory lasts only a second or two for similar sounds. If the time gets much longer than that, the difference has to be very large to discern. Solder and connectors aren’t something to worry about… neither are the cables, but you insist on comparing those.

You have some deeply ingrained misconceptions about how audio works. But if you take this opportunity to tighten your controls to remove that tightly held bias, you might end up learning something.

You might want to pick up another inexpensive cable and have that ready. If one of your two cables is defective, you can match the third one to the one that works properly to discern which one is defective. Cables aren’t likely to be defective in the exact same way. So if two match, those are probably the proper ones.
 
Last edited:
Nov 30, 2023 at 11:25 AM Post #203 of 476
I googled this to find out why it stopped working after only a few weeks:
TBH, I was going from memory of what I read many years ago, I don’t remember the exact details, only that it cost a fortune, hardly ever worked and died shortly after it did.
Whitehouse pumped up to 2,000 volts into the cable …
That should have provided some brain melting peak level punch, Whitehouse wasn’t an audiophile was he? :)

G
 
Last edited:
Nov 30, 2023 at 11:28 AM Post #204 of 476
The speed of switching is very important. Audio memory lasts only a second or two for similar sounds. If the time gets much longer than that, the difference has to be very large to discern. Solder and connectors aren’t something to worry about… neither are the cables, but you insist on comparing those.

You have some deeply ingrained misconceptions about how audio works. But if you take this opportunity to tighten your controls to remove that tightly held bias, you might end up learning something.

You might want to pick up another inexpensive cable and have that ready. If one of your two cables is defective, you can match the third one to the one that works properly to discern which one is defective. Cables aren’t likely to be defective in the exact same way. So if two match, those are probably the proper ones.
Real. Very right. An instant switch will be ideal for the young ears of our daughters, who will immediately perceive (youth, spring of beauty) what mine must savor with concentration, various tests back and forth, and indeed "feeling".Obviously as long as there are these objective differences. (subjectively: no doubt. The differences are there. The mind - my mind and let's even add an additional “Imo” raised to "n" to abound - is fooled)
 
Nov 30, 2023 at 11:33 AM Post #205 of 476
Forgive my question but I am no expert in this aspect, if silver and copper are two materials with different physical properties shouldn't there be changes in their result?

Is the world of sound a special world where nothing affects?

It would have an affect, but it would be like sitting in an empty football stadium, and expecting to hear a leaf land on the other end of the stadium from you.
 
Nov 30, 2023 at 11:49 AM Post #206 of 476
I think that two cables are enough, even if she hears any difference it must be so small that adding more variables could complicate everything and throw away any results.
I disagree with this assessment. If you alter your testing methodology to correctly conduct a DBT, adding a controlled experimental factor will actually reduce the chance of random choices or guessing to result in another perfect score. This would happen with a logarithmic relationship to the amount of factors included, so of course you don't have to go wild with it, but adding even just one more would very significantly improve your chances of having a statistically significant result with a relatively low iteration count.

As you designed the initial study, the probability of your daughter guessing each answer correctly would be 1.5625% given that the testing cycle was not contaminated by any uncontrolled variables. If you add another experimental factor while controlling for any possible contaminating variables in the next study and not changing the amount of iterations, the study will yield a 0.1291% chance of your daughter correctly guessing each one right.

This is assuming that you use 1 control (the stock cable your HPs came with), 1 experimental factor (your custom polish cable), and one red herring (a cable made of the same material as the custom cable but manufactured to stock specifications, ideally from the same manufacturer as your control cable). This is still not ideal given that you don't personally know how each cable was specifically manufactured, so if you wanted the proper level of control for the test, you would order a silver-plated copper cable and a pure copper cable from one manufacturer and use a cheap commercially available cable from amazon or whatever as the third factor.

Furthermore, this would add hesitancy due to uncertainty on your daughter's decision making process during the testing cycles. If there is any kind of contaminating variable that you still haven't addressed in this second test design, adding the third factor will further reduce the chances of a false positive by causing second guessing on a conclusion, meaning that a clear and present difference will make it through all of these introduced factors and serve as far more robust evidence of a difference.
 
Nov 30, 2023 at 12:00 PM Post #208 of 476
the protocol for a barely reliable comparison between cables is not trivial.
Actually the opposite is true, there’s hardly anything you can reliably test that is more trivial than cables. Unfortunately though, that requires an ABX switch box which the OP probably doesn’t own. The other fairly trivial way is to record the same signals through the different cables with a decent ADC and then use free software to ABX them. Probably not as much fun for his daughter though. The protocols only become non-trivial when doing a manual DBT, where someone is switching the cables and asking the questions.
An instant switch will be ideal for the young ears of our daughters, who will immediately perceive (youth, spring of beauty) what mine must savor with concentration, various tests back and forth, and indeed "feeling".Obviously as long as there are these objective differences. (subjectively: no doubt. The differences are there. The mind - my mind and let's even add an additional “Imo” raised to "n" to abound - is fooled)
Instant switch is ideal for everyone. Regardless of age or gender, just noticeable differences are more easily and more reliably differentiated with fast/instant switching, pretty much every ABX test since it was invented in the early 1950’s has demonstrated that.
For instance, my daughter might feel pressure to perform well and get nervous.
So don’t let her see the efforts you’ve gone through to set up the test. Hopefully she shouldn’t feel any more pressure than you just asking her normally in a sighted test. Very possibly she will get less nervous, as she’s already been through (and enjoyed) a similar procedure from the previous test. I would think boredom/loosing interest would potentially be more of a problem, especially if you want to do more tests to reduce the likelihood of random guessing resulting in a high/significant success rate.


G
 
Nov 30, 2023 at 12:04 PM Post #209 of 476
How many volts would have been sufficient?
With a headphone cable, probably 1,000 times fewer Volts! With the transatlantic cable I have no idea but presumably a great deal less than what destroyed it.

G
 
Nov 30, 2023 at 12:04 PM Post #210 of 476
I still think capturing the signals through the cables and doing a null test is the simplest and most effective way of finding out if there is an audible difference.
 

Users who are viewing this thread

Back
Top