This is like religion, everyone has got their own version. Some belive in god and others in science. But is there a thruth out there somewhere🙄
if we consider the physical reality of this world as some kind of believe...
Regards
Religion and science can coexist.
After all, some believe in God and double blind tests.🙂
Here? Unpossible.
You're talking cables, I was still on boconner's reasoning. The notion that the limit to scientific investigation is set by manufacturer's claims (as one example) has more to do with the social dynamics of online forums than science. Your statement, by sticking to the subtly of auditory effects, though arguably a value judgment, is far more coherent. My complaint is about the damage being done here to science by some well-meaning defenders.
Not sure I recall. 🙂
It's not the co-opting of knowledge, it's ignoring the process by which it sometimes arises. In 2 acts:
"There is no evidence, no science and no possibility that functional cables affect sound. You're deluded."
"But Polk cables with Naim sounds bad..."
Smoke, fire, warranty repairs leading to....
"There is no evidence, no science and no possibility that functional cables connected to well designed electronics affect sound except in the rare case of incompatibilities resulting in instability. You're deluded."
"But these Teflon interconnects are audible."
"There is no evidence, no science and no possibility that functional cables connected to well designed electronics affect sound except in the rare case of incompatibilities resulting in instability or dielectrics prone to triboelectric issues. You're deluded."
And on and on. It doesn't matter is the report is proven or not, deluded remains the judgment.
I'm not saying all reports of audible difference are credible. Far from it. That doesn't absolve however the odd transformation to knowledge that occurs when reports are proven credible.Those claiming audible cable differences remain uniformly hallucinating and intellectually inferior. Not my perspective, but read the posts here. Whether the reports are proven factual or not has no bearing on how the reporter is judged.
Thirty+ years ago, when my interests in all this started, the notions that cables could blow up amps or Teflon was audible was solidly in the domain of the demented. The dividing lines have shifted but the perspectives haven't.
I don't know if it is clear, BUT the NAIM amps needed zip cord like wire in order to add the inductor that it lacked. In those days, it was almost impossible to NOT use a coil, but Julian just insisted that the speaker wire was an inductor. Is that clear to everyone? Polk cable was really, really difficult to drive, with any amp and would still be difficult, today.
Thanks for your reply. Acts 1-2; To be fair I think the, "you're deluded" should be replaced by, "there is a real possibility that you're deluded". That aside you are right, but I don't have a problem with it.
Dangerous thought:
In the light of psychoacoustics you can't assume that the later measured differences were actually heard by the reporter. With so many people trying so many things and such a low hit rate, pure chance is a possibility.
But to summarise your point could I say that, subjectivists have been unfairly treated?
If so I disagree. Their track record is not good if you are impartial and include all their claims.
If you just select their sucessfull claims it is a different story altogether and I feel a great sympathy for them. 🙂
To expand on what I said earlier. Test results should line up with the claims made. So, if the claims are that a cable “maybe sounds better, maybe not” and the test results are similar, then everything lines up, all is good. Similarly, if the claims are that a cable “shows clear improvement in sound quality that is immediately evident” then the results should be the same and equally unambiguous. So that all, or a high proportion, of the listeners should easily be able to hear the superior results from the cable being tested. Again all is good, things line up.
But if the claims are for unambiguous improvement, and the results are “maybe yes, maybe no”, then the results don’t support the claims and the claims have not been confirmed.
This is, of course, completely separate from any investigations scientists and engineers may be pursuing with cables or indeed any other equipment. They can delve as deep as they like (to the quantum level if it helps) and explore subtle, low level measurements and what not.
But that is different from checking claims made about cables by manufacturers and the retail chain. Those claims are in the public domain, are potentially fraudulent, are made in a way that can be tested, and quite rightly should be tested vigorously, just in case they are false.
Five Contributors to Audio Interconnect Distortions
In my looking at audio cables over the past few years using special equipment I believe there are five areas of actual distortion as opposed to RLC effects.
The first and best known is the formation of oxides on the connectors. Oxide formations can in extreme cases require a potential voltage to reach a level of -30dbv before the barrier permits current to flow. Once current starts flowing the barrier voltage usually drops and lower levels can pass. When the current flow stops the oxide "heals" and the problem recurs. When I did my first experiments on cables I found that I had to use Caig De-oxit on all of the connectors to observe any of the lower level effects.
The second contributor is in some cases electroplated metals. In circuit board applications I have been able to measure performance differences in the through plated holes and the base copper. I have not been able to pin a precise value on precious metal plated contacts, but indications are thin plating looks nice and may reduce some oxide formation, but the overall quality of the connector is that of the base metal. There may be special plating processes to improve this, but I have not studied them.
The third contributor to distortion is the "work hardening" of the metal during the drawing process. It is well documented that annealing (Heating in an inert atmosphere and slow cooling) not only softens the metal but also in typical copper wire decreases the resistance by about 2%. Hard copper wire shows distortion around -140dbv for a 1M piece. This may be due to the change in crystal structure during the hardening. If you have access to a scanning electron microscope you may wish to take a piece of hard copper wire, cut it in half, run "conditioning currents" through one half and then compare them. You just might observe that the micro-fracture structure really does change between the two samples, at least for a while. The fracture structure of course changes every time you flex the cable!
Annealed copper wire shows similar distortions but 10 to 20db lower.
The fourth issue is of course the insulation surrounding the conductors. The effects of dielectric on capacitors is well documented and to a lesser extent this exists in many audio cables. There are of course methods to almost completely negate this problem. AX readers will of course be familiar with some of them.
The final method in my manuscript of the Audioxpress article on how to measure cable distortion began "Whack me up the side of my head with a two by four.." The editor removed this, as although I think audio should be fun, they prefer to present it as serious. This observed result was that there is a component of distortion that is influenced by the direction of power flow. Turns out there are actual artifacts that are a result of the manufacturing method! This effect is around -150dbv.
A valid question is if the distortion levels are so low, can they be heard? Note that some distortions are affected by current levels and others by voltages. The level is referenced to 1 volt and some distortions do not reduce with signal level.
In the world of CD music sources, the maximum CD player level is +6dbv. If we only allow 20db for headroom, set the passive volume control at half scale (another -22db typical) then our signal level is about -36dbv. If we have all of the interconnect problems we may have distortion at -138dbv or only 102db of signal above distortion at typical listening volume. Soft passages may have less than 60db of signal to distortion!
Of course if you use a higher signal level, a reasonable load impedance (10K typical) then the distortions can drop to -160 to -180db.
If you wish to try an AB test between a hard drawn copper cable and a dead soft silver one you still may not be able to hear a difference. In one of my early experiments I found that the fit of the RCA connectors was not well matched and this caused some A-B and B-A differences, so I tried a reversing switch instead. The switch was 10 to 15db worse than the cable under test!
In conclusion if you are building your own gear, don't use connectors when they are not mandatory, deoxit everything, use reasonable impedances and levels. If you are manufacturing gear for sale for tweaks' use, you may wish to use conditioned silver wire and other bits of almost magic, because you do not have control over the end user conditions.
In my looking at audio cables over the past few years using special equipment I believe there are five areas of actual distortion as opposed to RLC effects.
The first and best known is the formation of oxides on the connectors. Oxide formations can in extreme cases require a potential voltage to reach a level of -30dbv before the barrier permits current to flow. Once current starts flowing the barrier voltage usually drops and lower levels can pass. When the current flow stops the oxide "heals" and the problem recurs. When I did my first experiments on cables I found that I had to use Caig De-oxit on all of the connectors to observe any of the lower level effects.
The second contributor is in some cases electroplated metals. In circuit board applications I have been able to measure performance differences in the through plated holes and the base copper. I have not been able to pin a precise value on precious metal plated contacts, but indications are thin plating looks nice and may reduce some oxide formation, but the overall quality of the connector is that of the base metal. There may be special plating processes to improve this, but I have not studied them.
The third contributor to distortion is the "work hardening" of the metal during the drawing process. It is well documented that annealing (Heating in an inert atmosphere and slow cooling) not only softens the metal but also in typical copper wire decreases the resistance by about 2%. Hard copper wire shows distortion around -140dbv for a 1M piece. This may be due to the change in crystal structure during the hardening. If you have access to a scanning electron microscope you may wish to take a piece of hard copper wire, cut it in half, run "conditioning currents" through one half and then compare them. You just might observe that the micro-fracture structure really does change between the two samples, at least for a while. The fracture structure of course changes every time you flex the cable!
Annealed copper wire shows similar distortions but 10 to 20db lower.
The fourth issue is of course the insulation surrounding the conductors. The effects of dielectric on capacitors is well documented and to a lesser extent this exists in many audio cables. There are of course methods to almost completely negate this problem. AX readers will of course be familiar with some of them.
The final method in my manuscript of the Audioxpress article on how to measure cable distortion began "Whack me up the side of my head with a two by four.." The editor removed this, as although I think audio should be fun, they prefer to present it as serious. This observed result was that there is a component of distortion that is influenced by the direction of power flow. Turns out there are actual artifacts that are a result of the manufacturing method! This effect is around -150dbv.
A valid question is if the distortion levels are so low, can they be heard? Note that some distortions are affected by current levels and others by voltages. The level is referenced to 1 volt and some distortions do not reduce with signal level.
In the world of CD music sources, the maximum CD player level is +6dbv. If we only allow 20db for headroom, set the passive volume control at half scale (another -22db typical) then our signal level is about -36dbv. If we have all of the interconnect problems we may have distortion at -138dbv or only 102db of signal above distortion at typical listening volume. Soft passages may have less than 60db of signal to distortion!
Of course if you use a higher signal level, a reasonable load impedance (10K typical) then the distortions can drop to -160 to -180db.
If you wish to try an AB test between a hard drawn copper cable and a dead soft silver one you still may not be able to hear a difference. In one of my early experiments I found that the fit of the RCA connectors was not well matched and this caused some A-B and B-A differences, so I tried a reversing switch instead. The switch was 10 to 15db worse than the cable under test!
In conclusion if you are building your own gear, don't use connectors when they are not mandatory, deoxit everything, use reasonable impedances and levels. If you are manufacturing gear for sale for tweaks' use, you may wish to use conditioned silver wire and other bits of almost magic, because you do not have control over the end user conditions.
Last edited:
Why? What does the statistical accuracy of claimants in describing degree of effect have to do with the existence of that effect? Those are two separate issues. Isn't it approaching a scientific investigation like a crown attorney? What are you going to do should claims of subtle differences in cables be one day proven correct and 'momentous changes' not? Claim cables are inaudible because 'they' overestimated the degree?
Which ties back to fredex's question. Most likely what will happen is claims that are ridiculed right here and now will be treated as prosaic common knowledge and the claimants still dismissed without missing a beat. Simon7000's appearance couldn't be more timely. To dismiss even the possibility of accurate observations from 'them' on a statistical basis when some claims have arguably (?) lead to the discovery new flaws and distortion mechanisms I find to be inconsistent and contradictory. From a scientific perspective anyway, from a social one it makes perfect sense.
Simon, people like you and me, who have looked into this, find MANY potential sources of distortion, especially at low working levels. Often, when people measure a connecting cable, they put an extra amount of voltage, above and beyond, the actual working voltage. This appears to make certain distortions disappear. This is what we call, 'science'.
Well, some things (your word "effects") may be measurable by instruments and also inaudible to people. I said that some posts ago. Just because they are inaudible with proper audibility tests does not make them less real. After all, we can’t see unaided heaps of things that we can see with microscopes and other tools, but they are no less real for that.
But you misunderstand the need for audible tests when the claims being made are for audible differences. If the claims were for “noise levels less than -100db” or any other quantitative figure then we would just use the appropriate measuring tool and confirm if the noise floor was where the claim said it would be. Easy.
The public claims are not like that. They are claims that cable A “sounds better” than the cheaper cable. Now, “sounds better” is of course a subjective term. Putting aside aesthetic judgements, it must mean, at the very least, that there is a audible difference between the expensive and the cheap cable. No difference, then no possibility of even beginning to make subjective judgements of “sounds better”. If they can’t be distinguished then one can’t be better than the other.
Insisting that public claims are confirmed (or not) by tests is not going against science. Nor is it denying that knowledge grows over time.
And the testing of claims with protocols that seek to remove bias and wishful thinking is the hallmark of good science. Check some science journals, statistical analysis in “real” science is everywhere.
And vica versa... In Kunchur's papers on audible time-delay, it was clear that he had to go to great lengths to with his gear to get the results he measured... in many cases, i suspect, to the same degree as Simon7000 did with his measures. Kurchur is clear that he feels that his measures may well be limited by his kit, and that human detection thresholds could be even lower than he established (5 mS)
dave
I hope that everyone learns that just everything that they learned in a textbook, usually a number of years ago, is the ONLY technical knowledge that is possible or useful.
Before one cable can sound better it must sound different. Tying it to cost in defiance of so many posts stating otherwise here is telling. What started all this was your assertion that the only valid claims for testing were those of very obvious differences. If the test is designed to determine if claims of immediately apparent differences are grounded in fact, then no argument, though the actual test protocol becomes a question. Background noise? Short time limits on auditions? How will subtle ones be excluded?
If the test is to determine if, as SY calls them, non-pathological cables can sound different at all then limiting it to a subset of claims is an artificial restraint that corrupts the protocol. I'm more interested in whether cable audibility can be proven than confirming Stereophile ads.
You're also in this odd, if logically consistent, place where the hypothetical valid discoveries from investigations originating in reports of audible effects become divorced from and lend no weight to the claims originating the investigation. Rephrased: "You claimed to hear a difference. We investigated and found something that may be correlated and will be addressed. It doesn't prove you heard anything." One way that's possible is if significant issues can often be found just by investigating areas at random. However it's just as logically consistent that the testing protocols to prove that audibility are lacking, or loaded, or have never been rigorously applied. A bit of a dead end.
Last edited:
- Status
- Not open for further replies.