Kimber RCA Cable vs Amazon Basics (Video)

[daniboun]

Amir's video popped up over at Tom's Hardware.

Comparison of $4,000 boutique audio cable to $7 Amazon Basics cable shows audiophiles waste a lot of money — scientific audio equipment analysis with analyzer shows no difference in quality

'Black Pearl solid silver conductors drawn in diamond coated dies and insulated with virgin FEP dielectric' doesn't help.

www.tomshardware.com

Nice and well done to Amir! ) bravo as we say in French !
I would just modify this point: "Prestige RCA cables simply unnecessary" by removing the qualifier "RCA" and stating "Prestige cables simply unnecessary" considering the other tests available on ASR

 

[solderdude]

That’s really good you managed a deeper null than I ever expected well done , hats off.

You can get good nulls with cable testing as long as you record cable A on L and cable B on on R in the same recording. Clock drift (that is the usual culprit when comparing 2 stereo recordings) is not important here

There is no 'bias' with null testing as there is no brain.

 

[daniboun]

Actually, it's Hitchen's razor. But that sounded good too

Hitchens' razor is a translation of the Latin expression "Quod gratis assertur, gratis negatur," a formula from Roman law that has been widely used since the beginning of the 19th century.
Like Occam's razor, these epistemological tools are widely used in rationality...Spinoza and Descartes were masters in this field.

 

[Mnyb]

You can get good nulls with cable testing as long as you record cable A on L and cable B on on R in the same recording. Clock drift (that is the usual culprit when comparing 2 stereo recordings) is not important here

There is no 'bias' with null testing as there is no brain.

Thats really good , i was thinking of clock drift and random noise as likely to disturb nulling attempts .

That is interesting if you record exactly the same cable the next day with different room temperature and slightly different random noise they don't null this deep just highlights how small these diffrences really are.

 

[voodooless]

If the only rebuttal left is to check my "labels" rather than addressing why a steady-state measurement model is used to represent non-periodic musical transients, then we’ve moved from a technical discussion to a gatekeeping exercise.

There is nothing to address, because you brought exactly nothing. No cited papers, no measurements, no psychological studies to look up. You just make a bunch of assertions. This is why you were asked for your credentials. Because: why should we accept your assertions? No, it’s not about credentials, obviously. It’s about backing up your claims. So far we’ve seen nothing at all…

I for one, think we don’t even need the psychology in this debate. I think plain old biology should be sufficient to dismiss any audibility claims between these cables. Audibility thresholds are pretty well known, and have been for decades. What we see here is orders of magnitude below those.

 

[solderdude]

My question is whether the stress and cognitive load of the test state itself act as a filter that influences the outcome.

For a listening test... sure.
You can change cognitive load (as is often done in several fields of research) and stress can also be alleviated.

The thing is ... the null test does not experience stress nor cognitive load, uses music and the same gear as an audiophile would.
It says ... NO difference.

The why of interlink cables not measuring/performing as can be predicted based on simple aspects like resistance, inductance and capacitance in analog audio is simple.
It is not fast enough and the voltage changes are small (a few volts).
The dielectric does not have a 'memory' for a steady state signal (square-wave in this instance with wide audio bandwidth so no near infinite bandwidth.
When you claim it does you need to point us to the research you claim shows this. Not just coin it as a fact and then claim audio (a bunch of sine-waves) does not do it in an exact similar way. The null test by Amir (and Ethan's video) clearly show this.
I too have done null tests of cables and ran in timing issues which can be taken care of in several ways and then shows nothing but noise from the differential amplifier (unavoidable). That noise is below audible thresholds and can be listened to (made audible) by amplifying the null result.

The reason for people reporting differences between interlink cables (and even power cables) is caused by the brain of the listener.
Take that out (measurements, null tests, well performed AB(X) tests and no audible differences are found.

Another reason for people reporting differences between interlink cables can be too high (and or frequency dependent) source resistance, load resistance and cable capacitance (length).

Regarding RFI, the concern isn't necessarily about what the amplifier filters out at the input.

That actually IS an important aspects. The cable is linear (for audio signals) the following and preceding electronics are not. That's where RFI causes the problems.
(Have done many EMC tests on analog and digital systems)

It's about how the cable, acting as a reactive component, might modulate the noise floor of the driving stage via intermodulation before the signal even reaches the next set of filters.

To intermodulate it needs to have frequencies that are also present in that entire system and RF is well above those frequencies.

An analyzer loopback is a closed, optimized system, it doesn't necessarily reveal how that specific cable interacts with the output impedance of various real-world sources under a dynamic, non-periodic load.

The internal loopback is of course (just a relay and a few traces on a PCB) but loopback of a DUT is not the same situation. Amir showed no (discernible) differences between internal and external loopback for both cables and internal loopback.
The differences between the cables that DO exist (capacitance, resistance, inductance (different length) simply fall outside of the audible range (and test BW of the tests).

Of course, when the test would be repeated with an unusual high output resistance (complex or not in the audible range) and long lengths will certainly show differences.
They are caused by the cable having different capacitance loading the source. Listening to those will reveal differences.
One can claim those are differences caused by the cable and in a sense they are but solely because of a different capacitance combined with an unrealistic high output resistance not found in your typical audio gear (aside from pick-ups or some weird not buffered tube gear)

A real audio system is highly reactive. Speakers present a complex load that generates back-EMF, and crossovers cause phase angles to shift dynamically.

Yep very true ... BUT we are talking about an RCA interlink cable. Usually fed from resistive source (for audio frequencies) into a resistive load (usually with a small capacitance)
Those are resistive in the audible range.
Sure a coupling cap can change the lower frequency range.

Speakers and speaker amps are a totally different ballgame and not part of this discussion. When one changes an interlink cable in a system one does not change all that follows behind that interlink (amp output, speaker cable, speaker, listening position nor even the level).
Also acoustics do not change. Perception can change (and likely does).

 

[amirm]

Nice and well done to Amir! ) bravo as we say in French !
I would just modify this point: "Prestige RCA cables simply unnecessary" by removing the qualifier "RCA" and stating "Prestige cables simply unnecessary" considering the other tests available on ASR

Thanks. But I did not write that article! They took my video as the basis and wrote their own, summarizing my work.

 

[Beave]

When I click on the link, that user is not found.

Apparently the account was removed since I made my post. I don't know why.

 

[xlinknz]

I do.

Yes please it was really nice to hear and see you!

 

[daniboun]

Thanks. But I did not write that article! They took my video as the basis and wrote their own, summarizing my work.

I know Amir ) I was just pointing out that your review caught someone else's attention. That's great!

 

[DanielT]

Why don't you run this test on yourself without any need to publish the results. You can listen to your system however you want.

If you are a believer, you will always find far-fetched reasons why you cannot hear differences and can dismiss a test. It becomes more a question of psychology. Not the test itself but the psychology of religion around it, for example the cable faith.

 

[SSS]

I am not claiming to be an engineer. I am simply pointing out that a measurement of a component in a controlled loopback is not a complete representation of that component's behavior in a functional system.

Whether I have a degree or not doesn't change the fact that an interconnect is a bridge between two active stages. A static bench test into an optimized input is one thing, but how that cable interacts with the specific output impedance of a source and the input stage of a receiver during complex, non-periodic signals is where the "why" lives.

Checking labels or checking my schooling doesn't answer the question of why we accept a simplified measurement as the total truth of a signal path. I’m interested in the gaps in the current model, regardless of who is pointing them out.

Agreed that the measurements by Amir are not complete in the sense that various conditions and parameters were not touched. But I think that Amir wanted just show that for regular audio signals there is no difference between the two cables. Regarding non-periodic signals you could take white noise through the line and compare for coincidence between in and out. An easy test too. Regarding the influence of the dielectric isolation material there might be an effect but I have no experience nor did measurements on that topic. What I know from the literature is that there are effects when materials are used in capacitors.

 

[daniboun]

My credibility comes from the logic of the argument and the physics of the system, not a piece of paper from a university. A degree is a credential, but it isn't an absolute shield against missing a variable. In engineering and the trades, we focus on whether a model accurately predicts the behavior of a system under real-world conditions.

Out of simple curiosity, I read your posts and saw that you registered recently and that you had only interacted in the thread relating to Kimber cables.
Can you introduce yourself, please?

Are you affiliated with the Coordinated Research Center (CRC)?

 

[daniboun]

If you are a believer, you will always find far-fetched reasons why you cannot hear differences and can dismiss a test. It becomes more a question of psychology. Not the test itself but the psychology of religion around it, for example the cable faith.

Believers are the most affected by cognitive dissonance... If you have "faith in cables" (assuming that even exists lol), there's a good chance you'll feel discomfort when your deepest convictions are challenged.

 

[daniboun]

That kinda perspective could be way more interesting

I wouldn't venture into that territory... even if his ASR username could be misleading.
The logo he uses strongly resembles a stylized form of the Greek letter Phi (Φ). This symbol is often associated with philosophy, balance, or the golden ratio, depending on the context....

I think every new member should take the time to introduce themselves, which is why I asked @PhiDelicLabs to do so.

 

[jareinertson]

I wouldn't venture into that territory... even if his ASR username could be misleading.
The logo he uses strongly resembles a stylized form of the Greek letter Phi (Φ). This symbol is often associated with philosophy, balance, or the golden ratio, depending on the context....

I think every new member should take the time to introduce themselves, which is why I asked @PhiDelicLabs to do so.

...and now silence... Hmmm...

 

[diddley]

We won't hear from him again I think.

 

[Jcrow]

Lastly, regarding blind testing, I believe the industry-standard ABX protocol has a fundamental flaw in its execution, it fails to account for Belief-Bias. The flaw isn't necessarily in the ABX method itself, but in the person conducting the test. If you don't strip the bias at the door, the test cannot remove that variable on its own. A listener entering a test with a strong skeptical bias is under a specific cognitive load to satisfy their own internal logic. If the goal is to prove no difference exists, the brain is just as capable of filtering out an objective delta as a believer is of hallucinating one. Until we move toward a stealth protocol, where the subject is unaware a test is even occurring and the test-state stress is removed, we aren't measuring the equipment, we're just measuring a biased brain under pressure.

I’m interested in how we can expand these measurement models to better fit the reality of a system under complex loads, rather than just relying on what's easiest to capture on a bench. Looking forward to the technical feedback.

I don't really know about the electrical aspects of these arguments, but I do have some comments about the cognitive aspects here.

How do you know 'belief bias' has the effect that you mention? Belief-bias is generally studied in relation to a person's beliefs and how they might discount evidence that runs counter to it. But I'm not aware of it applying to sensory modalities, so would be good if you can provide some research that shows this relationship. If you ask someone to listen for a difference, I think they will generally try to hear a difference, and you can always force them to choose an option. Of course they could just answer randomnly, but I think most people would likely try, even if they have a pre-conceived notion that there is not a difference. Couldn't you also include a control group that has been screened as neutral to check if this phenomenom actually exists and account for it?

If belief-bias does in-fact diminish one's ability to objectively tell a difference between sources, you could say for a certain population with strong bias that they display a diminished ability to discern a difference. However, this would only apply to this group. A person who strongly believes that a difference between components does exist, they would not suffer from this 'belief-bias' as you defined it (someone who believes that not difference exists), and would therefore be objectively searching for differences during ABX testing. In practice, it has been this cohort where ABX testing has been most valuable, and has shown that these people cannot accurately discern between different components. By definition, the test elimnates bias from this cohort.

Finally you mention the test-state stress, which would generally be applied to diminished performance in tests where time-limits are involved. Is there any research to indicate that participants in audio ABX tests are actually under a state of stressed cognitive load? This seems to be pure assertion. In reality, I think most participants would be in a state of high-focus, not stress. So I'm skeptical that 'cognitive load' is actually relevant in this scenario, but if you have some evidence that demonstrates this it would be great if you can share it.

 

[Mnyb]

I don't really know about the electrical aspects of these arguments, but I do have some comments about the cognitive aspects here.

How do you know 'belief bias' has the effect that you mention? Belief-bias is generally studied in relation to a person's beliefs and how they might discount evidence that runs counter to it. But I'm not aware of it applying to sensory modalities, so would be good if you can provide some research that shows this relationship. If you ask someone to listen for a difference, I think they will generally try to hear a difference, and you can always force them to choose an option. Of course they could just answer randomnly, but I think most people would likely try, even if they have a pre-conceived notion that there is not a difference. Couldn't you also include a control group that has been screened as neutral to check if this phenomenom actually exists and account for it?

If belief-bias does in-fact diminish one's ability to objectively tell a difference between sources, you could say for a certain population with strong bias that they display a diminished ability to discern a difference. However, this would only apply to this group. A person who strongly believes that a difference between components does exist, they would not suffer from this 'belief-bias' as you defined it (someone who believes that not difference exists), and would therefore be objectively searching for differences during ABX testing. In practice, it has been this cohort where ABX testing has been most valuable, and has shown that these people cannot accurately discern between different components.

Finally you mention the test-state stress, which would generally be applied to diminished performance in tests where time-limits are involved. Is there any research to indicate that participants in audio ABX tests are actually under a state of stressed cognitive load? This seems to be pure assertion. In reality, I think most participants would be in a state of high-focus, not stress. So I'm skeptical that 'cognitive load' is actually relevant in this scenario, but if you have some evidence that demonstrates this it would be great if you can share it.

To counter some bias against the test , test subject should not for example be aware that the test is about cables , just simply a choice, is this different from that.
No why and how and any details disclosed to the testers.

 

[Jcrow]

Checking labels or checking my schooling doesn't answer the question of why we accept a simplified measurement as the total truth of a signal path. I’m interested in the gaps in the current model, regardless of who is pointing them out.

For medical advice, I wouldn't consult anyone who claims to have 'innovative' ideas but not formal training.

But that aside, you have not provided any evidence to back up any of your claims, so how could anyone verify any of your statements? You literally just make assertions and claim relationships exist without anything to back it up.