Those who own Audioquest cable...what do you think?

[Jud]

Getting back to cables, though the electrical performance of cables in isolation is quite well understood, a couple of factors that can make things complicated very quickly are the performance of cables as part of a system, and the environment. In a given system, are thinner conductors advantageous due to greater resistance to noise being conducted through ground, or are thicker conductors better for less signal loss? With regard to shielding, is a heavily shielded cable better to avoid picking up external interference, or are widely spaced conductors better to avoid crosstalk between power and/or signal and ground?

 

With regard to human comparisons, will some cables simply result in very slightly greater loudness, pretty well invalidating any attempt at a controlled comparison? Do the tests we run correspond with what makes humans think there is greater fidelity to the sound of a live performance? How much do we know and understand about what those things are? Are tests measuring simple variation of one aspect of a signal at a time adequate, or are tests of multiple aspects varying simultaneously in more complex ways more representative?

 

Think of an analogy to another sense, sight. Certainly aspects of the signal are well modeled and understood. But it was quite a long time until we knew enough to make CGI of the complex variations of light on hair, the appearance of a mountain, or the play of emotions on a human face look real.

[Jud]

The higher-level concepts like timbre are useful when discussing music and musical instruments, but DACs and amps don't care about such things.

 

Odd, then, that there are academic papers dealing with trying to adequately model such aspects of timbre as inharmonic attack transients in digital audio.

[mansr]

Odd, then, that there are academic papers dealing with trying to adequately model such aspects of timbre as inharmonic attack transients in digital audio.

 

That's useful for analysis or synthesis of music signals, such as in a lossy compression scheme. It has no bearing on reproducing a recorded waveform other than perhaps as a tool to assess the impact of measured imperfections in the reproduction.

[mansr]

Think of an analogy to another sense, sight. Certainly aspects of the signal are well modeled and understood. But it was quite a long time until we knew enough to make CGI of the complex variations of light on hair, the appearance of a mountain, or the play of emotions on a human face look real.

 

That's a flawed analogy. If you want a visual counterpart to hifi audio, you should be thinking of photography (or videography). CGI would be akin to completely synthetic sounds, and accurately mimicking a real musical instrument is indeed very hard.

[gmgraves]

There have been many experiments conducted over decades and published in peer-reviewed academic journals that shed considerable light on this. I hope to write something of interest on the topic for CA soon.

 

 

I for one, eagerly look forward to reading it.

[gmgraves]

Getting back to cables, though the electrical performance of cables in isolation is quite well understood, a couple of factors that can make things complicated very quickly are the performance of cables as part of a system, and the environment. In a given system, are thinner conductors advantageous due to greater resistance to noise being conducted through ground, or are thicker conductors better for less signal loss? With regard to shielding, is a heavily shielded cable better to avoid picking up external interference, or are widely spaced conductors better to avoid crosstalk between power and/or signal and ground?

 

With regard to human comparisons, will some cables simply result in very slightly greater loudness, pretty well invalidating any attempt at a controlled comparison? Do the tests we run correspond with what makes humans think there is greater fidelity to the sound of a live performance? How much do we know and understand about what those things are? Are tests measuring simple variation of one aspect of a signal at a time adequate, or are tests of multiple aspects varying simultaneously in more complex ways more representative?

 

Think of an analogy to another sense, sight. Certainly aspects of the signal are well modeled and understood. But it was quite a long time until we knew enough to make CGI of the complex variations of light on hair, the appearance of a mountain, or the play of emotions on a human face look real.

 

These are good questions, especially the one about perceived loudness. Conductors aren't supposed to attenuate either linearly or non-linearly (I.E. equally across the passband or in such a way that some part of the passband is attenuated while other parts are not). But it is a well known psychoacoustic phenomenon that everybody seems to always think that the louder of any two signals sounds better (all else being equal). So if one has two interconnects and one of them has higher resistance than the other, and therefore attenuates the signal somewhat with relation to the other signal, then the cable that attenuates the least will be perceived as sounding the better of the two.

 

Of course, as far as having greater fidelity to the sound of a live performance is concerned, That's a toughie. Since few of us have ever been in a position to actually be at the recording venue when the recording was made in the first place, knowing what the actual performance really sounded like is nigh impossible. Even if we were lucky enough to have been at the actual recording, human aural memory for sound characteristics is very poor. For example, I was present when when the album Jazz Samba was recorded with Stan Getz and Charlie Byrd at the All Souls Unitarian Church on 10th Street in Washington DC. When the album came out on the Verve label some months later, I bought it. I didn't remember whether the album sounded like the performances or not. The performances were familiar, and instantly recognizable as the performances I heard that day, but the specific sonic attributes? I couldn't tell. Since then I have made hundreds of live recordings and I have always been surprised by the fact that the recordings never sound as I imagine they would while I was making them. They sound good because I know what I'm doing, but they never sound as I expect that they will while making the recording. It's hard to explain the difference between the expectation and the reality. Another problem with cable sound. How does one define "better". I understand "different", but is the sound of an expensive boutique cable different better or different worse? And do you know when you buy a several hundred dollar (or more) pair of interconnects that they are going to sound better or worse when you insert them into your system?

[Jud]

That's a flawed analogy. If you want a visual counterpart to hifi audio, you should be thinking of photography (or videography). CGI would be akin to completely synthetic sounds, and accurately mimicking a real musical instrument is indeed very hard.

 

Why would that be, if we understood what creates a sense of "reality" in audio?

[jabbr]

You 'beg the question' by assuming that 'every measurement known to mankind' is exhaustive and that nothing new can possibly be identified and measured in the future. By that token, the only thing that is 'certain' is your opinion.

 

I see. So are you picking:

 

A) a new branch of as yet undiscovered physics

 

or

 

B) magic

 

as what distinguishes cables?, because I'm fairly sure I excluded most other explanations. What question is being begged? Again, I am not saying that cables don't sound different, just that differences can be explained by the physics and scientific principles we already know.

[mansr]

Why would that be, if we understood what creates a sense of "reality" in audio?

 

Understanding something doesn't necessarily make it easy. We understand how atomic fusion works, but there are still no reactors to be found. Also, I never said we fully understand "what creates a sense of reality."

[Jud]

Another problem with cable sound. How does one define "better". I understand "different", but is the sound of an expensive boutique cable different better or different worse? And do you know when you buy a several hundred dollar (or more) pair of interconnects that they are going to sound better or worse when you insert them into your system?

 

That's a tremendous issue. I'm sure many of us have known people who never seem happy with their systems after they've had them awhile, and are always looking for something new and different. I try to listen to new equipment with some idea of what sounds "real" in mind. I tend to keep my equipment a long time, as in decades (that includes many of my cables), and to be quite happy with the sound of my system over a long period. Whether that's success or self-fulfilling prophecy, it's impossible to know. Then again, as long as I'm happy I'm not sure I care.

[Jud]

Understanding something doesn't necessarily make it easy. We understand how atomic fusion works, but there are still no reactors to be found. Also, I never said we fully understand "what creates a sense of reality."

 

For both those reasons, particularly the latter one, it is difficult to be sure we are measuring the right things, or measuring them with adequate accuracy.

[Jud]

That's a flawed analogy. If you want a visual counterpart to hifi audio, you should be thinking of photography (or videography). CGI would be akin to completely synthetic sounds, and accurately mimicking a real musical instrument is indeed very hard.

 

OK, I'll take you up on that. Why is it that a caricature, cartoon, or drawing of a person can elicit from us an emotional response reminiscent of seeing the real human being it represents, while a computer-generated facsimile objectively closer in every visual detail can cause us to inwardly exclaim "There's something not right about this!"? (Akin to the feeling of seeing someone who's had plastic surgery - something subtly, irritatingly wrong....)

 

Is there any analogy to be drawn to LPs versus digital?

[mansr]

For both those reasons, particularly the latter one, it is difficult to be sure we are measuring the right things, or measuring them with adequate accuracy.

 

If two waveforms are identical within the accuracy of our best test equipment, they will sound the same. If one sounds real, so will the other, doesn't matter why.

[Jud]

That's useful for analysis or synthesis of music signals, such as in a lossy compression scheme. It has no bearing on reproducing a recorded waveform other than perhaps as a tool to assess the impact of measured imperfections in the reproduction.

 

Which is of course quite useful in itself - and perhaps also to help tell us what to measure, and to try to see whether our current measuring tools are sufficiently capable?

[mansr]

OK, I'll take you up on that. Why is it that a caricature, cartoon, or drawing of a person can elicit from us an emotional response reminiscent of seeing the real human being it represents, while a computer-generated facsimile objectively closer in every visual detail can cause us to inwardly exclaim "There's something not right about this!" (Akin to the feeling of seeing someone who's had plastic surgery - something subtly, irritatingly wrong....)

 

I don't have an answer to that, but I also don't think it matters to the topic at hand. With an audio cable, we can measure what goes in one end and what comes out the other. If there is no detectable difference, there will be no effect on the sound either.

 

Is there any analogy to be drawn to LPs versus digital?

 

I don't think so.

[esldude]

For both those reasons, particularly the latter one, it is difficult to be sure we are measuring the right things, or measuring them with adequate accuracy.

 

Waveforms we hear can be fully described by frequency, amplitude and phase. Any gear capable of measurably conforming to accurate reproduction in those terms over our range of hearing and then some can reproduce whatever is in the input signal to illicit reality. We don't need to have full understanding of how the brain does that to say this is so.

[Jud]

If two waveforms are identical within the accuracy of our best test equipment, they will sound the same. If one sounds real, so will the other, doesn't matter why.

 

Not necessarily so, if one is not testing for the right things, where "the right things" can be quite prosaic; and if neither sounds very real, what good does identical performance do?

 

For an example of a very prosaic "right thing:" What if perceived realism depends on fidelity to variation in *both* the time and frequency domains simultaneously? Do current commonly run tests on audio equipment (e.g., filters commonly used with DACs) measure for this? Is current commonly used test apparatus sufficiently capable to explore the limits of what humans can perceive with regard to such simultaneous variation? (Consider Phys. Rev. Lett. 110, 044301 (2013) - Human Time-Frequency Acuity Beats the Fourier Uncertainty Principle.)

 

Yes, we know an awful lot about the science and engineering of audio reproduction. Perhaps not nearly as much as we would like regarding human perception of same. (Though hopefully that day is coming with technological advances.) But even the commonplace quickly becomes complex in real world situations. (E.g., we can know plenty about circuit design and the behavior of cables, but have you ever had a ground hum that took an irritatingly long time to diagnose and resolve? I've had a ground hum I resolved by simply substituting a slightly older design of cable from the same manufacturer for the newer ones I'd installed.)

[mansr]

Not necessarily so, if one is not testing for the right things, where "the right things" can be quite prosaic;

 

If the amplitude vs time is identical anything else you choose to measure must also be identical.

 

and if neither sounds very real, what good does identical performance do?

 

That just means the input signal didn't sound real to begin with, and that's not the fault of the cable being tested.

[jabbr]

Not necessarily so, if one is not testing for the right things, where "the right things" can be quite prosaic; and if neither sounds very real, what good does identical performance do?

 

If the signals are the same the sound is the same.

 

 

For an example of a very prosaic "right thing:" What if perceived realism depends on fidelity to variation in *both* the time and frequency domains simultaneously?

 

 

If you have fidelity in time you necessarily have fidelity in frequency. Two sides of the same coin:

 

The Fourier transform describes the necessary relationship between time and frequency domains.

Do current commonly run tests on audio equipment (e.g., filters commonly used with DACs) measure for this? Is current commonly used test apparatus sufficiently capable to explore the limits of what humans can perceive with regard to such simultaneous variation? (Consider Phys. Rev. Lett. 110, 044301 (2013) - Human Time-Frequency Acuity Beats the Fourier Uncertainty Principle.)

 

Just because someone didn't happen to make the correct measurement hardly means it can't be done by the right person with the right equipment. You would be amazed at what measurements of signals can be done!

 

Jussi often publishes measurement plots, for example.

[jabbr]

For both those reasons, particularly the latter one, it is difficult to be sure we are measuring the right things, or measuring them with adequate accuracy.

 

Sure. "Sense of realism" might have a specific set of requirements. Entirely hypothetically it might need exactly 27bits of amplitude resolution or a noise background of -188Db at 64 kHz but no one has demonstrated one shred of evidence supporting the need for new physics.

[Speedskater]

If two reasonable interconnect cables have different loudness, then one has background noise, interference, ringing or oscillation problems. However if we are talking about 100 meter long cables then there might be small level differences.

[Ajax]

This is a slippery-slope. Most electronics engineers will scoff loudly and long at the notion that one USB cable is any different from another as long as it meets all the USB specs as put forth by the IEEE. I am of this opinion also. However, I must, in all honesty, say that while I can't see what could possibly affect a USB signal enough to change the sound, as long as the cable met specifications, I also cannot simply discount the empirical evidence of people here, who I trust are not either lying or imagining things when they say that they consistently hear a difference between one USB cable and another. I do however, reject the attempts I've read so far to explain what might be the reasons for the differences heard. They simply do not hold any water as far as I can see.

 

 

 

My point is that if you are relying on the opinions of EEs (even those from Benchmark), keep in mind that all of that textbook training in electrical theory has narrowed, considerably, their capacity for out-of-the-box thinking especially when that thinking flies in the face of all the digital quantization theory and conductor behavioral theory that they were taught in engineering school. My advice to you is that if you are interested in pursuing the matter, try different USB cable solutions for yourself and see if you hear a difference, and - and I can't stress this one too strongly - if you do hear a difference, is it a difference that you care about.

 

Hi George,

 

Thanks for your thoughtful response, which I interpret as "keep an open mind and listen for yourself".

 

However, if we are not to trust those (who have no possible monetary gain) such as the qualified and experienced engineers at manufacturers like Benchmark, then who do you trust? Many here at CA have been highly critical of the reviewers in magazines as many question how they can be objective if the magazine has a financial relationship with the manufacturer whose equipment is being reviewed. So how can we trust them?

 

I accept that the ultimate test of equipment is to listen to it. But in what combination, and in what room? You need to have a fundamental understanding of what you are doing before you can evaluate equipment .... why doesn't that sound good? - is it the recording, the software, the streamer, the DAC, insufficient amplification, the speakers, speaker positioning, the room's acoustics etc etc. IMO this is the main reason why this hobby will remain a niche. It's complicated and it requires a certain level of effort and knowledge - you need more than just a basic understanding to get great sound. It is also why we enjoy the hobby - there is no simple answer to a lot of questions because if there were it would become boring.

 

Audio is a big part of my life but it is only a part - I haven't the time to listen to every possible combination - I have to trust at some stage, whether it be the dealer or the manufacturer or the reviewer. I "listen" to all of them, as well as the comments here at CA, and narrow the field down. I enjoy that process. If you continually engage with CA you soon learn who's opinion you value, similarly the dealers and manufacturers.

 

Back to the engineers at Benchmark and the comments in your last paragraph, which to me imply that once they leave university they stop learning. What I feel many fail to understand is that the gauky engineer (whether civil, mechanical or electrical) starts of his professional life with a very theoretical base and then applies that theory to the real world. This is when the real learning starts. He gets constant feedback on what does and does not work and the good ones end up extremely practical and creative with an in-depth knowledge of how things work. They also know how to achieve an outcome cost effectively as they are usually constrained by budgets and the physical environment.

 

I started life as a site engineer and my job was to take the architects design, together with the consulting engineers detailed drawings, and try and build a building. How practical those designs were was directly proportional to the engineers experience. i.e. engineers are not just theoretical scientists, they are actively engaged in the real world and receive daily feedback which increases their knowledge and they adjust their designs accordingly.

 

There are several parts to the story - design, product development, manufacturing, QA, measurement, implementation, listening and finally feedback and the whole process starts again.

 

All the best, Ajax

[Allan F]

What question is being begged?

FYI, "begging the question" is a philosophy concept, i.e. a logical fallacy. https://en.wikipedia.org/wiki/Begging_the_question

 

The logical fallacy in your argument is the a priori assumption that every possible measurable characteristic is currently known, to the exclusion of any potential discoveries in the future. As I see it, your position can otherwise be stated as: Unless a difference heard between cables can be demonstrably measured, the difference heard cannot be real.

 

 

[esldude]

FYI, "begging the question" is a philosophy concept, i.e. a logical fallacy. https://en.wikipedia.org/wiki/Begging_the_question

 

The logical fallacy in your argument is the a priori assumption that every possible measurable characteristic is currently known, to the exclusion of any potential discoveries in the future. You are saying, unequivocally, that cables that measure the same must sound the same, and that any differences heard between such cables cannot be real.

 

 

 

With an approach like you display here, there is no question possible that would not be judged as begging. Which means there is a problem with your line of thought in this. Part of which is a misrepresentation of jabbr's views on this matter.

[jabbr]

FYI, "begging the question" is a philosophy concept, i.e. a logical fallacy. https://en.wikipedia.org/wiki/Begging_the_question

 

The logical fallacy in your argument is the a priori assumption that every possible measurable characteristic is currently known, to the exclusion of any potential discoveries in the future. You are saying, unequivocally, that cables that measure the same must sound the same, and that any differences heard between such cables cannot be real.

 

Yes I'm surprised you'd take my statement as a priori. It's not.

 

Electrodynamics is exceedingly well tested empirically. I am not overstating this point. for starters: https://en.wikipedia.org/wiki/Maxwell%27s_equations

 

What I said was:

...cables should be treated as purely mechanical entities whos function and behavior are actually incredibly well known and measurable.

 

Human hearing is not anywhere near this level.

 

If two cables behave identically in every measurement known to mankind, then will sound identically except for expectation bias and measurement error. That is for certain.

 

I am not claiming that there is any minimal audible difference, just that for a difference to be audible (aside from either measurement error or placebo effect) it must be measurable.

 

But the placebo effect is dramatically audible and so may swamp other effects. Basically those new shiny, big thick heavy and expensive cables that you need to be brought into a dark private room to feel and listen to, might sound different, and make you want those cables at home -- just throw out your everyday boring cables that have been with you for 20 years

 

Are you taking offense to "That is for certain"? I mean scientific certainty (of a very high level). I also leave (necessary) room for measurement error. In any case as far as we here are concerned:

 

Cables which measure the same, sound the same (aside from measurement error and expectation bias/placebo effect)