Things that can't be measured

[Koei]

I just learned that a large group of people believe cables transfer more than electrical signals (excl. optical cables) and these cause the transducers in headphones to act differently. I've heard the same for interconnections. These changes apparently can easily be heard, but cannot be seen in the frequency response of either the headphone or DAC output. Can somebody point me to the science behind that and how we would go about measuring these effects?

 

[71 dB]

I'm afraid people who believe cables transfer more than electrical signals aren't very scientific to begin with.
If anything audio cables transfer placebo particles invented by snakeoil sellers.
Also, expensive audio cable are known to transfer a lot of money from buyer's wallet to the bank accounts of snakeoil sellers.

The real frequency response is the Fourier transformation of the system's impulse response. It is complex-valued. It has amplitude and phase. Plotted frequency responses are almost always only the amplitude while the phase is omitted. This loses a lot of properties of the system. So, there are lots of stuff that we can't see in an "amplitude only" absolute value frequency response plot, but that doesn't mean we can't measure those. Impulse responses are theoretically full descriptors of a linear system. Complex-valued Fourier transformations simply take impulse responses from time-space to frequency-space (it is kind of like watching an object say a car from 90° different angle: You see different things, but it is the same damn car!) so no information about the system is lost.

There are four known forces of nature:

The strong force (keeps atom nuclei together)
The weak force (causes radiation)
Electromagnetism (electricity, magnets, light...)
Gravitation (makes apples fall into your head when you sit under an apple tree)

Recently scientists have gotten results while studying muon particles (similar to electrons in other ways but about 200 times more massive) that indicate there might be a fifth force we don't know about because muons don't obey completely so called standard model which otherwise has made astonishly accurate predictions. However, I doubt it people scientifically illiterate enough to buying into the nonsense told by snakeoil sellers would know more about this possible fifth force of nature than top scientists in the World...

 

[roskodan]

The brain likes to hear all kind of things.

The main argument against cables as a way to improve sound is that, theoretically, the measured difference would fall way short of being perceptible by human physiology.

But add in the psychological factor and... yes, the differences in sound between cables can be huge.

Once people get more accessible and precise EQ options, they will quickly learn to appreciate the possibility to actually be able to get any kind of sound, or change in sound, they want, whenever, wherever.

It's already happening.

 

[Northern]

Some people claim they can clearly hear a difference, and I believe them. Reflective autosuggestion is real, and it works.

 

[VNandor]

These changes apparently can easily be heard, but cannot be seen in the frequency response of either the headphone or DAC output. Can somebody point me to the science behind that and how we would go about measuring these effects?

As an example, nonlinear distortions won't readily show up in just the frequency response. The math behind nonlinear systems are far more complex than linear systems so often times, nonlinear systems (every physical system is nonlinear to some extent) are considered and analyzed as if they were linear for the sake of simplicity. This won't lead to any problems if the system isn't "too" nonlinear for its purpose.
Here's a paper that explains how to measure both the linear and nonlinear behaviour of a system (in this case speakers). The paper came out at 2000 but as far as I know a lot of measurements rely on the same principle the paper discusses.

Plotted frequency responses are almost always only the amplitude while the phase is omitted. This loses a lot of properties of the system. So, there are lots of stuff that we can't see in an "amplitude only" absolute value frequency response plot, but that doesn't mean we can't measure those.

There's a certain subset of systems called minimum-phase LTI systems. These systems can be completely described by just the frequency response because the phase response could be simply calculated from the frequency response. A lot of dacs, amps, probably all the cables are minimum phase, even headphones are minimum phase up to a couple hundred hertz so with some caveats, these could be pretty much summed up by just their frequency response alone.

 

[Koei]

As an example, nonlinear distortions won't readily show up in just the frequency response. The math behind nonlinear systems are far more complex than linear systems so often times, nonlinear systems (every physical system is nonlinear to some extent) are considered and analyzed as if they were linear for the sake of simplicity. This won't lead to any problems if the system isn't "too" nonlinear for its purpose.
Here's a paper that explains how to measure both the linear and nonlinear behaviour of a system (in this case speakers). The paper came out at 2000 but as far as I know a lot of measurements rely on the same principle the paper discusses.

There were also no differences in THD. Maybe IMD was different? Highly unlikely, but who knows. I think it's probably just bias and autosuggestion at work, but its very hard for people to admit their are fallible.

 

[bigshot]

I just learned that a large group of people believe cables transfer more than electrical signals (excl. optical cables) and these cause the transducers in headphones to act differently. I've heard the same for interconnections. These changes apparently can easily be heard, but cannot be seen in the frequency response of either the headphone or DAC output.

For some reason these easily heard differences vanish with blind testing. Distortion is easily audible with null tests.

 

[VNandor]

There were also no differences in THD. Maybe IMD was different? Highly unlikely, but who knows. I think it's probably just bias and autosuggestion at work, but its very hard for people to admit their are fallible.

Where? Is this yet an other cable thread in disguise?

 

[Claypole]

It's funny how these things that can't be measured result in changes to sound quality that are 'night and day'...

 

[Roland P]

I learned about electronics at school (got into computers afterwards...). Every piece of electronics has been designed by measuring things. You can't design something that can't be measured. It's weird to think that audio cables are the only single piece in electronics that has properties that can't be measured, but do affect the end result.
Of course, you could put a GHz signal in the cable, and it will be gone on the other side, but that's not within the audible range.

 

[Northern]

I learned about electronics at school (got into computers afterwards...). Every piece of electronics has been designed by measuring things. You can't design something that can't be measured. It's weird to think that audio cables are the only single piece in electronics that has properties that can't be measured, but do affect the end result.
Of course, you could put a GHz signal in the cable, and it will be gone on the other side, but that's not within the audible range.

You're forgetting audiophile fuses.

 

[AudioThief]

I just learned that a large group of people believe cables transfer more than electrical signals (excl. optical cables) and these cause the transducers in headphones to act differently. I've heard the same for interconnections. These changes apparently can easily be heard, but cannot be seen in the frequency response of either the headphone or DAC output. Can somebody point me to the science behind that and how we would go about measuring these effects?

Science can't tell you everything buddy! We don't know everything about audio and how it works - its all just magic to a degree - at the end of the day, trust your ears! I've experienced significant gains in my chain by upgrading cables to full nordost and audioquest (I make sure to keep the nordost cables on the analog signals and audioquest on the strictly digital signals to avoid any infetterence). I've also made sure to burn all my gear in - how can science explain that? Hah, thats what I thought. Look guys, we just don't understand everything yet. What happened before the big bang? We just don't know! Same with audio. Just trust your ears!

 

[vessel]

I just learned that a large group of people believe cables transfer more than electrical signals (excl. optical cables) and these cause the transducers in headphones to act differently. I've heard the same for interconnections. These changes apparently can easily be heard, but cannot be seen in the frequency response of either the headphone or DAC output. Can somebody point me to the science behind that and how we would go about measuring these effects?

I think you need to measure the brains response to the cables rather than the actual signals transmitted through the cable. We experience subjective reality as mostly linear but underneath that there's a tiny bit of lag where our brains are doing all sorts of nonlinear things with incredible amounts of data and then fusing it together into a coherent narrative that feels like it's happening "now". The signal in the cable is hopefully the most important factor but the reality of our brains is that other factors are also involved and our experience of the signal can be changed by things other than the signal.

 

[castleofargh]

Science can't tell you everything buddy! We don't know everything about audio and how it works - its all just magic to a degree - at the end of the day, trust your ears! I've experienced significant gains in my chain by upgrading cables to full nordost and audioquest (I make sure to keep the nordost cables on the analog signals and audioquest on the strictly digital signals to avoid any infetterence). I've also made sure to burn all my gear in - how can science explain that? Hah, thats what I thought. Look guys, we just don't understand everything yet. What happened before the big bang? We just don't know! Same with audio. Just trust your ears!

Science tells us to stop trying to justify our own belief, and to instead rely on the data we can trust(well controlled, well documented, repeatable). If you believe in something, the scientific method tells you to try and disprove it as a way of validation. Not to jump on the first thing that seems to agree with your belief because it's conforting.
Have you tried to measure the signal coming out of your gear before deciding that it couldn't be measured?
Have you set up a controlled experiment to demonstrate that the changes you felt were at all related to the differences in sound you could perceive, and not to how the cables look or other preconceptions you had?

If you didn't, all you really have is your more or less arbitrary belief that you hold the truth about some subject. Belief based on nothing more than trusting your guts. Maybe you're right, but remember that anybody believing in something false in the entire world since the first man, also thought that he/she was right to trust his/her guts. The strong conviction that we’re right should come from more than our personal feelings and our omnipresent desire to be right.
Science has limitations, like all things in the real world. But anytime I see someone bringing that up to justify his views that apparently contradict objective evidence, I can't help be read it as bad poor excuse.

”Just trust your ears!”, is what this subsection of the forum keeps suggesting anytime we mention blind testing. Stop trusting your eyes, your preconceptions, your desire to be right, and do trust your ears in a test that really allows it.

 

[bigshot]

Science can't tell you everything buddy! We don't know everything about audio and how it works - its all just magic to a degree - at the end of the day, trust your ears!

Psssst! Buddy! Some of the people you're talking with in the sound science forum are professional engineers. They know a hell of a lot more about sound reproduction than you do. If you have the right attitude around them, you might actually learn something yourself. Lecturing them to embrace ignorance is a waste of time.