They are in some of the amplifiers, like the traditional designs of Electrocompaniet and some others. Depending on how well the designers have done their audio history lessons.
If someone is designing amplifier THD+N figures in mind, they may repeat the mistakes that led to TIM becoming a problem. By the way, tube amps didn’t suffer from that, while of course THD+N figures were worse. So the overall best amp may not be the one having lowest THD+N.
Ahhh, a little Google sleuthing and now I know the scoop on TIM! Or at least an introductory window into the topic. It appears to be a solved problem with most modern amps:
Learn something new every day! So I’m still looking for clarity on issues that are human-hearable but not system-measurable…
That document is someone’s compilation ignoring whole lot of detail. Now it makes it sound easy and straightforward to design an amplifier, but it is not. Sure it is easy to write such statements “it is solved problem”. To do so, you need to read whole thing Douglas Self has written. I have, I have built couple of his amplifier designs too and I still have those. I still remember clearly all the things he was writing in a long article series in Electronics&Wireless World magazine in the 90’s. I was subscribing that for couple of decades, lot of good audio articles too.
Now he advertises his amplifier design that -94 dB distortion. A bit worse than what RedBook CD can do.
For example, see how many amplifiers have the thermal compensation transistor mounted on the heatsink? Wrong place! Read your Douglas Self! It must be glued straight to the output transistor. Best option is to use for example those Thermaltrak or similar ones that integrate the sensing transistor. Read also about correct resistor and capacitor types also mentioned there. Forget about most of the SMD ones…
Fair enough! But, but… I’m still looking for clarity on issues that are human-hearable but not system-measurable?
I think the question and a potential answer to it somehow ends in a circular loop.
There’s a major aspect of science and scientific theories. Science builds theories and tries to prove or disprove such theories by observing/measuring. This process is often quite hard to carry out properly/correctly because there’s always a chance of human mistakes. And for science it’s crucial to eliminate the human factor in this process. That’s also a reason why scientific studies are often rejected due to willing or unwilling flaws in their design or carrying out.
Assuming there is a phenomenon coming from a properly created theory which (currently) was not yet observed/measured in the process of a proper study the theory is not (yet) proven.
So, what kind of answer do you expect?
And keep in mind that the argument “prove me wrong” doesn’t count in science if you or others are not able to deliver a proper prove of your own theory. And others will check your study regarding any kind of flaws.
And if you’d like to see how humans hear, then be careful with what you actually measure. An external sound source or the brain’s processing or whatever. Whenever you like to understand the human brain it always gets a bit tricky.
To me it looks like you really try hard to find a “prove” for something (currently) only a human brain can “detect”. This would be no prove at all in a scientific sense.
And maybe thinking our brain is so far away from being (fully) understood and there’s so much happening in it is also no base for any kind of proper prove. Because you’d try to explain something you don’t understand with something you don’t understand either. This wouldn’t make sense.
Here is a related video from Lachlan about how we hear.
I am not saying he is right with what he is stating in the video.
Yep, rather over-simplyfied to my understanding.
“Simply anything in the audio chain within an audio system can be perceived” is simply wrong, neglecting existing limits of our hearing.
“Cannot be measured by regular equipment” doesn’t mean it can’t be measured at all. I hope they used professional equipment in the study.
He does mention that there’s lots of processing in the brain but he doesn’t really say something about the influences other non sound related aspects actually do have on hearing.
He concentrates on describing things interpreting hearing as a more or less isolated sound processing in our brain. That doesn’t reflect reality.
If the numbers about micro and pico seconds are correct, they measured them.
“Delayed signals in cables”. He doesn’t explain the electrical and mechanical delays in speakers. You can’t just pick one single aspect from a complex chain and state that this would have a sensible impact on hearing, omitting all the rest.
Also, we do not know anything about the sponsors of the quoted studies and meta study. Would be great to know.
Finally, the video’s title “Prepare to blow your mind!” doesn’t sound too serious to me, TBH.
Really final comment 
Download and try harman’s Windows listening/hearing app “How to listen”. This really opens your mind regardong things you simply don’t hear though, they are real.
I recommend watching the vid on 1.5x speed 
The Kunchur paper is actually quite good and worth reading, though Kunchur’s other papers on cables are notably lacking in details about the cable characteristics in a manner that would help with replication and are flawed in other ways.
That being said, the video discussion has a small flaw in the presentation of frequency domain vs. time domain. Timing transients do show up as high-frequency components and any slowing of signals show up as changes to the frequency response. One can take any signal and analyze its spectrum but, yes, the measurement systems are using continuous tones. By examining the multi-tone performance of audio equipment and insuring that the equipment minimizes noise and distortion across the spectrum it therefore also minimizes the impact on timing.
The suggestion that these measurements are not sufficient is an easy empirical investigation: simply show that two audio replication systems produce different outputs when they measure identically but are fed fast transient signals.
The speaker/headphone issues with coupling to the pinna, HRTFs, and mechanical behavior of drivers represent additional challenges for measurements, but I’ve not seen any definitive evidence that measurements are not useful and telling about the performance of these systems.
As I routinely ask cable promoters, if there is good evidence for some hypothesized effect it becomes a tremendous value proposition and why don’t we see those results? Same here: if the evolving set of measurements we currently use are not sufficient it should be fairly obvious and the practice of engineering these audio systems should be improved and the measurements evolved. I see a great deal of discussion but not much evidence.
transients are recorded by an impulse response measurement… this is perfectly common and nothing new. We know what frequencies a device produces, how quickly it can produce and recover from them, how it behaves when multiple frequencies try to interact, and how it behaves when load / gain are increased.
There’s not hidden magic here, it’s all in the graphs.
(let’s emphasize here first that I’m talking about analog cables)
I personally use mostly Supra’s cables. Pricing is reasonable for me and I agree with their design principles.
But how some cable works on some system and amount of effect it has depends on the involved electronics. Source impedance vs input impedance, etc. I always design cable capacitance compensation to my cable driver output stages. And I have a separate driver stage overall.
If we talk about recording, microphone cables are rather sensitive and can also become microphonic, for example someone stepping on a microphone cable can make sound of it’s own when high gain is used.
Another place where cables can make quite a bit of difference is vinyl decks, especially with MC cartridge.
I’ve personally had some fun with hydrophone cables and preamps. There typical preamp gain can be 120 dB with about 100 Mohm impedance. Just bending a cable can create enough internal friction that the electric field changes enough to make sound of it’s own. Good thing there is that EMI/RFI and cooling are usually not a problem…
Superstition is based on the human thinking they perceive something that can’t be perceived. Not only are we humans special because we are created in the image of God, ya God looks just like me, but we believe that many things about us are much better than anything else. We can hear better than machines with exceedingly more accuracy resolution than biologically possible. You may tell me your ears are better, but I can say, based on my superior subjectivity, granted to me by being a God created human, my subjective opinion is not only better than high resolution instruments, but this concept that I hold, lets me assume I am even better than everyone else and their opinions. Who needs objective evidence? And now I have a thing to sell you.
this yes, but even further I am fascinated by the fact that our species, evidently predatory carnivores per design, find particular sequences of sounds (music or not) more pleasing than the others. meaning that mere movement of gas particles (referred to as air) in some sort of order (referred to as waves) can through the stages of detection (receptors) and recognition (brain) drive pleasure hormonal releases. how’s that?!
(Speaker wire - Wikipedia) As an electronic engineering technician I can say you can not a step on an audio cable can make a noise, unless you make the other end come loose. The wire on a MC cartridge is irreverent. I work with MRI systems, incredibly small 123 MHz analog signals in wires, Bending a cable will not create internal friction that causes a change because electricity is only effected by resistance and reactance, not by friction. In audio cables there will be less than 1% attenuation in cables that are 12 gauge and 30 feet or less on a 2 ohm speaker, unintelligible. Well over 100 feet for a 8 ohm speaker before 1% attenuation. There is not a penny of diffrence in the expensive cables and that is not a subjective opinion, that is an objective fact based on the same science that makes your sterio possible.
What should really intrigue is the capacity of your brain to delude you into thinking you are hearing something…
It’s very easy to understand how speakers will inherently image if you have detailed quality measurements of off-axis radiation from the speaker, as from a Klippel.
“We hear differences in attack that are hard to measure” – This can’t be serious. It is trivial to measure any REAL changes to the waveform to far greater degree that can be perceived by human hearing.
Once you employ blind level controlled testing, you will see how quickly our so-called extreme sensitivity drops significantly.
“although our logarithmic frequency response is difficult to replicate in a microphone” – this makes zero logical/scientific sense. What exactly do you mean by this statement?
It’s not well phrased. I was intending to refer to two things. Firstly, the lower noise floor of human hearing, 0dB by definition, compared to all but extremely expensive microphones. Secondly the unreasonable efficacy of our hearing in distinguishing between sounds from different sources.
Both these things are measured using dB, which is logarithmic to start with, so my use of that word was otiose. A microphone’s frequency response will exceed human hearing, but it will have a higher noise floor and be less able to distinguish between different sources.
I’ve been designing audio systems, including measurement systems and underwater surveillance systems for the past 31 years… So I think I know something about the topic.
Well, you are free to read on the topic in internet… Just search for something like “microphone cable microphonics”.
You can find low microphonics mic cables:
https://www.performanceaudio.com/products/mogami-w2792-superflexible-low-microphonic-mic-cable-black-by-the-foot
Or you can just read here:
Bending causes internal physical movement in the cable, in this case hydrophone is purely capacitive piezoelectric sensor and is driving 100 Mohm input impedance on preamplifier. This preamplifier can have gain of 120 dB. Cable itself has capacitance too and can be tricky to deal with since it can become sensor of it’s own, especially because the preamplifier input is so high impedance.
The key in both cases is high impedance input and high gain. Static discharge properties and “condenser microphone” properties. Where latter is emphasized by the DC on the cable, such as 48V microphone phantom power.
Quoting from there:
Triboelectric noise within medical cable assemblies and lead wires is generated when the conductors, insulation, and fillers rub against each other as the cables are flexed during movement. Keeping triboelectric noise at acceptable levels requires careful material selection, design, and processing. It is also an issue with underwater electroacoustic transducers if there are flexing motions of the cables; the mechanism is believed to involve relative motion between a dielectric and a conductor in the cable.
One hydrophone manufacturer for example specifies about their cable:
No, for example capacitance of the interconnect used to connect LP player to the RIAA stage can have notable impact depending on the cartridge-cable-RIAA stage combination.
Our hearing is also logarithmic, both in terms of intensity and pitch, so using dB is quite appropriate. It’s what I think of as nature’s way of doing dynamic compression.
A microphone is similar to a ear. It takes two ears and a brain to do whatever our auditory system does. That would translate to two (or more) microphones and some serious DSP. With the amazing progress made by AI lately, I wouldn’t be surprised if machines will soon do a better job at what is now considered specifically human.
Does it need to? Surely, all that equipment needs to do is faithfully reproduce the original sounds. If, as you say, the human ear is sensitive to certain stimuli, e.g. distress–and I agree, see below*–this will be reproduced, too. And, I think science and engineering has already demonstrated that we’ve got it all covered. There are no discoveries in audio to be found. We already surpass the human ear in science and engineering.
*If so, this BBC / CBC video from forty years ago wouldn’t trigger an (evolutionary) emotional and physiological reaction then or today. It’s a real shame we still can’t solve those problems.
Who’s gonna plug their ears
When you scream?
Excellent, so we’ve collectively established that the correct cable for the job is important and that there is little evidence that human hearing surpasses measurement capabilities in any meaningful way (pedantically: such that hearing can be uniquely used to improve on the audio reproduction chain).
But I will beg to differ on the issue of “completeness,” at least WRT 3D reproduction systems. The thesis I linked to above does include an analysis of human subjects in a 3D wavefield synthesis system that involves arrays of speakers…not that such systems are readily available (Sphere in Las Vegas?) So there are still plenty of frontiers left to explore but hyperbolic assessments of the properties of human hearing and audio cognition don’t really advance the field.
