Cable Theory

[garylkoh]

for a very stimulating thread! Perhaps, I'd missed it earlier; however, I'm curious if energy has been alluded to in modelling/simulating and characterizing signal transmission...

The origins of my development of speaker cables came from the possibility that the cable itself might be storing and releasing energy. This theoretically should be measurable - but the amount of energy stored by capacitance, inductance of the cable is not enough to make an audible difference - as Amir's simulations has adequately (to me anyway) shown. That is, unless the ear is even more sensitive to time domain than we think that we know.

 

[garylkoh]

The problem is, who really has time or more importantly the incentive and motivation to do such extensive preference testing of so many cables to find one that may more ideal than the current cable you are using if the system is already enjoyable.

Orb, in a smaller way, we had done that over the past 8 years. We have to demo with someone else's cables, electronics, accessories, etc. for CES, RMAF, and other audio shows. With a system that is enjoyable, we stay in the demo room, spinning music for ourselves even when the room is empty. With a system that we don't like, we find ourselves hanging outside the hallway chatting to one another.

We find this behavior even though in most cases, we would have months with the cables and/or electronics prior to the show to test them out to be sure that they are "compatible" with our loudspeakers, and hence we are willing to share the space and borrow that equipment.

Not sure if this is heresy, but for balanced interconnects - especially long ones, my go-to stand-by backup for years has been the Mogami Stage Microphone cable. After 2 days at the show, when our ears are too fatigued for the 3rd day, I'd surreptitiously change the long interconnect that goes to the front of the room. Then, if any visitors come in and remark on how much better it sounds just as we have to break the room down and go home, I say "Oh, the cables have finally broken in nicely." and smile.

Cheers
Gary

 

[microstrip]

Not sure if this is heresy, but for balanced interconnects - especially long ones, my go-to stand-by backup for years has been the Mogami Stage Microphone cable.

Surely this is heresey . But everyone has hidden sins. I also use long balanced cables in my system, and like to try new cables and equipment. But sometimes, when everything seems unpleasant I look for my old carbon balanced van den Hul The Second's and listen for a few days with them...

BTW, from the Frequently asked questions about cables, Version 2,
By: A.J. van den Hul

3-26 Q: Is it possible to measure sound differences and are there measuring methods to determine the sonic quality of cables ?
A: With a very good spectrum analyzer you can measure sound differences. But still I have my doubt if it’s really possible to measure what we are able to hear. But just the simple technical parameters, like resistance, capacitance and inductance as well as other things like spectral balance and time are easy to measure. Other methods working with nonrelated frequencies and tracing the intermodulation so far failed to establish predictable results. Minor audible differences
between cables are hard to measure. Partly because the circuitry of the analyzer is not made of the same audiophile quality as your cables are. What always works is the critical ear, but in many cases this produces rather subjective results.
The complex ear is much more sensitive for sound quality compared to today’s electronic analysers...

The whole document can be found at

www.vandenhul.com/userfiles/docs/Cable_FAQ.pdf

It is a master piece of hifi literature, some sense of humor is needed ...

 

[c1ferrari]

This theoretically should be measurable...not enough to make an audible difference - as Amir's simulations has adequately (to me anyway) shown.

I'm very comfortable with empirical science and, indeed, rely on it every workday However, we may not be measuring what we are hearing...entirely

 

[c1ferrari]

Not sure if this is heresy, but for balanced interconnects - especially long ones, my go-to stand-by backup for years has been the Mogami Stage Microphone cable.
Cheers
Gary

Gary,

I'm running ~25' of Mogami Gold Studio for Microphones, aka Mogami 2534 "Neglex Quad Mic Cables", between my Studer and Rex preamp. The cable/IC acquits itself marvelously in my system

 

[Orb]

I think you could be right Tom,
BTW you just reminded about the Podcast John Atkinson did and it is a blinder-great one to listen to.
It is long but well worth taking time for members here, he also mentions potential of feedback amp design-cables-speaker, although in this scenario it is more about amp and speaker than cable from what I can remember.
Also it provides good background info on JA that I think many here never knew about (including myself).
http://www.ultimateavmag.com/content/podcast-17-john-atkinson

Anyway coming back to measuring what we hear and its challenges.
IMO this is really tricky because I am not aware of a single product combining software and both test generation-measurement tools that will allow for a time varying spectral envelope plot-analysis (showing all partials-harmonics against their FR and their amplitude over every ms) of say a musical chord and the attack-decay-sustain-release aspects of the sound/tone, including for this to be generated by the product into a DAC or preamp and then measured-compared to the real output inline (important and not isolating) beween any of these: DAC-pre-power amp-speaker.
For background on what I mean about spectral variations of time look at 2.10 in the useful PDF from Bill Buxton (appreciate quite a few here already understand this but for others its a really useful document).
http://www.billbuxton.com/AudioUI02acoustics.pdf

To do this and have accurate analysis and possible modelling-simulation of various musical instrument chords at different scale-FR may require something like Matlab combined with certain other measurement tools and probes.

However, even if this does show subtle changes in the spectral plot over time, this does not provide information to tell us why.
And IMO this is the reason such testing is probably not done by companies as they want a narrative to highlight the benefits-reasoning for purchasing a product.
So companies such as Nordost must go further in their analysis to not only show changes occur but how and why, and put this as a narrative to consumers to show why it makes sense to buy their product over others.
This is then made even more of a pig of an issue that if going for a narrative, they want to isolate the product from other factors in the real world such as other manufacturers preamps-power amps-speakers-source.
And this may be the downfall on ever trying to present more than slight measurement differences for reasons already suggested.
Good news though this is not what we really need here IMO, we just need to see if the various spectral plots (need to do different low mid and high musical notes) change and by what.
The details on why if there is a change can come later.

Cheers
Orb

 

[microstrip]

Hi microstrip,
I disagree with the statement they made "The complex ear is much more sensitive for sound quality compared to today’s electronic analysers",
Tom

Tom,
I think that the sentence should need a context and cannot be taken out of the main document. The main problem of electronic analyzers is not taking data, it is our capacity to analyze it. If you measure everything you can you will gather thousands of measurements and you must synthesize them in a few meaningful parameters that completely describe your work to present to other people. This is the hard job.
I have no doubts that most of the so called voodoo of hifi could be explained scientifically if someone dedicated his time and efforts to explain it. But the few people who know about it naturally use their knowledge to develop such products, as they must earn their salaries and pay there kids college.
BTW, my reference to "sense of humor is needed" concerned the linguistic style of the whole document - not the audio content, that I find very interesting.

 

[FrantzM]

Hi

I will not get drawn in a cable thread .. Well .. I already am ... I have no doubt that we can measure differences... That has never been a problem really .. Electrical parameters will be different for different cables... That to me is a normal premise. The real crux of the matter remains: Are these reliably audible? Why is it that in most instances with knowledge removed, the audibility of the differences disappear? More so than any part of the Audio gear we have to strain to find the differences , even on measurements... What with 0.01dB differences in FR with two widely different cables??? that we know and agree tha they cannot be perceived by the Human Hearing system as presently built (we might evolve in a million years listening to cables to be more sensitive but my money is not on it , by then, implants directly stimulating our brains will be more common than today's earbuds )
I am not saying that in some instances cables differences are not audible (Digital comes to mind for one, the wrong impedance is clearly and repeatably audible ) .. No! but there exist a threshold and it is known and it can be derived from known electrical parameters beyond which those differences are no longer reliably audible, thus the term "competently built" to which some cable vendors like Blue Jeans and manufacturers like Belden, Mogami, Canare and others abide

 

[amirm]

Hey guys. Here is some new data. Prompted by Andy's suggestion that I was measuring the speaker load relative to ground rather its two pins, I went ahead and made that fix. Wouldn't you know, now there is a rather significant difference of .2db at 20KHz:

So maybe speaker cable can shape the high frequencies. I know I can't hear 20 KHz anymore but if there are enough people who do, we could encode some test tones and make that change to see if they can hear the difference.

 

[naturephoto1]

Amir,

How about using something closer to 14,000 or possibly 15,000 Hz that many of us may still be able to hear rather than 20,000 Hz.

Rich

 

[amirm]

We could do that. The difference at 15 Khz is still around 0.2db.

 

[RBFC]

The difference at 10 kHz is still 0.2 dB if I'm reading the graph correctly! And 0.15-0.17 dB at 7 kHz!

Lee

 

[amirm]

That's correct (I think it is a bit less than .2db though). Sorry about the bad colors there. Didn't catch the visibility issue with blue trace until after I post it .

What test should we do? A single tone that changes back and forth 0.2db?

 

[microstrip]

The real crux of the matter remains: Are these reliably audible?
...
I am not saying that in some instances cables differences are not audible (Digital comes to mind for one, the wrong impedance is clearly and repeatably audible )

Frantz,

I agree 100% with your point.

But as the wrong impedance only affects jitter, according to current knowledge, it should not be audible, as decent consumer or professional DAC's will reduce it to levels below audibility.

See: Detection threshold for distortions due to jitter on digital audio, Acoust. Sci. & Tech. 26, 1 (2005)
http://www.google.com/url?sa=t&sour...sg=AFQjCNFmwSwQGAtR6hlmAgVpxWYeemnvag&cad=rja

 

[amirm]

But as the wrong impedance only affects jitter, according to current knowledge, it should not be audible, as decent consumer or professional DAC's will reduce it to levels below audibility.

See: Detection threshold for distortions due to jitter on digital audio, Acoust. Sci. & Tech. 26, 1 (2005)
http://www.google.com/url?sa=t&sour...sg=AFQjCNFmwSwQGAtR6hlmAgVpxWYeemnvag&cad=rja

You have a lot of reading to do this weekend! http://www.whatsbestforum.com/showthread.php?1151-Audible-Jitter-amirm-vs-Ethan-Winer

 

[The Smokester]

...I just re-ran the simulation while probing the equiv. circuit above and it made the same tiny difference when I varied the cable characteristics per above.

Many thanks.

 

[garylkoh]

Hey guys. Here is some new data. Prompted by Andy's suggestion that I was measuring the speaker load relative to ground rather its two pins, I went ahead and made that fix. Wouldn't you know, now there is a rather significant difference of .2db at 20KHz:

Amir, is there a more significant difference in the phase as will?

I think what is important is whether these differences are reliably audibly as FrantzM said. I don't think that a 0.2dB difference in FR is reliably audible. The threshold is much closer to 0.5dB before test subjects are able to hear the difference, and this is with very close intervals.

The conundrum is WHAT is the difference that we are hearing between cables if it is not FR and phase, and does what we can easily measure (impedance, capacitance, resistance) have any bearing to what we can hear? There are cable manufacturers who claim that their cables measure better because they track a square wave better.

So.... may be we really should try to build a good enough switcher to carry our DBT of cables. I'm curious!!

 

[garylkoh]

Gary,

I'm running ~25' of Mogami Gold Studio for Microphones, aka Mogami 2534 "Neglex Quad Mic Cables", between my Studer and Rex preamp. The cable/IC acquits itself marvelously in my system

Sam,
Try Mogami Gold Stage - at the expense of a little "body", you get gobs more micro-dynamics, tonal colors and nuances.

 

[amirm]

Peak phase difference is at 7 Khz but it is only 1 degree so I don't think it is worthwhile to look at.

 

[terryj]

So.... may be we really should try to build a good enough switcher to carry our DBT of cables. I'm curious!!

Fantastic. It is rare to see this from a 'cable guy', no offense intended.

Boy, hope you guys can get a well done thing happening!