The cable conundrum
- Thread starter DaveyF
- Start date
Yes - his laziness consists not in not working hard - certainly he did that. He didn't however work smart, so as I said its intellectual laziness (rather than physical).
What tests was I talking about? As far as I recall I've only talked about tests that he omitted to do - common-mode ones. I've not done those with measurement kit, only with my ears. That's how I know that common-mode effects make a difference to the sound. Pete Goudreau found the same in the link I posted up earlier.
Your question presumes that I want tests done. I have no such want, I'm just observing and remarking on the tests others have made. If I were to do tests I'd do my best to ensure that such tests arose out of listening results because that's how science works - observation, hypothesis, experiment. I was also taught this paradigm for teaching English (OHE) so its served me pretty well so far.
The kinds of tests that interest me probably need different kit - which would need to be designed and characterized. I note Mr Cooper didn't direct himself towards developing tests which correlate with the salesmen's observations. More evidence that he was lazy in just sticking with established tests like THD which have scant evidence (if any) of correlation with sound quality.
They didn't use any kind of simulated load so how could they calculate the effect on FR without knowing the load impedance?
Overall from the details so far given this does not look like particularly thorough work to me, and that's a typically British understatement
Right or wrong, the way they approach this is that the cable IS an electrical component, and they are simply measuring such an LCR network, all by itself. A load would add another dimension: In general, MIT views audio cable as a series inductor in parallel with a capacitor, as shown
Given that, they simply calculate voltage/current phase relationships at various frequencies, and presumably provide corrective networks and winding techniques to approach as optimal a relationship as possible.
@Don, thanks for reading the entire paper... still thining about your comments...
No quibbles with the cable being an electrical component, but the picture you show is a component with 4 terminals - top left, top right, bottom left and bottom right.
So if that's what they show, why are there four electrical connection points on the diagram? A series L with a C would have just two.
Obviously they measured lumped parameters and extrapolated from there, although I think the paper also treats everything as lumped rather than distributed. I agree a network analyzer would have been a better choice. I wonder what a TDR would show...
Regarding the earlier comment about 0.1 dB flat frequency resonse -- what was the driving source and load? That aside, it is possible to design an LC network that would have little loss and only change phase angles, so perhaps that is what they are doing (in addition to the special wrapping technique).
@ack: At this point I am merely ruminating, basically babbling out loud (or in writing) out of curiousity. If I had one of the cables in hand, I would hook it up to a network analyzer and get the transfer function, comparing to a zip cord or something simiular, and look to see what was different. A vector network analyzer measures the response through the cable, amplitude and phase over frequency, and so would provide an electrical description of what the cable does to a signal.
Regarding the earlier comment about 0.1 dB flat frequency resonse -- what was the driving source and load? That aside, it is possible to design an LC network that would have little loss and only change phase angles, so perhaps that is what they are doing (in addition to the special wrapping technique).
@ack: At this point I am merely ruminating, basically babbling out loud (or in writing) out of curiousity. If I had one of the cables in hand, I would hook it up to a network analyzer and get the transfer function, comparing to a zip cord or something simiular, and look to see what was different. A vector network analyzer measures the response through the cable, amplitude and phase over frequency, and so would provide an electrical description of what the cable does to a signal.
Testing a cable without a defined source and a defined load is meaningless in the context of an audio circuit.
In my view there are really only 2 types of cables. Ones that do the job properly with minimal noise/degradation. And ones that do a poor job eg. unshielded interconnects, poorly terminated, low quality connectors etc.
To Assume/believe/desire that a cable should add or subtract ANYTHING is to miss the point! Its a fairly simple matter of good or bad - cable companies have tried (successfully) to 'educate' us otherwise. Are there differences? - absolutely no question! Should you pay thousands for a cable simply doing its job well? NO. Cat5/6/7 do their job very well for less than a buck a meter!
To Assume/believe/desire that a cable should add or subtract ANYTHING is to miss the point! Its a fairly simple matter of good or bad - cable companies have tried (successfully) to 'educate' us otherwise. Are there differences? - absolutely no question! Should you pay thousands for a cable simply doing its job well? NO. Cat5/6/7 do their job very well for less than a buck a meter!
I guess that depends on what you define as "doing their job properly".
While I would agree that all cables that are poorly made and are of low quality will most likely not sound good,however, the differences in cables that are well made can be profound. IMHO, IF you have a highly resolving system, these differences become more profound as you dial in the cable.
I guess that depends on what you define as "doing their job properly".
While I would agree that all cables that are poorly made and are of low quality will most likely not sound good,however, the differences in cables that are well made can be profound. IMHO, IF you have a highly resolving system, these differences become more profound as you dial in the cable.
I do have a highly resolving system - within a purpose built room with proper inbuilt (as well ad add on) room treatments (even my carpet underlay has an acoustic rating)! Acoustic rated fitouts is part of what my company does, so I do know a little bit about acoustics and room treatments. I stick to what ive stated - well made cables should NEVER 'add' anything and should degrade in as small a way as is possible for copper or in my case silver. Mate, don't get caught up in all the 'if its more expensive it must be better' BS
I do have a highly resolving system - within a purpose built room with proper inbuilt (as well ad add on) room treatments (even my carpet underlay has an acoustic rating)! Acoustic rated fitouts is part of what my company does, so I do know a little bit about acoustics and room treatments. I stick to what ive stated - well made cables should NEVER 'add' anything and should degrade in as small a way as is possible for copper or in my case silver. Mate, don't get caught up in all the 'if its more expensive it must be better' BS
I agree. I am not opposed to spending money on the hobby and many of my non-audiphile friends believe we audiophiles have gone far overboard with equipment purchases and diminishing returns as we supposedly go up the ladder, but I really do find there is something very wrong when a power cable costs $10K (Crystal Power) or $17K (PranaWire). I can go to a West Marine store and browse their line (rope) display, and see the variations in fiber patterns in the ropes which is analogous to the conductor patterns in supposedly high end cables and interconnects and no where is there such a discrepancy in cost. Fancy twist patterns in wire are actually quite simple to produce and there just doesn't appear to be any correlation between cost and quality, or especially claims made and actual performance.
I agree with your observation however from an economics perspective, there's nothing 'wrong' here. its a bubble, they happen and they always burst. Think 'black tulips' for a not very contemporary example
I do have a highly resolving system - within a purpose built room with proper inbuilt (as well ad add on) room treatments (even my carpet underlay has an acoustic rating)! Acoustic rated fitouts is part of what my company does, so I do know a little bit about acoustics and room treatments. I stick to what ive stated - well made cables should NEVER 'add' anything and should degrade in as small a way as is possible for copper or in my case silver. Mate, don't get caught up in all the 'if its more expensive it must be better' BS
I couldn't agree with you more. The "If it is more expensive it must be better" is BS. OTOH, to say all well built cable sounds the same simply isn't my experience either. Is it yours?
More than any other device all cables should sound the same. That is to say they should have no sound. Why do we hold cables to a different standard? Because it is the job of every device to alter the signal albeit in a specific way. Transducers-convert. Amplifiers-magnify,etc.
The designer wold have two primary goal;. to counter the cables natural obstructions to pass the signal unaltered and to reject outside contamination. Simple in the design difficult in the execution.
Shouldn't it be?
The designer wold have two primary goal;. to counter the cables natural obstructions to pass the signal unaltered and to reject outside contamination. Simple in the design difficult in the execution.
Last edited:
my personnel experience is YES there are differences even with well made cables - but I have found the differences to be very subtle...certainly not '50 x the price' kind of differences/improvement. To be fair though, my ears are very well used/abused 45yr models - so I'm not entirely sure my findings/impressions are that reliable. I recall reading somewhere our hearing begins degrading around 21/22? if my also fading memory serves me
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |