Inter Connects - What I know and don't know

@nquery I don't see any figures in the document at the link which have the signal pins tied to 'REF'. If they were tied in such a fashion, I imagine they would be shorted. Perhaps you can list the page number on which the figure is found?

@atmasphere See figure 3 on page 7 of AES48 Draft. Maybe I was being presumptuous by assuming that the signals pins == "signal circuitry" in the diagram. The REF label is next to the black lines coming down from "signal circuitry".

Twisting cables is for blocking outside noise impacts from other nearby sources of magnetism (EM? too). Standard technique, so not sure why simplest twist would give more noise. See Twisted Pairs: Why are Wires Twisted? – Audio University (audiouniversityonline.com) for explanation. Most obvious company here is Kimber. They do seem to use more complicated twists in higher cost cables. Do not believe Cardas twists with their complicated geometry of different size conductors. But many cable companies use different shielding (foil, polyethylene, etc.) for blocking as well. 

Did read ASircom review of Odin 2 when came out. Is actually copper with silver overlay. He did actually test 100,000 British pounds of cable in a 10,000-pound system. Loved Odin 2 which seems to be the general consensus (like ghdprentice who tried them) and admitted likely used in more expensive systems. Have seen this before with the plating of silver over copper. Not due to cost, but many saying silver can be sibilant, emphasizing the highs. Some saying copper good for lower regions and silver for higher so the compromise. Not sure how that relates to the higher conductivity of silver unless an inductance game for reactance as that is proportional to frequency. So then maybe silver would enhance the highs.

Is certainly a varied game with monofilaments, Cardas the different sizes (golden ratio), Kimber (complicated twisting), anticables (naked), square conductors, oval conductors, etc. Questions about skin effects and high frequencies running there. Seems like number of manufacturers are able to use different constructions and still come up with great cables. That is why trial basis is best way to go.

Lastly, could also try Audience now acclaimed Forte f3 power cables. Considered very good for the cost. Am very happy with mine, even if considered entry level.

Do generally agree with excellent, detailed comment from williewonka. Classic movie by the way! (the original, of course!)

Also want to add that the OCC I believe is the Ohno continuous casting. This was a major advance by Japanese researcher Ohno with continuous casting in one crystal. Thus, there are not variable size crystal grains and grain boundaries to deal with for the electrical current. Also reducing the level of non-conducting elements like oxygen has been a major improvement. Annealing heat treatments likely reduces stresses (though heat treatments can impact overall composition if look at material diagrams). Perhaps that is what happens with the mysterious burn in. Some talk about moving cables even producing temporary stresses which may impact sound. I have found that raising cables off the carpeting does impact the sound, likely due to static electricity of the carpeting impacting the charged dielectric. 

Maybe I was being presumptuous by assuming that the signals pins == "signal circuitry" in the diagram.

@nquery You were. the lines from the bottom of the 'signal circuitry' box are ground connections. The signal passes from left to right in the diagram.

This document relates to how EMI circuitry should be used in balanced connections to deal with high frequency issues that may result in noise. So it shows a variety of hookups and shows how a number of them can be problematic.

In any even while the signal circuit does use ground as a reference, the signal current does not enter or use ground. For example in our preamps, which have a patented direct-coupled balanced tube output, if you ground one side or not, the signal level remains constant; IOW is the same amplitude.

I'd be very interested to see a circuit where signal currents are not present in the ground, yet when operated balanced you get double the output you do single-ended!