Bombaywalla, thanks. Your post went up as I was composing this one, the latter part of which addresses Bruce's posts above. But first:
Pc123v, have you tried both the 4 ohm and 8 ohm output taps on your amplifier? If not, I would suggest that you do that before changing cables or anything else.
Also, I would expect that tube-rolling in your amp is likely to have a much more pronounced effect on imaging, tonal balance, and most other sonic attributes than speaker cables. And likewise, of course, when it comes speaker placement, room acoustics, and placement of anything that may be located near or between the speakers (such as the system components).
Also, regarding tube rolling, I wouldn't be surprised if changing the relatively inexpensive 12AU7's the amp uses were to make as much or more difference than changing the more expensive power tubes.
That said, Audiogon member Kijanki has the same speakers as you do, although he uses a VERY different amplifier (a Rowland solid state design employing class D architecture). After trying some Audioquest cables, and finding them to produce objectionably thin sound with his particular amp/speaker combo, he settled on Acoustic Zen Satori, in a biwired shotgun configuration. He has been pleased with the added fullness and overall tonal balance it provided, although as far as I know imaging was never an issue in his application.
Also, based on many comments I have seen over the years and on my own experience with their interconnect cables, I would expect the suggestion of Mogami by Loftarasa and others to be well worth pursuing.
The rest of this post is in response to Bruce (Bifwynne). Good luck with your experiments! The results should certainly be interesting.
As you realize, 12 gauge Romex will provide adequately low resistance for most applications. The inductance of a speaker cable can also be significant, of course, and minimizing it will assume increased importance if the impedance of the speaker in the upper treble region is low (assuming that the goal is minimization of the sonic effects of the cable). IIRC your speakers are in the vicinity of 8 ohms throughout that region, not particularly low.
I don't know what the inductance of Romex is, but I suspect it is neither particularly high nor particularly low. If it were particularly high, I suspect that someone would have discovered by now that replacing it in the house wiring with something else would improve the dynamics of their amplifier.
Capacitance is usually not a major consideration when it comes to speaker cables, except in the case of a few cables (such as Goertz) which achieve ultra-low inductance at the expense of having ultra-high capacitance, which can adversely affect the performance (or even the health) of some amplifiers if the cable is used without a "Zobel network."
Other design characteristics of speaker cables, such as purity of metals, dielectric absorption,
"characteristic impedance",
"skin effect", "time alignment," etc., etc., are either:
(a)Arguably not quantitatively significant, or
(b)Not amenable to analysis or measurement that can establish a quantitative threshold separating what may be significant in some systems from what is likely to be overkill.
And as I recently said in
this thread:
1)As seems to be generally agreed by most audiophiles, cable performance is highly system dependent.
2)From a technical standpoint, it can be expected that cable performance will vary significantly depending on the technical characteristics of the components that are being connected, such as impedances. Even to the point of a comparison between two cables yielding exactly opposite results depending on what they are connecting. In past threads, such as this one, I have cited examples of situations in which exactly that can be expected to occur. (See especially both of the paragraphs in that post which begin with "one interesting example").
3)It seems to be generally agreed by most audiophiles that cable performance cannot be either fully explained or fully predicted based on generally recognized science. It follows from that, however, that the cable designers have no way to accurately predict the point of demarcation between optimization of a given cable parameter or design characteristic and what may be overkill of that parameter or design characteristic, which will accomplish nothing in most or all applications. Therefore it can be expected that what is likely to be a significant driver of the cost of many very expensive cables is overkill of some or all of their design parameters and characteristics, which will accomplish nothing in most or all applications.
The bottom line: I would certainly consider the experience with Romex recounted by Bombaywalla, as well as the technical points he cited, to be highly credible. At the same time (not inconsistent with his comments), I would emphasize that a high degree of correlation between cable price and cable performance in any given system should not be expected.
IMO.
Regards,
-- Al
