You are missing one important piece... which is damping factor.
How the speaker cable affects the damping factor is more likely to become an issue before actual "Ohmic losses".
Let's assume I have 8 Ohm speakers, and an amplifier that has a damping factor of "500 or more".
This high damping factor tells us that the amplifier is able to control the speaker pretty well.
(This includes both delivering power to the speaker and shorting out the back EMF from the speaker to stop its motion when the signal stops.)
I'm going to round our speaker up to 10 Ohms - because I'm lazy and the math is much easier.
If my damping factor is really 500, then that means that the source impedance of the amplifier "as seen by the speaker" is 1/50 Ohm or less.
(The damping factor is simply the impedance of the speaker divided by the source impedance - its output impedance - of the amplifier.)
NOTE, however, that, for this calculation, the resistance of the speaker wire counts as part of the source impedance.
(In the proper electrical model, the speaker wire counts as "part of the amplifier", and not "part of the speaker".)
Now, to pick an extreme case, let's assume that the resistance of our really thin speaker wire is 1 Ohm.
The "Ohmic power losses" will be about 10% (the wire is about 1/10 the resistance of the speaker).
BUT, since the source impedance, as seen by the speaker, is now about 1 Ohm, the damping factor will now be approximately 10.
To repeat that.....
Adding a one Ohm speaker cable will have reduced the power that arrives at the speaker by about 10% - which is insignificant.
However, it will have reduced the damping from about 500 - which is quite high, to about 10 - which is more like what you would expect from a tube amp.
That difference will be clearly audible with many speaker/amplifier combinations.
Most people I know would more or less agree that:
- differences in damping factor between "really low" and "about ten" are EXTREMELY AUDIBLE with many speakers
- differences between about ten and a hundred are OFTEN AUDIBLE
- and numbers above about 100 can probably be simply thought of as VERY HIGH (or "high enough so as not to matter")
Also remember to double the values you use in calculations.
(A ten foot speaker cable has ten feet of wire in each direction - for a total of twenty feet.
Wire resistance ratings are always for a single wire - going one way - so double it for both directions.)
Also remember that resistance in connections, like between the wire and the banana plug, or between the plug and the terminal, are often FAR greater than the actual wire resistance.
(And they're also affected by oxidation, and loose connections, while the resistance of a continuous piece of wire is not.)
I would personally usually use 8 - 10 gauge for runs longer than twenty feet or so.
I would be comfortable with 12 - 14 gauge for short runs (under ten feet).
And I would not be too uncomfortable with slightly longer/thinner runs with smaller speakers and surrounds (because they're less critical).
And, yes, I would adjust that slightly (towards fatter) for speakers that are know to be very low in impedance, because they will be more affected by wire impedance.
And, yes, I would consider most of those numbers to be overkill (but copper is cheap, so a little overkill doesn't cost much).
And, yes, I would consider anything more than $1/foot spent on speaker cable to be "for aesthetic reasons" (if you do it, do it because the fancy wire is prettier, or feels nicer, or looks like it will last longer, which are all perfectly good reasons)
This whole discussion is way off base. What a connection wire represents is insertion loss. Take the wire resistance divided by the sun of the wire resistance and the minimum speaker impedance and that represents the % of signal that is lost to the wire. Skin effect is not remotely significant at audible frequencies. Additionally an amp is a voltage source and if it is driving a short it is no longer a voltage source because theoretically the output voltage is zero and it is driving only unregulated current (but in reality any protection would have already kicked in so the point is moot). This is tech school ohms law stuff. Simple.
How is 7 AWG as apposed to 14 AWG wire going to improve his system? At 30', the differences in resistance is insignificant AND imperceptible.
