DonCarr
Head-Fier
- Joined
- Apr 13, 2010
- Posts
- 98
- Likes
- 46
LOL TREE LOBSTERS

Very loosely, speakers are low impedance, require high current and low voltage. Headphones are high(er) impedance, require low(er) current and higher voltage. Consequently, the types of wire used for each is different, though using 18AWG wire, which is just about speaker cable or power cable (!), for headphones, is pretty funny. The more current, the thicker the wire required, which is why power lines are relatively thick. It's more likely that large voltage swings would be affected by the quality of audio components, and less so, say, speaker cable that carries a large current.
The problem with this kind of discussion on here is that it's pretty far removed from science, involved in proposing theories, methodologies of testing, possible errors, performing experiments and writing them up correctly, but just people arguing about things they read online and taking sides because it suites their desire to be right, not to find out or know the truth. Unfortunately and presumably as a result, the people who are heavily involved in actual DBT testing and whatnot don't seem to be posting any more because of this.
cegras: thanks for the correction on power transmission lines, I was indeed wrong. You indeed appear to have a proper background in these matters. My "unscientific" comment is directed at people who (usually) have little or no background in science who parrot bits of science purely because it validates their belief system and are not interested the truth, nor any science that might disagree. Indeed, the thread was started in the manner of a religious quest, not a search for truth.
There's one thing I'm certain of: there's never going to be a test that convinces everybody. You'd have to have a statistically significant sampling of self-identified audiophiles (since we're seeing if there's a difference *to audiophile ears*, not in the general public) that runs a statistically-significant number of tests on each with each of a statistically significant number of cables (in case some cables are snake oil and others aren't) on a setup that is both standardized and that the audiophiles are previously familiar with. The question of "do audiophiles really hear a difference with expensive cables" is actually quite massive.
I'm content to conclude that, while cables may make a difference, they don't make enough of one to warrant my spending on them (unless a headphone I want off the FS forum happens to have been recabled, in which case I simply appreciate the well-built cable). I'm a very happy agnostic.
There's one thing I'm certain of: there's never going to be a test that convinces everybody. You'd have to have a statistically significant sampling of self-identified audiophiles (since we're seeing if there's a difference *to audiophile ears*, not in the general public) that runs a statistically-significant number of tests on each with each of a statistically significant number of cables (in case some cables are snake oil and others aren't) on a setup that is both standardized and that the audiophiles are previously familiar with. The question of "do audiophiles really hear a difference with expensive cables" is actually quite massive.
I didn't mean to appear harsh, but to summarize, asking me to believe there is a difference outside of psychoacoustics is as annoying to me as (I presume) me asking you to believe there isn't one. It's like telling me my education is useless, and I guess me telling you your hearing sucks isn't very nice either. Ho hum.
It's much simpler than that. Using two ears, two cables and a binomial distribution, the following number of correct blind discriminations will be adequate for a statistically significant result (p < 0.05):
5 out of 5 : p = 0.0313
7 out of 8: p = 0.0313
8 out of 10: p = 0.0439
Anyone who says that they can clearly hear a difference should be able to clear this absurdly low bar to achieving a universally accepted demonstration of scientifically real effects. Hypothesis testing is fun, fair and easy!
^ I think at this point the debate is at a much more fundamental level. As nick_charles noted, there has not been *any* demonstration of a statistically real effect, aside from between a very short and very long cable with an unamplified signal. A single demonstration of a real effect would be a better place to start (using whatever ears and setup that could resolve that difference). The rest is gravy.
For example, in my line of work, one person 53 years ago showed one statistically significant effect from one brain structure. Now there are hundreds of articles every year, and a scientific journal named and dedicated to that structure.
As nick_charles noted, there has not been *any* demonstration of a statistically real effect, aside from between a very short and very long cable with an unamplified signal.