The Science, and the Politics, of the Coat Hanger
December 4, 2008
Before we get started: no, there is nothing about Roe. v. Wade in this blog post. So sorry... but if you're interested in recording your own music and you've never been here before, then hey, welcome to my blog and I hope you enjoy the website.
Now, where was I? Oh yes.
One of the ongoing stories that tends to pop up a lot when people surf around the portion of the web that discusses audio technology is the Coat Hanger Cable story. You can Google it yourself and find it in at least a hundred different places, but I'll summarize it for you here.
Some guy wondered if the so-called "audiophile cables" out there were really worth all the money. So he got some audiophile friends together and had them do an A/B listening test to a stereo with audiophile cabling from a famous maker, vs. cables made from soldering audio connectors to straightened-out wire coat hangers. No one could tell the difference.
From this story, which has long since gone viral and spread through the Internet almost as quickly as that YouTube video of You Know Who doing You Know What to You Know Who... you know the one, right?... dozens of discussions have popped up all over the place, with people happily trumpeting one or more of the following opinions:
"HAH! I knew it all along. Those audiophile cables are ripoffs, and this proves it. Audiophiles are suckers with money to waste."
"What a bogus test—you can't call THAT company's cables 'audiophile', they stink! Try it with real audiophile cables and see!"
"How do you expect to tell the difference between cable types on a stereo that crappy? That's not audiophile gear... it sounds so bad that differences in cables will be inaudible."
"Yet another example of deaf people slamming things they refuse to understand, because they're afraid to admit they're deaf."
And so on.
But in all this clamor, there is one thread that thankfully turns up more than once, that goes something like this:
"Who are these so-called golden-eared people who did this test? What are their qualifications? What was the listening environment like? How was the test conducted? What was the control? What was the time lag between A and B? Where's the hard data? This is a great story for starting arguments, but no matter what you believe, it proves nothing."
And that's my point for today: sensationalism aside, tests like this prove nothing. The optimist sees the glass as half full; the pessimist sees the glass as half empty; and the ex-physicist sees the glass as twice as big as it needs to be.
When I read that story, a lot of thoughts went through my mind. The first was, as a big hunk of wire with air as a dielectric, a naked coat hanger would actually make a pretty damn good audio cable... as long as you were able to crimp it into shape and never wanted to move it around, and as long as you weren't worried about it picking up nearby radio stations. A more "fair" and less sensational test would be comparing two different insulated audio cables, one that sells for a few cents a foot and one that sells for a hundred dollars a foot, under otherwise identical circumstances.
The second was, without a proper instantaneous switching system with carefully matched SPL, any test of this kind is meaningless. The human ear's relaxation time, its time to reset itself internally after exposure to a certain type of audio, is very fast, on the order of a few seconds, tops. My colleague John Rossi III, who writes frequently for Recording, once did a test of people's ability to distinguish MP3 files from uncompressed audio in A/B testing. I don't remember his results completely, but one thing he told me that stuck with me was the fact that the ability to distinguish between compressed and uncompressed audio was greatly affected by the time it took to switch signals. If you had two copies of the same song playing back in sync and switched back and forth between them, it was easy to hear the difference. Turn one off, wait a bit, turn the other one one... and it became much harder. If a test involved detaching one set of cables and attaching another before listening again, the data are basically useless. (This mistake has been made by audiophile magazines, too, by the way.)
The third thing is that politicization of "audiophile" gear vs. "tin-eared Neanderthal" gear aside, not knowing much about the audio performance of the stereo itself also wrecked the data. You need to have a very well defined system, and change ONE thing, and listen for a difference. ONE. If you change more than one thing, you don't know which change did what. This is a classic mistake all engineers make at some time or another. "It sounded like ass, so I messed with the gains and changed a couple of cables and tried a different mixer channel and now it sounds different than before but it still sounds like ass." From this we get...what?
So if I had to comment on this coat-hanger test at all, it would be to reiterate one person's request to have this test redone under proper conditions. If you want to start a fight, you can say just about anything and accomplish that end. If you want to learn something, you have to be a bit more careful about what you're asserting and how you got to that conclusion.
I prefer learning to fighting. I don't do the former nearly enough, and the latter happens too often without my help.