Audio File: Musical chairs

In a large recording project, phase relationships can have a big effect

Some of you regulars may be wondering just what the heck I’m doing lately, writing about recording instead of about listening. This column is called “Audio File,” and we should discussing audiophile stuff, right?

While we’re at it — shouldn’t recording advice come from an experienced pro, and also include techniques to try and maybe a list or two of some moderately priced gear?

Well, there are tons of articles like that already, and I actually am an experienced pro with plenty of work under my belt. I have my own thoughts about how to capture sound, but because I’m also a rabid audiophile, my perspective on recording is often driven by audiophile interests, and perhaps stands out a little from the norm.

So I hope my takes are useful in some way. Last week’s topic about maintaining proper phase leads us right into this week’s examination of phase effects from the listening side of things. It’s a rare meeting in the middle between audiophiles and mixing engineers, where at least one of these gives a hoot about the other.


Google “audiophile absolute phase,” and click on the “Fact or fiction” link from Stereophile, if you have any desire to see the kind of fraught handwringing goes on about phase in the audiophile quarter. The focus in this article is about a preferred polarity for instruments and recognizable sounds, and whether the opposite polarity induces an audible change for the listener.

My experience says it does, but this effect is more pronounced with more tracks added to the mix. In a world with computer-based mixing and the possibility for hundreds of tracks in a given song, I think there is too little attention paid to the “absolute phase” of the tracks, relative to each other.

When lots of instruments are placed in a song, the chances of having some of those tracks be inverted from the others rises. This has to do with varying distances to microphones, among other things. Given that most instruments have asymmetrically shaped top and bottom halves of their waves, my recording and listening has led me to believe that for many sounds, polarity does matter.

The effect of this, to a stereo listener on the playback end, is that certain sounds are in front and some are behind those in the stereo field. Flip the polarity of the system by inverting the wires on both speakers, and if everything is set up correctly and the room is good, the instruments or voices will flip positions, front to back.

I can do this on my system instantaneously with a press of a button, and this effect is real. I can hear it on plenty of recordings, although less so on some recordings and definitely less on “audiophile” recordings where few microphones are used. But John Mayer’s “The Search For Everything,” for example, shows this well on my system, and clearly changes character when I press the button.

It varies song to song, and moment to moment within songs, but it’s real and it’s there. It’s something I think the modern recording engineer should be thinking about when many tracks are in play. Take the time to flip some of those, and listen closely to the differences.