Dynamic Range Compression

Pete Bilderback at Flowering Toilet has a fascinating series of posts about dynamic range compression, and how it’s making music sound awful.

It starts here, and it makes sense to at least take a quick look at the graphs before delving too deep into my comments, so that you’ll have a better idea of what I’m talking about.

“You listen to these modern records, they’re atrocious, they have sound all over them. There’s no definition of nothing, no vocal, no nothing, just like — static.”

-Bob Dylan

“Loudness is killing music, and I’m not talking about the 80s Japanese Heavy Metal band.” You’ve probably heard some audiophool, musician or music geek say something like that over the past couple years. But what are they talking about, and what exactly is wrong with loudness?

Loudness is actually a somewhat misleading term…. What is actually going on is that most contemporary pop music is getting all of the dynamic range squashed out of it by means of dynamic range compression. Dynamic range is the difference between the loudest possible undistorted sound and quietest sound that is audible above the noise floor of the recording. Dynamic range compression limits the difference between the loudest and the quietest sounds. Subjectively, compression makes for a louder sounding recording. It is also a practice that is currently being taken to absurd extremes.

The net effect of this is to strip subtlety out and make the music exhausting to listen to, in a manner similar to how some cheaply-recorded albums can be exhausting to listen to–your eardrums just get tired.

While I don’t claim to be an audiophile or anything close, I like to think I’ve got a pretty good ear, and I’ve noticed this, too, without necessarily being aware of the reason. Based on a good review, I bought the Cold War Kids’ self-titled debut. I think I made it all the way through twice, both times in my car with stops along the way. I never made it through uninterrupted. I can’t make cool graphs like Pete’s because I don’t have the right software, but I suspect that if I ran songs from that album through Rip Edit Burn or something similar, I would find that they bore a great resemblance to Pete’s examples. To me, it never sounds like the music ever gets a chance to breathe.

I think one of the best-sounding albums I’ve heard in the past few years is (don’t laugh) Pink’s I’m Not Dead. The single “Who Knew” opens with guitar, then vocal and drums, then bass, each element distinct–they have been allowed to breathe. The song has a soft-loud-soft structure, but it isn’t so ridiculous that you have to turn down the stereo when it hits the loud parts. I wonder how that one would look graphed out? I’ve listened to it straight through many times, and it didn’t exhaust my ears. Maybe I’m way off base, as this is precisely the type of album where one would expect to find a lot of compression. Is this a rare example of sonic restraint in recent pop music? Or just an exceptional production job that overcomes the limitations of excessive dynamic range compression? I’d be interested to find out from someone who knows more on the subject than I do.

If you’re at all interested in this geek stuff, check out the series. It’s quickly getting longer, so if time is limited, start with the inaugural post, followed by the one on Mudcrutch. If you want more, read the piece on Born to Run and proceed from there.

I was thinking about buying the deluxe re-release of Steve Earle’s Copperhead Road. After reading about Born to Run, I’m nervous about doing so. Maybe I’ll just keep my old copy and spend my twenty-five bucks on other albums.


5 Responses to Dynamic Range Compression

  1. Pete says:

    Gordon–thanks for noticing the series. I do tend to get a bit long-winded at times. I justify this by telling myself I am going into a subject “in depth,” when in truth I am in need of a good editor.

    I’ll see if I can take a look at the Pink song you mention (I don’t have it).

  2. Pete says:

    I found an MP3 of “Who Knew” on another blog and took a look at it. Your suspicion was correct. There is a tremendous amount of dynamic range in the song. While it does get pushed very loud at certain points, there are lots of clear peaks and valleys in the music, much more so than is typical in pop music today.

    Of course when the song gets played on the radio, the radio station will add an additional layer of compression, and rob it of much of that range.

  3. Thanks, Pete! I actually wanted to just outright say it, but suffered a crisis of confidence and hedged my bets. I’m a wimp! I should have trusted my ear.

    Incidentally, what software are you using? You’ve written a fun series of posts, and you’ve inspired me to want to screw around with some songs myself.

  4. Pete says:

    Hi Gordon. It is good to trust your ears, but there is probably a reason we are hesitant to do so. People say “seeing is believing” so often for a reason. For me, looking at these tracks in a sound editor, it was shocking to “see” what was being done to the music.

    I have a few programs that allow me to look at sound files. The best of them is Adobe’s Soundbooth. It is a very powerful professional editing program with lots of features. Unfortunately, it’s also expensive. Another good program is Rogue Amoeba’s Fission. It’s a much simpler editor, and much less costly. Unfortunately, it doesn’t allow for quite the same “close up” view of files as Soundbooth, and it is Mac only. Audacity is a free, cross-platform program. I find it somewhat difficult to use, but you certainly can do the basics with it, including the ability to “zoom in” on particular sections, and detect instances of clipping in a track. For a free program it’s quite good really.

  5. […] music industry, cut it out!  Or at least have the decency to release a second, respectable-sounding edition like Tom Petty […]

%d bloggers like this: