Thursday, September 06, 2007

More on Compression

I had a small post in June about my concerns with the overuse of compression in audio recording. Richard Friedman of All I Know^2 now points to an article in the Electrical Engineering Society's Spectrum magazine laying out this problem further, in short, the drive to record everything at a undifferentiated loud level and the equipment with which we listen to music is turning everything into undifferentiated noise.

(Money quote: This might be one of the biggest reasons why most people are completely unaware of the loss of dynamics in modern music. They are listening to songs in less-than-ideal environments on a constant basis.)

Contemplate, for a moment, the first images in this article, comparing a typical pop waveform from a generation ago with a contemporary example. I believe that the uniformly dense level of information contained in the contemporary example is a further illustration of my argument that "more is often less" in music. It's a bit like comparing Christian Wolff's Trio 1 with Milton Babbitt's Composition for Four Instruments: the uniform distribution of pitches, attacks, and rhythms in the Babbitt example creates a noisier environment than that of the Wolff, with its irregular distributions and lack of emphasis on the consumption of complete aggregates in each parameter. For all the insistence on the values of details in Babbitt's compositional thinking, the necessary end of the works in audition is one in which details are lost.

It may be argued that there are examples of radical music characterized by minimal materials that also display a uniform dynamic distribution; I can only counter that the best works of the sort use the minimal material state explicitly to compensate for this, in that the works are better framed for perception of details in whichever parameter the composer has chosen to focus his or her work.

One more thing: I am famously bad at predicting anything other than poker hands and thoroughbred race finishers, but doesn't this article suggest that if popular music producers wish to create a product that is more attractive to an audience through its lack of fatigue, they will have to start using sound environments that are more varied? Not completely varied, which would be equally if not more taxing on ears, but admitting a greater degree of differentiation.

BTW: I really like Bart Kosko's book NOISE, for an excellent introduction to the subject of physical noise in general and to its costs and benefits. Every musician ought to have two books on their shelves with that title (the other by Jacques Attali, part and parcel of 1985, but still very cool).

1 comment:

Ben.H said...

I've just started playing with a handmedown iPod and wondered about the parallel rise of dynamic compression in recordings, and in the uniformity of dynamics in a lot of postwar music (not just in pop, the minimalists and downtown composers who write music with an unbroken "surface"). It's interesting how you've found a way to group Babbitt with that lot!

In particular, I wondered if this was because more and more music is heard in competition with other sounds. Constant loud dynamics so nothing is lost against background noise, or at least an unvarying sound level which creates a figure in negative against an irregular background of noise.

Cage, Wolff et al started writing music which was intended to coexist with noise, but this idea seems to have lost out against people trying to fight it with the technology that help create this situation in the first place.