I'm compelled to write my thoughts on this issue, something a lot of us probably rarely ponder upon, disturbing trends that appear to be here to stay.  I'm talking about how digital audio recording and distribution has actually made audio quality a lot worse, rather than better, like it was supposed to.

First, let's talk about digital music distribution.  These days, most people get their audio fix online, whether it'd be via mp3 downloads, streaming audio (such as on the MySpace player or on Youtube), or, very rarely via an uncompressed .wav file purchased at Beatport.  This is disturbing, as an entire generation is now growing up on inferior sounding audio, and doesn't give a rat's ass about it.  Let's face it, the audio quality on most streaming websites is extremely inferior, and, by the way, a lot worse than audio cassettes in the 80's (and there, at least, you had the satisfying physical interaction with your music by pressing that clunky "play" button on your walkman). 

Now, I personally think that digital distribution is fantastic, and especially streaming music could potentially be the saving grace of the ailing music industry and a viable answer to illegitimate downloads.  Music can now reach every person in all corners of the world with the press of a mouse button.  And I'm sure that not everyone has lightning speed internet, some even still dial-up, which is why audio needs to be compressed to a data rate that can accommodate such connections.  Point taken. 

But what is disturbing is that people don't actively seek out superior-sounding audio in the form of 320 kbps mp3's, or uncompressed wav files (and I do laud Apple for offering 256 kbps files now on ITunes, instead of the original 128's).  Whatever happened to high fidelity, and the gratification one got out of owning a superior quality of music.  When CDs replaced cassettes and vinyl, people were flocking to buy everything on CD, giving up tapes especially, for the superior CD.  Now we're in a worse place than we were in the 80s.  Perhaps, in the end, it's a matter of economics--people rather listen to crappy quality, free stuff, than paying for the actual high-quality original.  Which defeats the argument that people are willing to pay a premium for better quality (like they used to, at least for audio).  And, I suppose, it's not just music we're talking about--how else would one explain the extreme success of Walmart.  Just something to think about.  But remember, club systems are a lot less forgiving than earphones, for example, so an mp3 under 256 will probably not sound very good in a club, as the high-end frequency spectrum such as percussion, are effected most by compression, and you will hear a fizzling effect.  All we can hope for, is that, as internet connections get faster throughout the world, audio will improve once again.  All infant technology eventually improves--the only stump is that digital audio is not particularly in its infancy stages, so we'll have to see...

Now if you can't hear the difference, fine.  And you may not even hear the nuances of a mix that only come out in an uncompressed audio file.  The reason why sample rate was set at 44.1 kHz (or 48, depending in which medium you work in), other than  the 2 formats being completely indivisible by any of the other divisors because of format wars in the 80s, was because you need 2 samples to sample a frequency, so 2 to sample 1 Hz, and 40,000 to sample 20,000.  Human hearing generally fizzes out between 17-20 kHz, so people figured if we set the sample rate to sample slightly higher than that, we can sample all frequencies that humans can hear (which is why it's complete and utter nonsense to raise the sample rate to 96kHz, for example, like they try to do on Bluray discs now, etc--it's just a marketing gimmick, nothing else, unless they want dogs to listen to the higher frequencies). 

What DOES, however matter, is bit depth, and this brings me to my second point: the limitations of digital recording.  Now, of course, when I write tracks, most, if not all sounds are created digitally.  No arguing there.  However, I do not mix them all together digitally, like so many producers do these days, on Fruityloops or other products, but rather on a good ol' fashioned analog mixing board.  The problem with mixing digitally is, that every time you mix 2 instruments together (say, for example, a kick drum and strings), you're in essence adding the 2 sounds together.  Now, if you're adding these 2 instruments digitally, you're performing a "subtractive" summation.  That may seem counter-intuitive, but think about it this way: You have a set number of bits (in most cases 16, sometimes now 24 or 32), and when you play one instrument, and add another, by default, since you work with a finite number of bits, some bits need to be taken away for the other sound to play on top of the original sound.  If you have 20-40 instruments going, like you tend to have in dance music, each individual sound suffers, and actually changes character as opposed to when you play it by itself, since bits are taken away to play the other sounds.  To get around this problem, professional studios are now actually acquiring analog summation devices.  And you thought we were done with analog.  So once again, digital proved to be not an all-out improvement, and people are going back to some technology that was thought to be going extinct.  The beauty of analog, that digital cannot reproduce, is that it's "additive" summation.  I.e. no matter how many instruments you add, they just add in with the rest of the instruments.  It's like a stream combining with other streams to form a river, or even the ocean.  You can keep adding, without having to take something away somewhere else. 

The difference in sound may not be so obvious to most casual listeners, but it is very obvious to me and other audio professionals, and tracks mixed poorly digitally (and believe me, there are many, many out there) have a very sterile, uniquely "digital" quality to them.  Of course, more expensive workstations such as Logic or ProTools have much better digital summation algorithms than Fruityloops or Ableton.  And I'm not saying one cannot create good 100%  digital mixes.  But there is a lot more to it, than many producers realize, and care needs to be taken in the use of EQ, compression and limiting on all individual instruments in a sequence to make a good digital mix.  And if you must mix digitally, consider mixing at a higher bit depth, such as 24 or 32, and export your master into the standard 16 bit format.

In the end, of course, a great musical idea will always override a bad mix, crappy internet streaming or digitally distorted instruments.  And though I've mixed many projects in purely a digital domain, I would never want to part with my analog tube compressors, preamps and mixing board.



YOU FOUND GEORG @ MY TOPFRIENDs ... :-)