Rank: Advanced Member
Groups: Moderators, Registered Joined: 13/02/2009(UTC) Posts: 3,094 Location: Probably not here Was thanked: 113 time(s) in 76 post(s)
|
Recording music is exactly what I've been studying the past year.
Not in the "here's a backing track now sing over it" way, but in the "okay now all your instruments are miked up, so, whenever your ready let's go for a take" and "okay, that was good, but you hit the mic-stand with your leg, so don't do that."
A lot of people (myself included in the past) assume it's as simple as running some software on the computer, or plugging an instrument into the computer soundcard. It takes a hell of a lot more than that to make anything not sound like ass. Plus, a lot of people don't even realise that in order to make a stereo track, you're recording a lot more than one stereo file. In extreme cases hundreds of recorded elements will make up a single song, each one is its own individual recording and must be dealt with both on its own terms and more importantly as a part of the whole.
On the technical side of things, you basically have three signal levels, mic, line, and instrument, and on the mic you usually have phantom power if it's a condenser, or battery power for a valve mic, and instrument is just a hi-gain version of line level used for direct input guitars etc. Next up on the signal chain you have the channel path, which on an analogue desk goes through dedicated gate, compressor and EQ, as well as patched in external processors and effects, possibly requiring further routing within the desk, before it is then routed out of the desk to a recording medium, ie. tape or hard disk, while on a digital desk it is immediately routed from the input into the software, whether that be pro tools or whatever else, which allows you to plugin any software dynamics processors and effects. At each stage of the signal chain you have to be careful to keep the level within a certain gain range to prevent loss in signal quality through a rising noise floor, or simply clipping the high end. In practical terms, this means you've got to make sure the mic is set up right in a quiet insulated space, there's no headphone bleed from any cue mix or buzzing electronics or whatever else, and then make sure the mic level peaks somewhere between about -18dBfs and -6dBfs both as it enters the system, and as it is actually recorded.
Simple really.
Then you've got to work primarily with gates, compressors and EQ to reduce bleed, compress dynamic range, overall volume, and then cut and boost specific frequency ranges to enhance certain qualities of the recorded performance. Though this is largely a mixing stage issue, it's worth doing a once-over in the recording so that you've got something approaching what it'll actually sound like. Because in software this stuff is non-destructive, you don't have to commit to anything.
Then the horror of mixing when you're trying to resolve frequency clashes, instrument separation, trying to make every element clear and distinct, and yet also not disparate and disconnected. First you're working on the dynamics as mentioned earlier, just trying to get the most out of the actual recordings. Then once you've done that you have to build on this and turn it into something more than just a bunch of instruments. This is when all the great fun effects get involved, delay, reverb, chorus, pitch-shifters, pitch-correction, and my personal favourite (perhaps unsurprisingly), distortion. Fortunately, if you've gotten this far, it's not as easy to fuck up this part as it is the dynamics - provided you take a less-is-more approach. And then finally, all your levels and panning, left, right, up, down, and you're done.
Ah, but wait, not quite. The subtle hidden art of Mastering, which is quite honestly a gigantic pile of bullshit because it's all about compromising dynamic range (the difference between the loudest and the quietest sound levels in a track) in favour of overall loudness and consistency. More compressors (this time multiband), subtle fiddly almost imperceptible amounts of EQ, and a great big mastering limiter strapped across the whole thing to squash the top few dB. Good mastering helps a track sound professional, finished. Okay mastering you don't even notice. Bad mastering really ruins good music because it squashes the music so much that you can only listen to about half an album before your ears get sore. It'd be like turning up the colour to 100 on your TV - it's way too much. Yet it happens, because on commercial radio, on television, in clubs and wherever else popular music becomes popular, the louder songs get noticed more. It's a constant battle between sheer volume and dynamics that they call the Loudness War, and many a major album release has fallen prey to it. The problem is simple - in digital audio, you cannot go above a certain point because there's not enough bits to store that much information. If you do go above it, it causes digital clipping, which is a harsh unpleasant fizzy distortion. In the old days, if you wanted a louder album the solution was simple: TURN YOUR VOLUME UP.
Apparently the last Iron Maiden album came back from the mastering engineer so horribly that the band's producer said "fuck this" and put it out unmastered. Now I don't give a shit about Iron Maiden, but man, I wish more producers had the balls to do that.
So where was I...
Oh yeah, recording. It's pretty simple really. |