Mixology: Mixing and production
I've said this a thousand times before: our biggest obstacle is muddy audio. We will go to any extreme necessary to prevent this. This means that you will have to trust yourself and, to some extent, trust me during the mixing process. You may have a wonderfully lush guitar track laid down that you think sounds great, but which eats up so much sonic space that it refuses to play nice with your other tracks. This happens. But with careful EQ, you shall overcome. Keep in mind that each track is part of the whole mix, and has its sonic space to fill, and you'll be set. It helps, of course, to have golden ears, but it's not necessary.
We've discussed EQ and drums, briefly, in the drums section. We want to make the snare drum sound sharp, the kick drum bassy, but tight, and the cymbals actually sound like cymbals. I've given some EQ guidelines in that section which you should review now to get the most out of your drum track(s). Keep in mind that the drum kit takes up an enormous amount of sonic space, but (usually) isn't meant to overpower the other instruments. You should try to empasize the crack of not only the snare drum, but also the kick. Your toms should sound as clear and transparent as possible -- that is, cutting out many of the midrange frequencies. Finally, you'll want to EQ your cymbals to be bright, but not overly harsh in the extreme high frequencies. If you've got this all in one track, you've got a slightly harder job, unless you really did a great job recording them. If this is the case, the best you can do is just fiddle with it until it sounds best. As for panning the drums, it's customary to pan the drum kit as if it were a real kit within your sonic field. Most recordings put you in the drum throne, but there are also plenty that put you facing the kit. This is your choice.
Obviously, your bass is going to take up most of your bottom end. Depending on how new the strings on the instrument were, this track may be either extremely bright or extremely dull. You'll want to compensate in either direction: you probably want your bass to sound snappy, but not harsh. Boosting it in the 1KHz range might be a good idea if it's too deep (along with cutting below 400Hz or so). If you've got a really bright track, you will probably want to cut a little above 3KHz and boost just a bit around 250Hz. In either case, you might want to make sure you're not too heavy on the sub-100Hz frequencies. Just the right touch here can make the difference between a warm, bassy mix, and a muddy one, especially since your bass will be panned center and probably be dry, except for a little compression.
Many -- maybe even "most" -- of the rock recordings you listen to will have many more guitar tracks than the band actually has. The texture in much of popular rock is created almost entirely with layer upon layer of guitars. If you're confident in your ability to keep them from blending all together -- not too difficult if you are careful -- you might want to do this. If you're dealing with a song with nothing but rowdy, distorted guitars, you might only need 2. If you've got an arrangement with more delicate, clean guitars, you might want many more (or, you could go the spartan route). Whatever the case, there are a few things to remember. First, when you've got a distorted guitar -- the sort that's just playing power chords or double stops or the like -- you'll probably want to double it and pan each one hard left or right. This will lend your recording "size" and "depth." Of course, if you're going for the garage-punk sound, you might want only one, panned slightly off-center. But if that's the case, you probably shouldn't be reading this.
Panning is an important part of your guitar arrangement. Since there are many things that should be panned center -- vocals, bass, much of the drum kit -- putting your guitars out toward the edges can help give you a clean, crisp recording. As noted, it often helps to double parts and pan them all the way out. When you've got only one guitar playing, you might want to consider a short stereo delay that gives you a little more open space. Alternatively, you might want to pan it slightly to one side (when doing this, I tend to base it on where my drums aren't: if I'm riding on the hi-hat, I'll pan right; I'll pan left if riding on the ride cymbal). Of course, these are just guidelines. What works for one recording won't always work for another, so experiment.
EQ on the guitars should emphazise whatever it is that makes them important -- the bite, the gruff, the sparkle -- and de-emphasize everything else. If you've got a wildly distorted guitar you're using as a rhythm section, you will probably want to boost it in the 1KHz range and cut below that very severely. Your bass will fill the lower bits of the sonic spectrum, and having the guitar there will just make it muddy. The same goes for bright, clean guitars; boost them where they're strong (maybe 2-3KHz) and cut elsewhere. There also tends to be a lot of harsh bite in many overdriven amps in the 3-4KHz range. Use this to your advantage if you want your guitar to cut through the mix. But remember to compensate by cutting it elsewhere, or you'll wind up with guitar soup. Again, you'll just have to play with this to see what best emphazises your track's strengths while keeping it from being overpowering.
Most people hate the way their voices sound in recordings. The rest are narcissists. I fall firmly into the former category (for good reason, really, but that's beside the point), so I spend a lot of time trying to make my voice sound less awful. One of the easiest ways to do this -- as evidenced by Foo Fighters (the album, and, to a lesser extent, the band's other albums) -- is by layering your vocals. Recording a vocal track a number of times has the effect of smoothing out any inconsistencies and making you sound like a better singer than you really are. The danger here, of course, is losing definition and making your vocals muddy. Smoothing out inconsistencies also has the nasty side-effect of smoothing out the crisp bits. You'll need to be careful with your enunciation and careful EQing these tracks. But it works -- prudent layering can make you sound more professional. When layering, you might also want to think about panning. Many layers all panned center -- even when done well -- can sound muddy. Pushing a few of them slightly left or right can really open up the track.
As with guitars, the EQ here is going to depend on the focus of the track and the qualities inherent in your voice and microphone. I find that I do best to cut my vocal tracks sharply below 500Hz and boost them very slightly around 1.2KHz. Your voice might be "tuned" higher or lower, so experiment to see what brings it out best without mudding up the mix. Boosting in the 8-12KHz range can also give extra sparkle to your vocals, which you may or may not want.
Backup vocals might need to be mixed somewhat differently. If you're trying to emphasize them, you might want to boost them way up around 10KHz. If you're trying to hide them a little, you might want to cut the same range, but boost slightly around 4KHz so they're still punchy. You might also want to pan them slightly to one side or another. Unfortunately (or fortunately) it's all a matter of taste, and you'll have to try to see what suits you best.
One of the biggest mistakes you can make when mixing a home project is using too many -- or too-dense -- effects. A little reverb can be very forgiving, and really improve your tracks. A lot will just make it sound like karaoke or, worse, make it unbearably muddy. The best way to add effects to tracks is to use your multitrack's effect sends. Most will let you have as many AUX sends as you need, and will let you send to DirectX or VST plugins. You can then alter the amount of each track that's sent to the effect, and the amount of the effect return that gets mixed in with the master mix, for reasonably precise controls. You will probably want to put a little reverb on your drum and vocal tracks, as well as your lead guitar track. Many acoustic or clean electric guitar tracks could benefit from a light reverb as well. You might also want to consider a very subtle chorus or short stereo delay to increase the stereo separation of some of your tracks. The key is -- now repeat after me -- keep it from getting muddy. Every track has its place, and too many effects will drown out the more subtle instruments.
By now, you should be able to come up with a workable recording to show off to your friends. They will be impressed, I promise. There's nothing quite like putting on a CD of yourself that doesn't sound like it was recorded in a basement inside a tunnel filled with springs. The next session will cover sweetening the recording to make it sound polished and (at least semi-)professional.
<<Make Me Scream -- Masters of the Universe >>