Excellent posts from Greg and Glen.
Im not keen on doubling parts...but after working on it a bit its sounding better but it took some pretty drastic EQing
Honestly, it's almost shocking how much double-tracking a guitar part can do for the richness of the sound. This was the single-biggest improvement in all my long history as a low-budget recording neophyte, even bigger than moving from a mic into a 1/4"-1/8" converter straight into my computer's soundcard to a proper firewire interface. Done right, two separate takes can sound way bigger in concert than either of them on its own turned up to the same volume.
It's tough to say why, but someone around here (I forget who, sorry!) once theorized something that made a lot of sense to me. Basically, we're used to "hearing" a guitar in the room, where it's reflecting off the walls and floor and reaching us (where we pick it up from two separate sources, hard left and hard right) at ever-so-slightly different times, with ever-so-slightly different room coloration. Doubletracking, especially if you vary up the tone a bit between takes (say, a bit gainier and heavy on the mids left, and a little cleaner and brighter and not quite as mid-heavy on the right), more closely replicates the way our ears perceive the sound of a guitar in real life.
A couple caveats - one, to do this right, you really need to be pretty damned tight of a rhythm guitarist. This is especially true if you're going for something more adventuresome like quad-tracking your rhythm parts (kind of the modern metal standard these days), where even slight inconsistencies start to snowball into one big muddy mess.
Personally, I rarely use more than two tracks for one simple reason - I write instrumental guitar music, and another guitar will be handling all the melody duties, as opposed to a human voice. I don't WANT a huge wall-of-sound guitar tone, because I need to leave ample room for the melody/solo guitar. So, I guess that's caveat #2 - consider your ultimate objective when you're thinking about the "size" of the guitar part, as in a dense mix a smaller but better-fitting guitar will sound as big if not bigger than a huge sounding guitar.
Caveat #3 is related - this depends on arrangement of course, but generally its worth thinking of the bass guitar as part of the guitar sound. So, you often don't want a ton of low end in your rhythm parts (and I say this as a seven string player, no less) because really the bass will need to be holding down that range. Likewise, you should also try to get a handle on where most of the "energy" is coming from your bass guitar sound, and leave some space there in the guitars - maybe a few EQ notch cuts where the bass really seems to poke out, etc. A huge, deep, super-bass-heavy guitar sound will sound massive on its own, but will just be a nightmare when you try to bring in a bass and a kick. This is only partly about post-processing, too - a lot of it is how you dial your amp and mic up. Why crank your bass up on your amp, if you're just going to have to strip most of that away eventually to get it to play nice with the bass?
Caveat #4 - it's becoming a bit of a cliche, but remember that when you're doubling, if you want clarity you probably need less gain than you think. Again, two doubled tracks will sound much "bigger" than either one on its own, and adding gain tends to compress the guitar, robbing it of a bit of its dynamics. Taken too far this can be a problem, but it's worth experimenting with. Likewise, the guitar is a midrange instrument, and scooping the living shit out of your mids will leave your guitars sounding hollow, empty, anemic, and generally nonexistent in the mix. So, sort of another variation on the theme of "evaluate your guitars in the context of the full mix (including other guitars) rater than how they sound as solo'd takes."
I'm rambling - hopefully some of this helps.
