Since a few people asked:
The Drums
The first thing I do when I'm mixing is I get the drums sounding good. A good, balanced and solid drum sound is the foundation of the song, for me. For this song most of the drum tweaking work I did happened to the overheads, so I'll go over those first. The whole reason I wanted to get the Kick Start source from you in the first place was to try out a ludicrously and anticlimactically simple processing technique on the close mics, which I will go over second.
- I started by panning the Overheads hard left and right and giving 'em a listen. First things I noticed: the snare was slightly out of phase, coming in a little sooner on the left than the right. I'd guess the left mic was a bit closer to the snare. To recenter the kit around the snare, I slid the left OH back so it was starting 2/3ms later than the right, at which point the snare moved to the center of the stereo field.
- The next thing I noticed was that the crash was way over in the right OH and really, REALLY loud; to compensate for this I panned the right overhead a little closer to center (about 3/4 of the way to the right), leaving the left OH panned hard.
- I then took a fine parametric EQ, boosted the peak filter as high as it would go (something like 18db) and put the q up as narrow as it would go (somewhere around q=10) and manually swept it up and down the full frequency range, listening to hear if anything really ugly jumped out at me. I found what I was looking for up around 4khz: a tinny, painful treble overtone, probably coming from the cymbals. I notched this out with a 9db cut and a medium width Q of 3.
- OK, the overheads were sounding pretty good, but that crash was still waaay too loud. For scenarios like this, where I want to turn down just one element of a kit, it's time to bust out the split-band compressor. I centered the frequency band that I wanted to tame up around 6khz where the bulk of the crash sound was. This is a tricky operation, actually, because a lot of the clarity of the hihat lies up in that region too, and if I didn't chose my compressor values carefully I would smoosh the hat. In the end I used a medium attack (10ms) and a fast release (50ms) with a fairly heavy handed ratio (3:1), and set the threshold juuust right so that the hi-hat was barely getting attenuated, but the louder crashes would get pushed down by about 6db. To replace some of the highs that I would be losing due to the split-band compression, I used a makeup gain of 1.5db.
- Alright, NOW shit was sounding usable. I added in the close mics, and listened to how they all sounded together. Not enough hi-hat, so I boosted at 2.5khz by 3db (with a normal width of q=1.5) to bring it up. There was also a bit too much junk going on in the lower frequencies, so I did a high-pass at 120hz. At last, I am happy with the overall tone and balance.
- But, with the close mics mixed in, I realize that I'm getting too much snare in the overheads. I like for the snap of the snare to mostly come from the close mics. To solve this, I reached deep into my bag of tricks and pulled out the side-chaining expander. Side-chaining is when you use one track to control the dynamics of another. In this case, I used the snare track to "duck" the overheads: every time there was a hit on the snare track, the overheads quickly got pushed down by 3-6db, then slowly rise back up to their original level over the course of ~250ms. This cut out most of the snare's hit, leaving only its roomy decay/ringing sound.
- OK, last step. This just wouldn't be a rock mix without some conventional (single-band) compression on the overheads. I used a fast attack (1ms) and fast release (100ms) to tame any hits that were peaking too loud, mostly the toms and some rogue crashes. I usually say "there are no rules when it comes to mixing", but for some reason I find that a ratio of 1.7:1 always works perfectly on overheads. 1.5:1 is too wussy, and 2:1 (or even 1.8:1) is too constrained. I set the threshold so that I am getting 1-3db gain reduction on the loud crashy sections and 4-6db reduction on the very loudest tom hits.
As for the close mics, I must sheepishly confess that I did not actually use them. Instead, I used a peak-detector-to-midi converter to turn your close drum mics into a midi track. The plugin detects the volume of each drum hit, so that the louder you hit the snare in your original track, the higher the midi-velocity for that note would be. Once your drum grove was converted into a midi file, I had the freedom to trigger any drum sounds I wanted. I went with a dream theater-y kick and snare that I had, and some nirvana-y toms (which I had to tweak a bit to get them to be the same pitch as your toms). Listening to the midi-drum tracks solo they clearly sounded artificial, but once I brought in the overheads they kind of blended in and sounded more natural, albeit definitely larger than life. If I were trying to make a more realistic sounding drum track I probably would have mixed your original drum tracks back in a bit. It's a little surprising to find out that probably 99% of Top 40 radio-rock uses some sample-replacement or sample-augmentation like this. I didn't EQ any of the synth-drums at all, and instead of compressing them I just tweaked the midi file a bit to give the drum hits a more consistent level (without completely sucking the life out of them). Finally, for a little faux-room sound I added a touch of short (0.5s decay) reverb to both the fake drums and the overheads.
So, that was my experiment with this mix, to see if I could get away with replacing your drums with samples, and see what the reaction would be. Everyone seemed to like the result, so, GREAT SUCCESS! For the record, I also subtly autotuned your vocals to fix a few sketchy notes, and edited the timing of your backing vocals to make them line up a bit better with the lead vocals.
Last few notes, even though no one asked
For the bass I recorded 3 tracks simultaneously: one DI straight into the mixer, one clean track coming from a bass amp, and one more aggressive track coming through a guitar amp. So, that's actually just one bass take, but split into three signals. I then compress the fricken' snot out of the DI, and leave the other two alone. The DI and the bass amp mixed together make up 90% of my bass sound, but I also like to ride the faders up on the distorted track during the more rockin' parts of a mix. For examples, you can hear the clean combo at the beginning of the song, but right before the solo section the bass plays by itself and you can clearly hear the distorted track mixed in there.
Guitars are always the last thing I think about in a mix, and I tend to use radical EQs on them. The guitar processing I used here wasn't anything special, so I won't go into too much detail. Your guitars had no highs at all, so I did a huge (like, 12db) high-shelf boost, and used a shwack of that BBE Sonic Maximizer again, which is very good for adding "presence" to guitars that lack it. Unfortunately, BBE can exacerbate annoying treble frequencies, so I did that sweep-a-peak-back-and-forth thing I talked about (when I was looking for annoying frequencies in the overheads) and ended up notching out about 12db of hissy crap near 4 khz. The guitars still felt hollow so I added a wide, beefy boost around 1.5-2khz to try and bring out the mids. I could take or leave the results. I gave the organs a similar treatment, and didn't compress either.
I hope someone gets something useful out of all that.
Assuming funds are limited, I'd re-order that list to
- -1. Bass
- 0. Anything other than Audacity
- 1. Room treatment
- 2. Monitors
- 3. Vocal Mic
- ...
- X. Cubase
You need that bass, like, last year. Audicity is a joke;
Reaper is totally decent, extremely lightweight (2mb installer, and half of that is the project file for a Brad Sucks song that comes with it), supports all the plugins I use, and it's free*. Room treatment clearly depends on your mixing environment, but I hear a fair bit of room sound in those vocals. If you're in a small room like mine, damping the room reflections will have a larger effect on your mixes than getting better monitors will. Cubase is fully featured, but you really pay for it. Plus one of those features is a steep learning curve. Aside from BBE that I mentioned earlier (and a great tape-saturation plugin called Magento) their built-in effects aren't anything to write home about. I really think that money would be better spent on gear, at this point.
My 38 cents.
* 
