Smash the lead vocal. Depending on the situation, I don't feel bad smashing the crap out of the lead vocal. I'm talking about sometimes 12-20dB of reduction. I'll use the fastest attack and the fastest release I can. Hitting the lead vocal pretty hard can help recordings a ton. There is a point of too much, but in this case, I usually feel safer with too much than too little. Experiment with it. I don't have many plugins so I usually go with the Waves Rcomp. I say the EMI Compressor / Limiter turn a vocal that needed to recorded into a vocal that was worth keeping.
Parallel drum compression. I always send drums to Bus A in Vegas. I like to toss a compressor on Bus B. I'll solo that bus and send the kick and snare top to it. I all out ass rape the tone. The mega critical issue here is the attack time. If you let just a little attack through while absolutely bone crushing the rest you can increase the crack in a source, but you most be careful with this method because some cracks will be way louder than others. I sometimes will use a limiter after this to catch the louder pops. This technique is pretty much required for rock drums.
Deathcomp ? I invented (more like discovered) the deathcomp technique myself (probably because no one else would ever want to use it). Anyway, you put the threshold of a compressor on the most aggressive setting. (maybe -60dB). Then you boost the output by a tone (usually at least 20dB). Then you play with the ratio to figure out the level you want. I run almost everything through this when the mood strikes. I mean I'll run kick, snare, toms, bass, guitar, vocals, etc. You have to be very careful because it can be a total nightmare. When done right, it can totally thicken up a track. It could destroy a track much easier. This is extreme and stupid. I'll usually send a soloing instrument to this thing as well. It really helps fill out the solo and makes it ?more there? without actually being louder.
Smash the guitar ? Sometimes I can't get big electric guitars to sound right. It's a frequency thing. I'm not sure what it is. I've found that sometimes hitting the guitars with 6dB of reduction or so will make a tremendous difference in their tone. It's actually sort of subtle, but in some ways the difference between shit and something usable. It has a way of bring out the ?grind? in guitars. Note: This is only used if no compression didn't work for me.
Smash the reverb send ? I'm not going to lie. I don't have a great drum room. I'm a broke ass like everyone else. My drum room is vomit times 7. To simulate what the big boys are doing, I will create a room mic. I do this on an Aux with a mono reverb (I guess stereo would work too, though). I mess with the sends on various tracks until I get the the solo'd reverb to sound like a good room mic should. It should sound like a drum kit X distance away. Then, I smash that pretty good with a compressor. I play with the attack times a lot on this. Depending on how much attack I let through can make all the difference in the world. You have to be careful. If you over smash this, you'll get too much cymbals and not enough snare in your room sound. Then again, that's part of the reason I compress it to begin with.
I'm lazy ? That's right. If you have a saxophone, an acoustic guitar, a clean electric guitar, keyboards, and a million other things going on who has the time to ride the level for the entire song (for 16 songs). Just hitting each source with 2 or 3dB on their loudest parts will make a huge difference in the making everything fit in the mix. Sometimes, you can not get this tone by simply riding levels. Note: This is not implying that you should be lazy. It actually brings up issues with local bands on mega tight budgets recording way too many songs. Compression, if used properly, can save you a ton of time.
If a tone doesn't sound right ? If I don't like something about a certain tone, the first thing I'll try is hitting it with 3 dB of reduction. This usually makes quite a difference in the tone and gets it close to where I want it to be. If not, oh well. It's always worth trying.
Smash the bass ? Sometimes bass needs next to no compression. In that case, I give it next to no compression. Sometimes bass needs an ass ramming. I don't feel shy about hitting it with 6dB of compression if that's what the song calls for. Get aggressive. It's fun.
Gate on the Drums ? Compressers and gates are the same things only backwards. Actually, you can create a gate with a compressor on drums. Just hit the drums with aggressive compression and slow down the attack time. You can let the attack through and reduce the hi-hat bleed, room sound, or whatever.
Using it on ?pretend mastering?. If you are reading this, you are probably no mastering engineer. If you are actually a mastering engineer and you are reading this, you have been demoted. Sorry! I don't know what I'm doing when it comes to mastering, but my clients expect something that sounds mastered. These days I'm using the trusty Rcomp. I like to set a fairly long decay time (maybe 200ms) and an attack time of around 20ms or so (but depends on the song and mix). I usually set it to reduce about 3dB during the peaks of the songs. Once again, this is what I do. I'm confident there is a better way. Feel free to enlighten me. Update, now that I'm reading over this 6 months after I wrote it, I note use a Waves C1 and I set the attack and release to be as fast as possible. You must be very careful with this, but it gets me closer to that mega loud sound the kids always want.
When reverbs get cheesy ? Sometimes you'll mix a tune and the reverbs won't be right. They may pop out in places that they shouldn't. You have to be careful using a compressor in this situation because it could make the problem worse. In this situation, you won't be bumping up the compressors with make up gain. You are just going to knock the peak 2dB or so off the reverb send. This should help tuck the loud reverb stuff quite a bit to avoid the sound of your favorite 1991 production. (Not that there is anything wrong with 1991 production).
Compress delays ? Just like the reverbs, sometimes certain delays will jump out a tad too much. In order to keep them ?tucked in? you sometimes just need to knock a little off the top. Sometimes, bring up the later delays is a good thing is well. In other words, use the compressor to knock down the loudest delays and bring up the quieter delays. Again, be very, very careful with this one. It can make the problem worse in a hurry.
Distortion is compression ? Sometimes, I'll put a distortion plugin such as Anteras Tube on a vocal or whatever. This will help level out the vocal quite a bit. There are times when clean compression isn't enough. Sometimes you need a little more harmonic content in whatever your doing and the distortion can be sort of the best of both worlds. Of course, a very aggressive compressor will add distortion too. Experiment with both.
Put distortion in front of a vocal reverb ? I've had cases where the vocal reverb sounded way too clean or boring. I didn't want to put distortion on the lead vocal, but I wanted to add some grit or ?harmonic content?. What should I do? In a couple of cases, I've had great luck putting a distortion plugin in front of my reverb on an aux send. It keeps the lead vocal clean, but adds ?stuff? to the reverb sound. Try it. You'll probably hate it 90% of the time.
Going back to my Deathcomp idea, sometimes you just need to totally destroy an instrument with compression and mix that in with the original signal. For example, a guitar solo is sort of being lost in the mix, but you are not sure why. Well, before you tear your hair out, try running it to an aux and smashing the living crap out of it. I'm talking 20dB of compression at least. Blend that in with the original track and see what happens. You may be surprised to find that it makes a huge difference in a very subtle way.