An acquaintance of mine was recently mixing a contemporary rock record, and sent me some draft mixes to get some advice. He had some really good tones going, and the musical balances made sense, but overall the mixes were getting a little crowded. After I wrote him back with my thoughts, I realized the advice I was giving him was general enough that it may be useful to some other students of the craft out there who can make great sounds, but have trouble getting their mixes to hang together as a unit. I’m glad I had the chance to write the email, because it sums up a number of ideas that have been vaguely kicking around my head for the last couple years. Let me know if these ideas are useful to you, or please ask any questions you may have. Cheers.
The main thing I’m hearing is that it sounds like most things are competing for the same space, both frequency-wise (high mids and highs) and dynamically (the “compressed” or “sustained space”).
As to the first question (frequency space), I get the feeling with some of the mixes that you are boosting the drums in the high mids (or cutting the lows) to get them to punch through, and then boosting the guitars in the high mids/hi end (or cutting lows, etc) to get them to cut through, etc on down the line (though it’s not as exaggerated as I make it sound, because I think you’ve done a good job sitting the bass in the low end). In particular, the drums sound thin to me, but then I compared to something off “Songs for the Deaf” and realized the drums on that album may be even “thinner” in some respects – but they are complemented by guitars that are very focused in the mids and low mids and don’t have too much hi frequency energy. Conversely, you could keep the guitars focused the way they are (cause they sound exciting) and allow the drums to sit more in the low mids, etc.
Keep in mind when dealing with these frequency issues: a) I never bought into that “rule” of boosting freq X on instrument A and cutting freq X on instrument B – sometimes, both instruments need a boost or cut at the same frequency and still will complement each other. b) No instrument lives in just one frequency range, in fact most have important information through most of the bandwith. It’s just about highlighting certain areas where it can come through with more definition or body and de-emphasizing areas where it clouds the mix. c) Be very careful when boosting or cutting that you don’t create a sound that is so scooped that the lo and hi range of the instrument are no longer “connected.” This is especially a danger when EQing in solo (though I do that all the time of course). For me when the midrange information is so deemphasized that you no longer hear it in the context of the mix, the low information just becomes random boominess or woof – you need the midrange to connect it to the sound so it serves as actual body. This is also an issue if the freq range you’re emphasizing in the low end is too low for that instrument.
If you do tend to EQ in solo, I highly recommend this technique: After you finish working in solo, put the sound back in the context of the mix, and bypass the EQ. Now put the EQ in. Is the mix better or worse? Rarely is it just better (in that case, move right along!) or just worse (zero the EQ and start again), it’s usually better in some ways but worse in other ways. Ask yourself what you like better with the EQ in and what gives you pause about it. Maybe you only need one or just some of the bands you’re using, and can zero the others. Maybe you just need to less boost and cut on all the bands you’re using. Maybe a frequency you cut actually needs to be boosted, and vice versa. Test out these possibilities and make your final decision listening in the context of the full mix. If you can’t decide, bypass the EQ and keep working on the mix, and eventually you may just hear in context, “you know I think I DO need that boost on that snare drum.”
The second question (dynamic space) is something that I learned from mixing live sound where I usually only had 1 or 2 compressors if any. A good mix is a balance of different frequency ranges AND different dynamic ranges; a mix of more sustained material with more transient material. Live I had to find the nice balance between the punchier transient material and the more sustained material, and I had no control over which was which. In the studio you have all sorts of lovely dynamics processors to change the dynamic shape of various sources and make sustained material more punchy, and transient material more sustain-y. The danger, especially given the contemporary drive for “LOUD,” is turning every instrument into a sustained pad, and robbing the mix of exciting punch and dynamics. In particular you’ve got to be careful with how you compress drums (both in amount and type of compression), because they are a very transient instrument. It may seem that in order to get a louder mix you’re going to have to compress the drums hard, because they are so transient they’ll eat up all your headroom. I find a few things happen when drums get too compressed: a) they start to lose tone, b) they don’t punch through the mix, so you either end up turning them up too loud, or c) they lose their attack, so you try to EQ the attack back into them and d) because they don’t punch through the mix, mastering compression and limiting tends to make them disappear.
My general advice for compressing drums:
a) on individual drums, try lower ratio compression with slower attack times and faster release times. If a drum is too big and woofy, try increasing the release time. If a drum is getting lost, try bypassing the compressor – it may punch more without! Or try adjusting the attack time or threshold.
b) be very careful with drum buss compression. If you’ve got a heavy-handed drum buss compression that you like, try it in parallel to an uncompressed bus and adjust the balance to favor the punch of the uncompressed bus, with the compressed bus bled in for excitement.
c) in some rock mixes, the overheads may sound better uncompressed.
d) in general, try not to be heavy-handed with both bus processing and individual channel processing. Try to compliment heavy bus compression with light (or no) individual channel compression, or vice versa.
e) be careful when setting up your drum mix to allow some headroom in your compression scheme. For example: you set up some a cool sounding drum mix, with some moderate compression on the main drum bus or a parallel bus. You move on and get the rest of the mix thumping, and your snare is no longer cracking, so you turn up your snare fader, but it doesn’t pop through any more than it was, so you turn it up louder, and EQ it some more, and compress it some more, only to get slightly more volume and a steadily degrading tone. The problem is when you turned up the snare channel it just started compressing the drum bus more. This is less of an issue when using parallel compression instead of compressing the main drum bus, but depending on how much you were relying on the parallel bus for your punch it can still be an issue there as well. Sometimes it makes sense to have a group made up of all tracks, but not busses, that you can trim down just slightly to reduce the overall amount of bus compression on everything (or trim up and increase) – you can move the fader just a little bit and hear a dramatic change as your mix alternately opens up and saturates, and find a sweet spot.
f) all rules were meant to be broken!
Hope some of this is helpful.