More recording thoughts - eq and panning
posted on 06 Apr 2007 under category Meta
As I finish the mixing of the Venom Tribute tracks and dig into the mixing on Bellum Aeternum, and in talking to people in the wake of my earlier ‘advice’ article on EQ, I’ve had a few more thoughts. Some revisit the EQ topic, but others move on into the area of panning, which is a topic that can be delicate but often isn’t given much thought. First up, EQ:
I got quite a few feedback posts on newsgroups after publicizing my earlier EQ article. Most were positive or commendatory. A few had other, more helpful ideas that I’d like to pass along.
First, when it comes to EQing, you can do two things with each band - boost or cut. In the examples I gave, I boosted frequencies to give the new ‘EQ shapes’ to the instruments. Technically, this is improper technique. As one of the follow-up posts eloquently said, “use boosting to change the quality of sound, use cuts to make it fit”. Of course, they said it more eloquently than that, and managed to convey more of the essence of why.
Basically, you can’t boost frequencies that aren’t there - you end up adding noise. Also, you can introduce phasing issues. Think of a Wah - basically, a Wah is doing what I did with those EQ curves - it takes a narrow point, boosts it way up, and cuts the rest. That frequency point may or may not match what is playing, so you’re pushing up different partials than what is going on - this is how a wah gets its sound. However, if you’re doing this on fixed bands, for entire tracks, its easy to see how you can inadvertently get yourself into trouble.
Instead, use the same shapes, but move the entire curve DOWN, so that you’re not adding anything, and are instead cutting out what you don’t want there - sort of like a sculpture.
Next up - Panning:
Between the second mix of “Bloodlust” and what I ended up doing for the final mixes, I changed the pan of the instruments. As I did so, I thought of some of the comments I saw regarding my mixes - comments which were working from the assumption that I was doing a ‘traditional’ pan-out of the tracks. Perhaps it’s best to start there. Since most people do different numbers of tracks, but use basic algorithms for their own works, I’ll describe my method and you can translate to your own.
I use a basic method for a standard drum-bass-two-guitar instrument mix. My drum software automatically pans the drum pieces so the snare and kick(s) are in the middle, hats are on one side, ride on the other, two or three crashes arrayed randomly left, right (center for a third, hard right/left for the china), and toms descending in one direction or the other. Bass goes straight up in the middle. Rhythms get panned left and right by equal amounts. For leads, if it’s one lead, put it in the middle; if trading off, pan them left and right as well.
But how much? And Why?
Things that are constant in the song like bass, kick, snare, and vocals, generally should be put in the middle, since that is where the ‘focus’ of listening attention is, and most people when listening subconsciously focus on the vocal lines and the kick-snare pattern that drives the rhythm of the song. Bass goes there because there’s just one bass track (usually), and it’s effect should be spread over the entire song spectrum. The reason most metal guys use two guitars is the sonic reinforcement they give each other when playing in unison. However, putting two guitars on top of each other in the middle hides a lot of that unifying power. By putting some pan distance between them, it actually sounds stronger because that reinforcement is ‘spread out’ over a greater area, mixwise, than just sitting in the middle.
The catch comes in that, if you pan too much, they don’t have a lot of reinforcement. See, think of ‘dead center’ not as a single position, but a ratio of two numbers - left-side volume and right-side volume, in percent. In a center pan, that’s 50-50. If you pan partway right, it’s, say, 30-70. The sound sounds right because the right side is louder, but you hear some bleed over in the left channel.
Now imagine that the two guitars are even as I said above - 70-30 on the left and 30-70 on the right. You still get a 30% bleedover for reinforcement from each side on the other, but each side still has a clearly dominant single line. You get some sonic space and the reinforcement. But, as you pan farther out, the amount of bleed-over decreases, both because the dominant amount is getting larger and the bleed-over from the other side is getting smaller. Even just 10 percent more, 80-20 and 20-80, you get a DRASTIC decrease in reinforcement. Instead of reinforcing each other, they sound separate, and the subtle differences that cause that reinforcement instead distract from each other, since they’re not working together. Obviously, there’s room for experimentation and personal tastes, but for basic mostly-doubling/occasional harmony rhythm/basic parts in my mixes, I try not to vary more than about 30-40 percent away from dead center.
This becomes even more important when you’re doing a Power-Trio type of mix, as I did on the Venom tracks, and even earlier when I was recording Festering Sore all those years ago. When you have a live situation, or a band who likes to do the studio work in the live style, where there is just one guitar and bass, the differences in instrument tones and frequency responses must be taken into account as well. Since it’s not two things basically doing the same thing, you can do just ‘fine’ by panning them dead center, but you still leave yourself sounding a bit ‘empty’ because there’s all that side room going unused. In these cases, I do a bit of panning between guitar and bass, but I try to keep the split at 25 percent or less. Since it’s just two instruments, of different tonal characteristics, they need to be closer to work together, and even a small spread sounds larger because of that. Also, you won’t have that bass in the middle to ‘bridge’ across your sonic spectrum. A little more trial and error is in order, but do try a bit of pan spread rather than straight-up when doing this kind of a mix. (Just don’t go overboard - I think a lot of the criticism that Venom’s “Possessed” album gets is because of the ‘muddy’ sound, which is really a good guitar sound and a good bass sound, just panned too far apart to support each other.)
And now the ‘however’…
However, there are times when a wide pan spread can be used for neat effects. I’ll share two examples where I did basically the same thing, but in different applications. Remember how I said that, in a wide spread, the parts don’t reinforce each other, because they don’t bleed over to each other? Well, the way that works is, in principle, not that different from a chorus effect. “Chorusing” is the way that two parts independently try to do the same thing without feedback from each other to ‘correct’ toward unison. Think about a vocal chorus - several people, each trying to hit exactly the same notes while singing live. It instantly adds some ‘space’ to the sound, since not everyone’s timing is perfect and their pitch might not be, but, if most of them are reasonably on pitch and on time, it makes it sound much bigger and fuller than if just one person were singing it perfectly.
Well, you can do that with instruments as well, and multiple tracks.
Listen to the clean guitar intro of Death Beast’s “Psychosis”. I wrote that originally as one clean guitar part, playing a low, powerchord-based arpeggio pattern. I wanted it to sound bigger, so I laid down a second part, playing exactly the same pattern but an octave higher. I did my best to double as exactly as I could. It sounded neat, somewhat like a twelve-string guitar, since that’s basically what I was playing between the two guitar parts. Then, I got this weird idea - what if I pan those parts? So I split them, about half-half, and listened, and it sounded bigger. Then, on a lark, I split them fully hard left and hard right, and listened again. It gave the effect you hear on the album. It’s almost like you’re in the ‘middle’ of the guitar, and though it gets lost when the distortion kicks in, that early part has a spacey quality that I really like.
A few months later, when I was recording “GHB”, the instrumental intro to the new Rampage EP, I got the idea to try this again. The intro starts with with distorted guitars, bass, and drums doing a rhythmic accent-type of a pounding intro. As a melody line, I came up with a singlenote guitar riff that plays over it until the regular song kicks in. I recorded that line once with a chorus setting on, then wasn’t sure I liked it, so I recorded on another track a setting with a touch of delay and the wah put on, but left in one position. I intended to A/B them to see which sound I liked better - and then I thought about the “Psychosis” thing I did earlier. So, I panned them hard left and right, then started it. As it happened, I doubled the lines just about perfectly, and so the differences in tones between the two guitar tracks really gave a weird effect. It’s kinda like a chorus, but different, and hopefully will give people who don’t read this something to scratch their head over.
So, in general, for rhythm parts, try to keep your pans from going too far out to the side, but do experiment with very wide pans if you’re looking for some new take on an old effect. If anyone else has any tips or ideas on wacky things to try with panning, let me know.
(Another of those gems beyond price, the nuts-n-bolts of process and the snapshot of where I was back then, combined in one perhaps-a-touch-too-long post. I didn’t expect “Misogyny II” to be so long in getting completed and released, which is why I sound like you’d all be hearing it soon.
I really need to figure out what to do about that.)