Inside the Producers' Studio. How to mix extreme metal music
Welcome back to Inside the Producers' Studio. In this series, top producers from across extreme metal share their knowledge and wisdom about recording and production. We’ve hand-picked some of the best and most interesting producers; some from large studios others from more niche circles, whatever their situation, you’ll be sure to pick up tips and get an insight into the minds that are shaping metal in the current era.
The first article was dedicated to studio set-up, the second one was about pre-recording, the third one – about recording drums, the forth one — about recording bass, the fifth one — about recording guitars, the sixth one — about recording assorted instruments, the seventh one — about recording vocals.
This new 8th article in the series is focused on how different producers mix their projects.
Dan Swanö… you know I can’t really say that I’ve ever heard of this guy, apparently he’s from Sweden and recorded a few bands. Who are we kidding!? One of the most well known producers in metal, Dan has worked with bands like Dawn, Dissection, Opeth, Marduk and Katatonia to name a select few. Contacts: unisound.se, facebook.com/unisoundmixandmastering
Tom Kvålsvoll has worked with some of the biggest names in black metal, acts such as Dimmu Borgir, Emperor, 1349, DHG, Darkthrone, Zyklon and Virus to name just a few. He runs Kvalsonic Lab in Norway and also plays guitars for Sarkom. Contact: facebook.com/kvalsonic
Christian 'Moschus' Moos has been in the recording game for many years, he has recorded and worked with a variety of bands in both metal and other genres like Avulsed, Haken, Pestilence and Delain. Christian has also worked with Dan on a number of projects and re-masterings! Contact: spacelab-mixing.com
Which frequencies/instruments do you find most challenging to deal with?
Dan Swano: I find that instruments with so much energy around 3 to 5 kHz is the nastiest stuff. Distorted rhythm guitars with too much "all strings at once" action and cymbals can be a real bitch to get to that point where it is still "edgy" enough to give you that raw 'n' real vibe, but not fuck up your eardrums. Some vocalists have nasty peaks around 3-4 kHz (myself included) when they sing in a higher pitch, and before there were tools like Spectral Dynamics or Soothe, it would drive me fucking mental!
Tom Kvalsonic: Is it wrong to say none? In the past I might have had difficulties deciding just how much of the harshness from 2,5-3,5 kHz was needed in a full-blown extreme metal mix, but not anymore. I am 100% confident with my monitoring, and I know how to avoid ear fatigue. That mentioned area is also very differently utilised from band to band, while an organic death metal band such as Execration needs a warmer sound and not too much in the harsh areas, more black metal-related bands such as Evighet and Sarkom will want more of that. How much, is a combo of what I suggest and what the band tells me they prefer after the initial mix. Much of this can be further honed in during mastering, as long as the balance in the mix is right.
Christian Moos: Hi mids around 4-5k where the guitars are killing everything else. Also, treble/presence (cymbals!) because after a while the ears get tired and you tend to lose objectivity here, that’s why you should mix at low levels.
What is your best tip for getting the easiest mix to work with?
Dan Swano: Know your speakers! If you are at a new studio, make sure to A/B to a really good sounding record, one that you know the sound of, each time you need to set a new sound. Just make sure what you consider being "album quality tones" is somewhere in the ballpark of a "ready CD" in terms of the amount of treble etc. The more everything sounds like a mixed record already at tracking, the less unnecessary overdubs will need to be made to compensate the fact that the band doesn't get a boner from what they are hearing with the basic drums/bass/guitars thing. It's also easier to set a bass and lead sound that works with the guitars if the guitars really sound super-close to what you want as a final sound. 99% of the time the rhythm guitars have too much low-mids and not enough treble than the band later want for their mix.
Tom Kvalsonic: Please clean up your tracks before sending them to me. Remove what's not supposed to be there. The exception is hiss. If you have a constant hiss on some tracks, leave it there so that I can clean that up. Don't clean up the drums. I prefer to do that. Probably needless to mention, but anyways: All tracks must have the same starting point. If you have particular ideas about the mix and effects, send some wet tracks with your suggested ideas and even a rough mix if possible, makes communication and a shared vision easier.
Christian Moos: It’s not the mix that makes a song a hit. It’s the song! Anyway — in case you record everything yourself, just make sure to always track parts as perfectly as possible, especially the more complicated it gets. Many bands record very sloppy tracks — and in the end, when everything comes together it’s a complete mess. BUT, don't over-correct everything: try to keep your song alive!
Mixed and mastered by Dan Swanö
How much of your mix do you do in mono?
Dan Swano: I constantly flip to mono and also listen to the "side" only, using my TC BMC-2 controller thing. I have a pair of small, passive, one-way, Yamaha Hi-Fi speakers standing next to each other, which is like a "super small stereo", that helps me to emulate that new "Bluetooth speaker" thing where you have stereo, but in a very short space. I sometimes also listen to the mix in mono in my phone's built-in speaker. Gotta love the low-end punch in those things...
Tom Kvalsonic: Many engineers say that to get a balanced mix they prefer to do most of the initial mix in mono then spread it after the mono signal is coherent. I can understand that, but I actually rarely do this. After getting a general impression of all the tracks featured in a song, I prefer to set up a soundscape in my mind and mix from that. I have a vision of where I want to go quite early in the process. So if I want a certain guitar to be, say 65% left, I first get the sound of it in mono then place it where it should be immediately after that.
I do of course check mono compatibility and try to avoid the "big mono" syndrome.
Christian Moos: I barely use the mono button, because I actually know how it should sound in stereo to also work in mono. I have to say that I only mix on monitor speakers and never on headphones. I actually hate headphones — contrary to Dan Swanö for example (hi Dan!). But I may use them at the very end of a mix just to make sure that there are no problems regarding the left-right balance of certain signals, but only for that short moment.
When starting a mix which instrument do you start with?
Dan Swano: It depends. There have been times when I have started with the drums, but lately, I have been working a lot with finding the perfect rhythm guitars + bass "wall" and then make the drums "sit" in (or maybe on) that wall, because drums tend to sound one way when soloed and completely different when "filtered" through pretty much a hi-cut, slightly scooped white-noise wall of sound! The amount of EQ and compression you need to make drums sound cool in the "environment" is sometimes insane, but after watching Chris "God" Lord-Alge mixing his drums with really nasty boosts and spanks, I felt like..."Oh, so that is how he does it too, then I guess I am not completely lost after all!" ;)
Tom Kvalsonic: The drums are the skeleton, and the instrument which takes the most time to mix. I start with them and other percussive instruments, then it's the bass. Then I move on with guitars, synth / piano / keys / strings / brass / other instruments before I do the vocals.
Christian Moos: Doing a mix is like building a house for me: I start with the cellar (drums), then comes the first floor (bass) and so on. Then I add either keys or rhythm guitars — depending on the style of music. When the basics are done I add lead guitars; solo stuff and vocals come at the very end. Of course, I always tweak older elements as new instruments are introduced.
What tips would you give aspiring producers trying to get the best sounding mix they can?
Dan Swano: Get a pair of Yamaha NS10m Studio and set them up in a way where most records sound good through them. I use my NS10's standing up, and I sit pretty close to them, which works best for me. A subwoofer that can be turned on or off to check what's going on below 100hz is a good thing, but try to get your mix to sound mega-awesome without a sub first. These speakers make you focus on let's say 160 to 8khz. Some of the super subs and that super shiny treble will fool you to believe that a sound is super ready when it really isn't. I just love the positive shock I get whenever I have a mix going in the NS10 and switch to my very Hi-Fi sounding headphones, and it just sounds so much better than I ever thought possible. Never really had that experience working with "friendlier" monitors.
A/B a lot to make sure you are close to a reference album in terms of overall EQ balance. It's easy to mix with too much bass in NS10's at first. It's also good for letting the ear hear something else than your mix. Kind of "erase" the aural memory a bit and when you return to the mix, it sounds different and you hear new things to fix. And of course, tune the instruments before recording. Make sure it is as tight as can be. Any clashing notes or bad playing technique (such as the ringing of "the wrong string" during palm-muted etc.). I could go on forever…
Tom Kvalsonic: Be humble. Learn. Practice, practice. Be critical. If you get stuck, try a different approach, maybe try find some clues to how others are doing things.
Christian Moos: Forget that saying "we'll fix it in the mix". Just try to record everything right from the very beginning of your recordings. If a take is crap, delete it. Don't keep every bad part just for the sake of it or "just in case". Learn to make decisions! Judge a mix on the next day, because when you hear it with fresh ears, you will find some issues instantly!
Recorded, mixed, and mastered by Tom Kvalsonic
Which instrument do you think is the most problematic to fit into a mix with extreme music? And what are your tips for resolving it?
Dan Swano: In extreme music the drums are a bitch because they tend to go from "let's hit this one super hard" during the heavy parts "and let's hit it as soft as possible" during the blast. No other genre has so much difference in the volume coming from the drums.
Before I start triggering stuff, I try to get the acoustic sounds to behave and have a good punch and pretty much sounds like they were samples, but with all the life and vibe of an acoustic drum. One trick I can share is to split f.e. the kick drum into 6 groups.
Each has a multiband with a "non-ringing" cross-over on it. I use the DMG Limitless for this. I find that crossover mega! Then you set it up in bands like lowest rumbling bass, chesty punchy bass, ugly low mids, ugly mids, low click area (2-5k) and high click area (5-12k).
First, just view the energy of the various frequency bands and try to set a cool balance, using the faders as an EQ, then insert a noise gate on each band, each triggered from the same "mother kick" using side-chain, set the release time to around 300 ms for the "non spill stuff" and the top frequency bands can have a really short release and hold time; like 5-10ms, then insert a DrumLeveler from SoundRadix on each band and after a while, you have found a way to limit the dynamics of each band so it cuts through the mix and "stay the same" for pretty much whatever the drummer is playing. Excellent for "ironing out" inconsistencies between left and right foot on one kick with a double pedal. You clearly see the difference in volume between the left and right foot in the "treble bands" of the setup and the smart thing about the drum lever is that you can choose to only raise the volume of the softer hits to the level of the harder ones, without actually "touching" the hard hits at all! Not many compressors can do just that! This trick works fine for snare and toms too, even for cymbals!
Tom Kvalsonic: Ideally, none. The challenge with extreme metal can sometimes be that there are a lot of elements "fighting" for space. Make room. Have a vision and go from there. I have received raw tracks from bands that wanted me to do the mix, and in one of these, their original mix engineer told the band, "This synth has to be removed because it's frequencies argue and get in the way of the guitars". Nonsense. Humbug. Separation is the key. Stereo placement. Width. And, of course, frequencies. Although many of these frequencies were indeed almost in the same places, I found it easy to get this particular synth separated and at the same time fit naturally in the mix. My starting point: The very timbre of a synth is quite different from any guitar, be it distorted or not. I was already pleased with the guitars, so I did not change the guitars one bit to accommodate the synth. (OK, maybe, I don't remember, but maybe I lowered some guitars a dB or max 2, but no eq changes, no more compression). I ran the synth through various saturation, tubes, gave it a unique place in the stereo field, and voila. No problem.
To further elaborate on this question, any "mellow" instrument (viola, cello, a soft female voice) requires special attention in a metal mix. I don't want to drown these elements, but it's also important to make sure these elements don't sound artificial, over-processed or simply louder than they should be: They should be audible and have their unique soundscape, but never challenge guitars, drums, and extreme vocals volume-wise.
Christian Moos: Haha, drums! In case you use all midi, most people quantize everything or program ridiculous patterns or fills, it sounds like a machine in the end (I am a drummer so I hate drum machines anyway).
When people record a real drum kit (and they always should prefer this), it boils down to 3 questions:
- is the drummer capable to play everything tight
- can you tune a kit and do you know shit about drumheads
- how to set up drum mics.
You have to know how to use compressors, limiters, reverbs and sidechain compression to make a drum kit sound cool. Not to mention, that an acoustically treated professional recording room will always beat your (muffled) rehearsing space.
Record the (acoustic) drums at a good studio, and also have them edited there (timing correction, sound). Use drumheads that fit the music you play! Tune them perfect, take your time for the mic setup, because it they turn out crap, you have to replace them with samples later…
How do you pan the guitars in the mix?
Dan Swano: Usually Full L and R when both are playing, and when there is a break with a single guitar in one speaker, I like to back off a few % so there is a little bit of guitar also in the other ear when you listen in headphones.
Tom Kvalsonic: How many guitars? A mix can have anything from 1 to 50 guitars. Roughly speaking, if there are two rhythm guitars, I prefer to balance them to the sides. Never hard left and hard right. Remember phase issues and mono compatibility. Leads closer to the center, not always dead centre. Sometimes the mix will sound more interesting if a lead is leaning slightly to one side. The way I stereo enhance guitars is also important. I do that with (almost) all guitars, even the ones that are panned, to make them fatter. With leads I might even have an LFO going on the stereo to make it come more alive.
Christian Moos: I always refer to PAN positions according to knobs on an analogue desk: hard left is 07:00hrs and hard right is 17:00hrs. I never use % or something here because my mix happens analog on a desk and that's where I define the pan positions. Do the math yourself :)
In case there are 2 rhythm guitar tracks I pan them 09:00 and 15:00. If there are 4 tracks, I keep 2 on 100% left and right. The other 2 I pan 10:00 and 14:00 and dial them in to make everything sound more mono again. These are rules of thumb and I never work within fixed parameters.
That's the perfect moment to check with a mono button or headphones! Try to use the same amp sounds for left and right, to avoid problems on the headphone. But you may blend different amps of course. But avoid phase issues!
Engineering, mixing, producing by Christian Moos
How do you bus your tracks?
Dan Swano: I usually end up with 1 bus for each drum and then 1 bus for all of those buses.
Sometimes a bus for parallel compression, if none of the compressor plug-ins with a "mix" is doing what I want it to do. Then I have rhythm guitar buses, and all of those get one "all rhythms" bus, bass has a bus too. Solos have a bus too, and I usually side-chain the rhythms to be a bit lower during the solos. The fx of the guitars have their own FX sends, so the tails are not triggering the side-chain ducking. Vocals have buses too. One clean and another for growl (if that is the musical case). I very rarely have anything on my master bus. When I do, a soft touch of M/S compression with Voxengo Soniformer in a 50ms attack/500ms release 1.5:1 ratio setting, just shaving off a dB or two when things get really busy! If the band send a reference CD with a very "pumping" bus compressor approach, I join in and slap a UAD Fairchild 660 and a UAD FatsoSr, both with very high values on the sidechain filter so it works more like the ear and the snappy attack of the Fairchild and a slow attack around 30m (on the Sr Plug version that is possible) and both just working a bit, it really pumps stuff up in a super cool way. I set the Fairchild in the LAT Vert mode for a relentless smack in the Center channel. Doesn't work on all music, but when it does, it's amazing!
For some albums with a lot going on, I tend to group "the hard rock band" and "the orchestra+synth band" separately and the vocals also separately so I can sidechain compress certain things with other things to create sonic space for stuff to work better in.
Tom Kvalsonic: Varies from project to project. Often I bus the rhythm guitars and lead guitars to separate busses. Depending on how much the audio spectre is filled, I might even split that up more. Drums: I never run all of the drums through a single buss for compression and eq, only for gain staging. I have a firm conviction that many albums would sound a lot better if mix engineers would refrain from using buss compressors on every buss! That is ruining it for the mastering, it can ruin an otherwise great mix. Maybe I am being this critical to compression because of my background: I have been mastering for over 20 years, and I know what I am talking about when I say that buss compression often does more damage than good. If you cannot balance and make impact with the drums without a bus compressor, you are novice. Back to school with ya. And by that I don't mean school. By that I mean every engineer is best served figuring out things for himself. You can get so much wrong advice at schools, it's insane. I do, however send all the toms to one buss and process that as a unit. Same with overheads, ride and snare. The kick and snare I never buss with anything. I have better ways of making them fit in the mix.
Vocals: varies. If it's just a few tracks, all in one buss. But 90% of the processing is already done on the strips themselves. If there are backing vocals, clean vocals, those go to separate busses.
Bass: on its own. I sometimes make one "clean" and one dirtier or distorted and blend these accordingly.
Christian Moos: I am old, I’m coming from the analogue days! So I work very old-school!!! All digital tracks are partially summed in the DAW and then assigned to 24 DA converters; these are routed to my analog desk.
Summing happens all analogue. Like this, I can use some very nice 19-inch analog devices like Urei 1178, SSL X-Logic Bus-Comp, Klark Teknik DN500, and some SPL made analog stuff.
2 Lexicons are in use (PCM80 and LXP15-II) as well as a t.c. M3000. All Drums are mixed from 6 analog channels (coming from the DAW). I do parallel compression on the drums using the 1178.
On the desk I have my channels organized like this:
Drums: 1x kick-bus, 1x snare-bus (just mono busses panned to center) 1 stereo bus for all toms, 1 stereo bus for all overheads/cymbals. I also route the entire kit to a stereo bus on the analogue desk where I add the parallel compression using the 1178 on 20:1 (compressed-to-fuck).
Bass: 1 mono bus
All rhythm guitars: 1 stereo bus
All lead guitars: 1 stereo bus
All clean guitars: 1 stereo bus
All keys & orchestrations guitars: 1 stereo bus
Plus 2 stereo choir busses, one special mono bus only for lead vox, 1 stereo buss for DAW fx (like reverbs or delays). Then there is the master output from the DAW on the last stereo bus. Makes 24 altogether.
So a mix is not "rendered" in-the-box because it happens all analogue in realtime and is recorded using my RME’s. And the mix is sightly processed with the SSL before its digitized. That adds some "glue".
By Dan Thaumitan