Mixing Metal

Updated: Apr 5, 2019


 

I haven't had a lot of experience mixing metal music and I love metal, especially death metal at the moment, so I wanted to get some practice in so that if I had the opportunity to mix or even record some metal bands in the future I'd be ready for it. I grabbed a free multitrack stem pack from Cambridge Music Technologies that fit closest to the style of metal that I've been wanting to mix and start to look for a reference track. To be honest, I would've wanted something that was a bit more in the vein of bands like Cannibal Corpse, or Cattle Decapitation, but I found something that still had similar elements like double kick, heavy guitars and growled or screamed vocals. I had been listening to a lot of death metal before I began mixing this track so I had a fair idea of what I wanted out of it. The main things I had in mind were to make sure the guitars filled out the majority of the mix with a nice chuggy and chunky sound in some sections and that proper ripping fuzz in other sections without swallowing up important musical elements like the vocals or drums, without just having the vocals or drums louder in the mix. The most daunting thing about mixing and probably recording death metal is how precise the mixing is; with so much going on, with such gritty and fuzzed tones, it's hard to get everything to pop through appropriately. After taking a listen to the track with all it's elements in place was find a reference for the kind of mix I wanted to achieve. Evisceration Plague by Cannibal Corpse kind of fit the guitar tones I was working with so I decided it would be my best reference.

And for reference as well, here is the original multitrack session without any processing, levelling or mixing of any kind.

The first problems, or issues I would have to address, that I noticed were the guitar DIs and how unnecessarily loud they were, and the plosives that were riddled throughout the vocal tracks. But before any of that let's get to the first stage of my mixing process.

 

Levelling

After organising all my tracks into groups, my first step when mixing is always getting the levels right, and there a couple of ways I will approach this. One technique is using pink noise to set levels. This technique involves loading a noise generator that produces pink noise, turning down every track in the mix and then bringing them up in level until you can only just, barely, hear that element above the pink noise. The reason for this technique is best described by Eddie Bazil in a Sound on Sound article titled Mixing To A Pink Noise Reference; "A more ‘human-friendly’ distribution of energy results when we filter white noise so that it loses 3dB per octave (going upwards). Each octave then contains the same amount of energy as the next one — so it sounds naturally ‘balanced’. Well, this is precisely what pink noise is." So that's what I did straight away, turned everything down, turned down all my tracks and then brought them back up one by one until they could just be heard. This definitely gave me a decent idea of where everything should be but wasn't my finally step in mixing. I thought about my initial idea about wanting to hear all my elements clearly without having anything too loud and noticed that the vocals were sitting pretty in line with the guitars and weren't really in a good spot to be heard clearly, so I moved them until they could be heard just slightly more than the guitars. I didn't want to bring them all the way up because I wanted to later EQ them to suit each other rather than boosting one above the other; I didn't really want to boost my vocals as they were because I might lose the balance in my mix. Everything else kind of sat where I wanted it, however there were way too many vocal tracks all doing the same thing, so I adjusted them so I had one main vocal and some backups (because the vocals are pretty breathy I wanted to fill them out a bit, I even duplicated one of the main vocals so I could fill it out a bit more), and a stereo vocal for a later section.

Stereo vocals for a later section in the song (left) Vox 3, Vox 4 and Vox 4 duplicate are main vocal while Vox 5 and 6 are fillers for a later section.

You can see here the 3 main vocals (including duplicate) and the other filler vocals (above and below main vocal)

Regarding the loud DI tracks for guitar and bass, I figured they would be more of a foundation for their respective instruments, so for the guitars I turned them down and left them mono so that they would sit under the distorted tracks, giving them a bit more body in the middle, while the distorted guitars are panned hard left and right to give it a spacious effect. If you consider what you're hearing on stage when a band plays live, if there two guitarists, you're hearing from either side of the stage, so for this reason I usually give any doubled guitar tracks a hard left and right pan. For the bass I left the DI track fairly loud to get a nice clean sub tone through, and then mixed the distorted bass in for a bit of flavour.

The main guitar section is split into two parts between the distorted and DI, and a single part for the bassFX (distortion) and DI. The guitars down the bottom are top lines for the bridge, they are also split into two channels each for the distorted and DI tracks.

The drums were fairly simple to mix as most of the elements aren't played over the top of each other and so separation is easily achieved. My main focus was to make sure the kick and snare were both coming through at a similar level so that the groove of the track is intelligible. The easiest way to explain the relationship between the kick and snare is to look at a beat in typical 4/4 time. The kick will play every 1st and 3rd beat in a bar, while the snare will play every 2nd and 4th; without the snare we would simply only hear the 1st and 3rd, without the kick we would only hear the 2nd and forth, and there would be no way to define the time signature. Getting both of these sounds to a level at which they can be heard against each other isn't too difficult because again, for the most part they are played separately, but being able to hear them in the mix is a different story. It can be borg useful and also a bit overwhelming to have multiple tracks for these elements as I did with this song.

The kick is broken up into just one track of the actual kick drum played during the session and a sample that follows the exact timing and velocity of the original kick. A sample is often used for percussion elements in Metal as its hard to keep a decent velocity when a drummer plays quickly, like double kicking for example, and therefore the sound can lose clarity. When it comes to the EQing stage I will talk more about this. But for now I just levelled them so that they can both be heard, I don't want to mess around with them too much in this stage because I know that I will change the levels later. It will be a similar situation for the snare as it has three elements that make up the sound, a top and bottom from when it was recorded in the studio, and a sample track of it's own. I will go through the difference these different tracks make to the over all sound of the kick and snare when I discuss mixing. I kind of liked where the toms and overheads sat on their own so I didn't do much to them other than panning the toms slightly left and right so that they kind of roll across the stereo field when they're played, and the overheads hard left and right for stereo image.


After levelling and panning everything I created individual master faders and VCA group faders for the Drums, Guitars, Bass and vocals. This meant that I could now tweak the overall level of each instrument as one piece rather than adjusting multiple faders each time (e.g. all faders for the drum tracks move with one fader.)

Now that I have all my levelling in place, it's time to move onto processing.

 

Processing - Drums

Rather than going through every little bit of my processing I will try and pick out important points that I think are worth mentioning, and go into decent detail about them. I'll still go over what processing I have done on every instrument or channel but I don't want to spend too much time on one thing.


As I mentioned earlier, there are certain elements in the drums that need to be processed so that they can work together to achieve the best sound. I'll walk you through my processing for the two most important drum elements, the kick and snare.



The first thing I did to the kick was cut out all of the spill from other parts of the drum kit with a pretty hard hitting gate (left). This allows only the kick to come through.





Then I immediately noticed how muddy it was (especially when it came to sections with double kick) so I then took an EQ and attenuated bit of the sub end, around 50hz downwards, and amplified the frequencies around 100hz with a fairly open Q to promote a punchier feel.



This cleared up the muddy tone of the kick but I still needed more presence so I made use of the kick sample and ran it through an EQ that cut everything in the low end from almost 450hz downwards, and boosted the high end slightly with a shelf at 5khz. This allows the real kick to provide most of the low end, while the sample provides presence in the high.


Here's what it sounds like after my processing, without the sample;

And here's what it sounds like with the sample;


It has a lot more attack and a lot more consistency than the original, and although it doesn't sound pretty on it's own, it makes a massive difference in the mix.


I applied this same principle to the snare and used the snare up for majority of the low content and the snare down for a majority of the higher content. I basically listened to see which one offered the most of a particular frequency range and EQed them to suit. I adjusted the level on these to fit my processing and then added in the snare sample for extra impact. I took almost all of the lower frequency content out of the sample, as well as a bit out of the mid range at about 450-500hz. Then I boosted it at 5k with a shelve to give it more presence and once mixed in with my other snare tracks I had a nice and punchy, consistent snare.

I actually didn't apply any processing to my toms because I thought they sounded fine the way they were, and for my overheads I added a high pass at around 100hz to drop out any low end, a small dip around the 500hz, a tiny tiny attenuation at 5khz and a shelf to bring up the high end at about 10khz.

I then created a bus for a bit of drum room reverb and one for some sidechain compression, which you can read about it one of my previous blogs. I am quite happy with this drum sound for the track so I didn't try and make it sound too much like my reference, if anything I wanted to take that chunky, full guitar sound from the reference.


 

Processing - Guitars

After messing with it for a while I noticed that I wouldn't be able to achieve exactly the same chunky guitar sound as my reference because a lot of what goes into the tone of a guitar is mostly done pre-tracking; meaning that the amp, the pedals, settings on the guitar, everything like that will affect the tone of a guitar and once it's committed to tape it's hard to emulate it without the right amp simulators or distortion plug ins. I am however, happy to try and get the guitar sound I have to fill a similar space in the mix as my reference.


The main guitar tracks, as I mentioned, consist of a DI clean guitar track and a distorted track. The distorted track had a lot of high end content so keeping with my theme of working with what has been recorded rather than trying to change it completely, I dropped a lot of the low end out of both tracks and raised a bit of the low-mids and high mids, leaving a little "valley" in the middle of my frequency spectrum so it doesn't sound too blocky. I also dropped just a bit of the high end above 10khz because I was going to use my DI guitar for a nice top end. As I said, the DI guitar is a clean signal so I had to dirty it up. I was looking for a super crunchy sound the had some nice low end response, but also some top end, so I went through the distortion plugins that I had available and came across Eleven Lite, which is a vintage amp simulator with the exact sort of sound I'm looking for. Top that off with a very heavy application of the compressor Smack! to get it back to it's original level after adding distortion (and also to add a bit of thickness and weight to the sound), and I've got a nice, full, crunchy sound to mix in with my guitars.


Distorted guitar EQ

DI guitar EQ

Eleven Lite Amp Sim

Smack! Compressor

So you can get an idea of what the difference is, here's an example of the DI guitar without this processing an with the processing;




After mixing them in with the distorted guitars I decided they still needed a bit more fuzz and thickness to fit my reference, so I fiddled around with a saturation effect on my guitar master bus to glue the two tracks together and again, add that all important harmonic content that creates the hefty sound I'm looking for.


 

Processing - Vocals

I'm not sure if it was the mic choices that were made or just the vocalists skill level, but I really don't like the vocals on this track. They're kind of airy and don't really sound like a good performance. So trying to get a good sound without focusing too much on the performance was a challenge. I decided to just try and focus in on how my reference track sounds in way of vocals and match the timbre and placement as much as possible.

I sent the vocals all to their own reverb to begin with,because I wanted to put them all in the same theoretical space without having to individually put reverbs on each channel. Then I started with a bit of compression just to even out the volume and bring a bit of power through in the performance. I didn't do too much to the EQ of these vocals but it was necessary to take everything down from about 300hz at 12db/octave to try and remove the horrible low end pops and booms that were throughout the performance (it sounds like they haven't used a pop filter). I kind of made my own decisions regarding which tracks were the leads and which were fillers, but it's fairly obvious to see which vocals are present for the majority of the track.

Yellow Box is Main Vox, Blue and Green are Filler

I mostly left the tracks highlighted in green as they were, with the exception of the EQ and Comp that I've already mentioned. For the main vocals I chose one to add a bit more low end around 250hz - 1khz and one to cut those frequencies completely out.

Then I decided I would try and add some distortion to the third vocal track, again just to add a bit more power and reinforcement. I got this idea from a band called Anaal Nathrakh who use quite a bit of overdrive distortion on their vocals. This plugin, which basically simulates a guitar or bass amp, gave me the tone I was looking for.

I repeated this processes for one of the lower back up vocals to give them some more presence where needed.


Over all, the vocals just required a bit of messing around with, but at the end of the day, not too much processing. I just wanted them to fit nicely into the mix without messing around too much with the overall sound of them.

 

After I had finished my mixing I popped a fairly strict compressor onto the master just to tie everything together and then popped a limiter on that kept everything at about -2db. This wouldn't have been my usual choice but for some reason I could not get the snare drum to come through the mix without blowing out my master levels. Initially I cut the snare sample all together (which was causing my master to clip), but then the snare sounded too rattly so this was the solution I landed on.


 


This is my first real shot at mixing a metal track and I have to say, I'm fairly happy with the results, but I'm aware that there are more things that I could have done. I mostly went in unarmed when it came to knowing much about metal is mixed, I essentially went in with only my knowledge of mixing and understanding of how metal or death metal should sound. While I did follow the reference track, I didn't refer too it an awful lot and I think I may have benefited more from carefully looking at each instrument and deciding what I should do with it based on the reference. I would like to record a metal band, or at least sit on the process because I have a feeling that a lot of the sound and feeling that goes into a metal track is done in the tracking process; most of the timbre and texture is established through the recording process with use of different amps and outboard effects. At least that's how I would want to approach recording a metal or death metal band, that way the mixing process is left just to polishing the sound. God help me I would do anything to re-record those vocals.


I know this is a lot of writing so thank you for reading, and I will definitely be doing a vlog instead next time.

 

References


Bazil, E. (2014). Mixing To A Pink Noise Reference. Retrieved from https://www.soundonsound.com/techniques/mixing-pink-noise-reference