This breakdown of the mixing process for ‘Take Her Away’ is not intended to be a complete “how to mix” guide; it is simply an insight into my mindset and methodologies at the time I was mixing the track. The accompanying videos at the bottom of this page will give you an insight into how different elements of the track sound in isolation, but the written content is more about starting to learn how to go about fulfilling your vision for the mix from the parts that have been tracked. It is also important to give some thought to the context in which the song was produced, and to think about the successes of the production (as well as any areas that, in hindsight, could have been improved). In the near future I plan to write my second book, which will be a comprehensive but extremely practical guide to mixing, so keep an ear out for that!
If you need to reacquaint yourselves with the song it is available here.
‘Take Her Away’ was fairly demanding to mix because of its progressive nature; the song needed broad dynamics and a strong sense of ‘journey’ whilst maintaining a perceived loudness level which competes with modern releases. The overall loudness of the track wouldn’t make or break the mix, but it nevertheless had to be considered. To achieve this, my goal was to mix the loudest points of the track as I would a modern commercial rock track, using wide panning, heavy use of compression/limiting, drum augmentation, and plenty of tape-style saturation and distortion to create a larger-than-life sound-field. To create the contrast in the quieter sections was to use inordinate amounts of automation – this would give the song the almost theatrical sense of progression I believe it needed.
I used the Mars Volta ‘Amputechture’ album mixed by Rich Costey as a reference point for the heavier sections, particularly the songs ‘Vermicide’ and ‘Viscera Eyes’.
I also wanted to capture some of the sentiment of the more laid-back verses and the middle section by referencing some of the more subtle parts of Pink Floyd – ‘The Wall’. For this, the song ‘Hey You’ seemed perfect, as there was a great contrast between the main part of the song and the explosive guitar solo, which in some ways was consistent with the middle section of ‘Take Her Away’.
Starting the Mix
When starting any mix, I will begin by cleaning up the session. As recording sessions are often frantic, even those projects I’ve produced often need some tidying up. I have an organization and colour-coding system which I use on every mix. It is up to you to create a scheme you are comfortable with, but I usually organize the drums and percussion at the start of the mixer window, before moving on to bass guitars, electric guitars, acoustic guitars, piano/keyboards, samples, strings, horns, lead vocal and then backing vocals last. As for colour-coding, I find it best to create distinct and consistent colours for each type of instrument (e.g. drums, guitars, vocals etc.).
Once the tracks are organized and colour-coded I like to tidy up the audio files themselves. I strip out unwanted noises, look out for parts which may warrant splitting (“multing”) to their own channel. A lot of the time I will also decide on a blend for multi-mic’d sources (especially guitars) and print them to mono channels to reduce track count and make the session more manageable. (I inactivate and hide the original channels but I never delete them.) I would also listen out for any editing or tuning that might be better to correct. Although it is not the primary job of the mix engineer to be editing audio, sometimes it is a necessity and I’ve often found these elements grind away at me, and fixing them when you first hear them leads to a more enjoyable mix and thus a better end result.
Accessing the Production
Once the session is neat and tidy and the audio files have been “fixed” it is time to actually start thinking about mixing. Before starting the mix in earnest, I will usually listen to my chosen reference material, which gives me a chance to get to grips with the room I’m mixing in. This can be beneficial even if it is a room with which you are familiar. You would be surprised what a difference “refreshing your ears” can make – for instance, if you have been listening to music in the car on the way to the studio, just watched TV or even been to a live gig the night before, all these acoustic environments can subtly skew the way you subconsciously “want” music to sound.
Once I have acclimatized to the surroundings I will play the track I’m mixing in its entirety and make notes on it. If I have a rough mix (e.g. a monitor mix) from the client I’d choose to listen to that. When listening to the rough mix I want to work out what elements of the song, the production team and artist deem important. Usually, what the client wants is something that resembles the rough but just sounds bigger in every way.
In the case of ‘Take Her Away’ I had live-engineered and produced the track so I knew exactly what I wanted to achieve. However, I started mixing several weeks after recording the song, so I needed to access the production with fresh ears. Giving enough time for your preconceptions of the production to fade is probably my number one piece of advice for someone mixing a track they produce. It’s funny quite how different you feel about it once you are separated from the adrenaline, excitement and, on some occasions, the stress of the session.
Some of my notes taken on ‘Take Her Away’ when assessing the mix were:
- Snare drum doesn’t have enough attack or “beady” sustain. Maybe use a sample of a more metallic snare drum.
- The solo section before the last chorus lacks definition and “thickness” in the low-mids; maybe an organ pad will fill that gap?
- Contrast between verses and choruses needs to be exaggerated.
- Drums need to feel more aggressive in the choruses.
- Vocal needs more “space” around it.
Gain-Staging / Basic Balance
Next I like to achieve the best possible mix by only using panning and level adjustment. I call this the basic balance. It might come as a bit of a surprise how just adjusting the basic balance can
improve the intelligibility of a mix. If you dive right into processing individual tracks before you have built the basic balance, you will find yourself adding processing you don’t need and thus reducing the quality of the overall mix. Remember, every move you make should be made for a reason, and that reason should be based on what you hear in the context of the whole mix, not in isolation. That doesn’t mean that monitoring channels in isolation is bad, it just means that you should use the solo button to search and correct the faults you have already heard.
When creating a basic balance I will often start by just playing around with the faders. This gives me a little insight into how well the tracks are fitting together. At this stage I will sometimes make some notes on extra processing that I might want to implement later. Once I am beginning to get to grips with the song, I will flatten the faders and use clip gain/audiosuite or a trim plug-in to create the basic balance. Once you are finished it should be gain-staged to have the master fader peaking between -6dBFS and -8dBFS (or use zero on the K-meter if you are implementing the K-System). This allows me headroom not to clip, and also allows the faders to stay closer to unity by the time the mix is finished.
The Static Mix
Once I’m happy with the basic balance, I like to process the audio in such a way that the mix holds up well from start to finish without having to use automation. This stage usually takes three or so hours and is comprised of most of the processing usually associated with mixing: EQ, Compression/Limiting, Gating/Expansion, Distortion, Reverb, Delay and Modulation. I normally start with the densest section of the song and work backwards, using various techniques to create contrast between sections. Like I said initially, this is not a how-to guide to mixing, so explaining these tricks is beyond the remit of this post – but rest assured I will be writing something comprehensive very soon.
Nevertheless, I feel I should mention that I created a three-tiered mix structure:
- Individual channels (for instance kick in, kick out, snare top etc.)
- Sub-groups. The outputs of the individual channels are fed to sub-groups for each type of instrument (drums, bass, guitar etc.)
- Mix buss (or 2 buss). The output of all the sub-groups feeds the mix bus which is where all of the parts can be processed as a whole.
I then used a “top-down” approach to mixing ‘Take Her Away’. A top-down methodology involves applying processing at the highest group level possible. For instance, when beginning the static mix you might have felt the whole mix lacked a little presence; therefore the decision might be made to add some more 3-6KHz, overall. The most sensible way to achieve this with the three-tiered mix structure is to add it to the mix bus. Whilst you could add this using an EQ on every channel, there are significant drawbacks:
- It uses significantly more processor power.
- It is likely to introduce more phase shift.
- A sense of cohesion occurs when processing tracks together (especially with compression), rather than individually.
Making changes to the mix buss isn’t always the most sensible approach; you may find that adding presence to the overall mix would leave the guitars sounding a little too piercing. In this case you might want to add the corrective EQ to each of the sub-groups (except the guitars), which is still markedly better than adding an identical EQ to every individual channel. Even when using a top-down approach there are still plenty of occasions where you will want to apply processing to individual channels, so don’t be afraid to do that if that is what the music is calling for.
Please note that the static mix doesn’t need to be perfect – and in fact shouldn’t be – but it should mean that you can play the track from start to end without feeling like it completely falls apart. If this cannot be achieved it is highly likely you haven’t given enough serious thought to how you multed the audio tracks when you cleaned up the session.
On a moderately dense contemporary track you should find that creating a kick-ass static mix will take approximately three hours and, once complete, the song should sound good – and, to some, would be passable as a finished mix. In reality though, you are only 90% there. Unfortunately however, the law of diminishing returns applies and the last 10% will take as much time as you’ve spent on creating the rest of the mix.
At this point you are likely to have already invested around 3-5 hours of your time. What happens after this point can be described as fine-tuning your mix and is likely to take you the same amount of time again. Most of this time will be spent automating, creating even more contrast between sections, and adding spot effects to create extra interest.
During the static mix stage, your use of compression and limiting is likely to have added a well-needed sense of consistency and control. However, an unwanted side effect of this is that some of the dynamics inherent in a human performance may be lost. This can make a mix a little uninteresting, especially as it develops.
Automation is an important tool to recapture this lost interest. When mixing it is not simply a case of riding the faders to add back any dynamic lifts lost from the human performance, but also a chance for you to create a greater sense of cohesion and performance. When automating, I am asking myself several key questions:
- Has the processing performed on the static mix lost anything from any of the performances?
- Are there any parts of a performance that I want to draw more attention to? I.e. a nice guitar lick or arpeggio buried in a rhythm track or a ghosted snare note that doesn’t cut through the mix.
- Does the song have a suitable sense of journey? I.e. does the chorus sound bigger than the verse? Does the song feel like it progresses throughout?
Automation doesn’t just have to be about levels either; sometimes adjusting panning or adding extra compression for a certain section might do a better job. Even muting instruments can help give a song more progression – for instance, muting a guitar part from the first verse but then including it for all proceeding verses might give the song a sense of build unachievable with automation alone.
On most contemporary tracks it is likely that you will want to add more than just level, panning and mute automation, and spot effects are a great way to help keep a listener’s interest, especially when they might have heard a section several times already. Spot effects are most often spatial effects like delay and reverb (particularly with long feedback or length settings) but can also be any processing that results in special or extreme effects like distortion, EQ filters or pumping compression. Utilizing effects in this way is likely to take a bit of time to set up but it can be the difference between a good mix and an amazing mix.
There are several examples of spot effects in ‘Take Her Away’. Most notable is the vocal delay with long feedback times during the end lines of verses (commonly called a vocal spin). Other examples of spot effects are the Leslie speaker effect on the guitar at the end of the first guitar solo, the bass distortion just before the last chorus, or the way that the distorted guitar’s gain settings increase as the middle section progresses. In the case of ‘Take Her Away’, these other spot effects were actually implemented at the tracking stage, but these sorts of effects can be added at the mix stage to breathe new life into a recording.
The End Result
Overall, the mix for ‘Take Her Away’ stands up well against commercial references, despite the self-imposed restrictions (e.g. only using a small studio space, with mainly budget equipment and no rentals). The track is in some respects a little dry and lacks some “sparkly” room energy; part of this was due to actively avoiding capturing excess room tone, as the room sound was poor and was only reducing the quality of the recording. The biggest challenge was the snare drum sound – the type of drum and tuning was, in hindsight, a little too weak and lacked the attack to cut through the guitars and also lacked the sound of the beads you would expect to make it larger than life. These days I own several snare drums and would have perhaps chosen a brass-style snare drum, tuned to a note sympathetic with the key but not as high as I did during the ‘Take Her Away’ recording session. The sound of the snare was successfully augmented with samples to give the aggressiveness and thickness required, although due to the fact that it was higher in level than the natural snare, it sounds a little more “triggered” than I would usually go for. Everything else seemed to sit really well, without too much processing. As most of the recording was done without a console or outboard gear, I did expect the track to lack a little “analogue mojo”, therefore it did require plenty of analogue-modelled EQ and compression, as well as tape emulation and saturation to give it that larger-than-life character.
In terms of perceived loudness the song is comparative in volume to commercial tracks without the dynamic range being squashed beyond belief. I was also extremely please with the dynamics between sections. The use of narrower panning in the verses and then a wider L-C-R style panning structure in the chorus gave the track the explosive feel it needed when the chorus kicks in. Another success was the way that the middle section utilizes a lot of slow-rising automation to help build and crescendo into the guitar solo sections.
It is important for any mixer to develop their own taste and feeling towards others’ work, so take some time to write down what you thought were the strengths and weaknesses of ‘Take Her Away’. Once you have done this, try and theorize why any weaknesses may have occurred and then think about possible solutions.