Our engineer figures out how to apply just the right amount of polish to an indie-rock track with a DIY aesthetic.
It's generally far easier to refine and enhance something that you already like than to completely rebuild it — and fortunately this was very much how things worked with this month's Mix Rescue track 'Who Are You' by young Cambridge band 28 Boulevard. Listening to the band's self-recorded and mixed version, I liked several elements, and it had a nice overall garage-style DIY vibe, complemented with plenty of bold, but not unpleasant, effects on the vocals. You can't underestimate how important and effective this 'DIY' sound can be for an up-and-coming guitar band. As listeners, we can subconsciously buy into the sound and the image of a band that is raw, new and exciting. Think early Arctic Monkeys or the Strokes' first record. The tricky challenge for a producer or mix engineer tasked with developing this type of act is that you instinctively want to develop the sonics and make the track more able to stand up on radio and the outside world in general, but it's all too easy to apply too much 'polish' or 'refinement', with the result that the all-important vibe starts to change — and you losemuch of what excited people in the first place!
In an ideal world every mix engineer would have an assistant to prepare everything for the mix, presenting us with a DAW session that's all sorted and colour coded, and then nicely laid out over our SSL console, with a cup of our favourite coffee waiting for us when we decide to swan in and work our magic. Alas, we mortals don't have this luxury and have to do all this ourselves, but as I've mentioned in previous Mix Rescues I still like to treat the prepping of a session as a separate process from the mix. Why? Well, it can be truly mind-numbing stuff, and it's really not the best way to get into the creative mood you need when mixing! I started by grouping and arranging everything in my preferred order — drums at the top, bass, guitars, vocals and so on, and then spent a while familiarising myself with the components of the track and identifying any potential problems. It doesn't matter what order you use, but it does help to have your own common setup, as it just makes every mix so much easier to tackle.
At a basic level, everything seemed reasonably well recorded. The only real excess was with the guitars: despite the seeming simplicity of the track it contained 35 guitar files! I worked out pretty swiftly that I could potentially divide this number by three, as each part had three mic options, with the differences being fairly subtle. I felt confident that I could do some heavy pruning with these, and that the mix, along with my own sanity, would be all the better for it. So I whizzed around, picking the tracks I initially liked best, and then tucked all the other parts out of view, in another session, which I could easily retrieve if I felt later that they were needed.
This particular mix session coincided with me a having a little more outboard at hand than I might normally have, as I was trying out a few modules in my recently acquired Lindell 500-series set up. I can quite happily mix 100 percent 'in the box' but I enjoy picking particular projects that might benefit from some extra analogue processing, or allow me to try out some new toys! As I was mixing the track in Ableton live, I set up the handy External Audio effect plug-in which helps simplify the process of getting the audio in and out of your software. (Most DAWs have an equivalent tool, and many are described in this SOS article about using outboard with your DAW: http://sosm.ag/may-10-hybrid-systems).
I could tell on my very first listen to the track that the drums were going to need a lot of attention. To be fair to the band, that's one of the hardest things to record successfully outside of a studio. I had a pair of overheads to play with, along with a mono room mic and close mics on kick, snare top and rack and floor toms. Listening to the overheads first, the drums had a very overpowering, boxy quality to them, that would need attention, and there seemed very little by way of stereo image. The mono room mic, which was quite prominent in the band's mix, had a nice trashy quality to it, but once again there was way too much boxiness imparted by the space in which the drums had been recorded. The close kick and snare signals were actually quite nice in isolation, but getting a rough balance with the raw tracks and playing with some EQ meant I was doing so much cutting to get rid of the boxiness that I couldn't see a way of getting them right without at least a little help from some triggered samples.
I often use drum samples in a mix, but I like to blend them in quite subtly, almost as an effect, just to help the drums cut through or to take on a different feel in a particular part of a song. I personally find it pretty unpleasant when a drum kit has been obviously or heavy-handedly replaced, so I was keen to find a way of using more samples than I normally would, but in a realistic and natural-sounding way.
My current go-to drum replacement tool is Slate Digital's Trigger 2, which comes with an excellent library. Sometimes, though, I find that the samples can be a little too bright and clean — or to put it another way, too good — for a particular track and, this being the case here, I decided to take a different approach. I was due to record some drums for another project later in the day, and as I'd already gone to the effort of setting up the mics and prepping for that session, I had a golden opportunity to record some of my own drum samples with minimal extra effort. All I was really looking for were some kick and snare samples that would help fill in some of the sound that I'd removed from the overheads, and could blend with the existing kit in a convincing way.
I had an old wooden Premier snare that had a pretty old head on it and was generally in need of a little attention. Ordinarily that might not sound great, but it seemed to fit nicely with the slightly 'trashy' aesthetic that I had in mind, so I set about recording a series of single hits of the snare, at maybe six to eight different velocities. I did the same for the Mapex Saturn kick drum that I'd set up for the recording session, and started preparing the files to be used as samples.
When you start this process, and start thinking about what proportion of close and distant mics to use and so on, you start to get a feel for just how much work goes into creating the many excellent drum sample packs that are available. I have a fairly short attention span for this kind of thing, though, so I tried to keep things as simple as possible! Referring back to the mix I was working on, I balanced the drum mics in what I felt was an appropriate mix of close and distant sounds, bounced out six snare hits of different velocities as a single stereo WAV file, and then repeated the process for the kick drum. I then loaded the files into a stereo editor, carefully chopped the drum hits up into individual samples, and saved them with appropriate file names.
One of the great things about Slate Digital Trigger is the included free editor, which allows you to make sample presets with your own files. This editor provided a pretty effortless way for me to position the samples to create different velocity layers. You can also select a suitable Trigger 'algorithm' to trigger the samples depending on which drum you're replacing. There are plenty of alternatives to Trigger, including WaveMachine Labs Drumagog, but for this sort of work — where you want results to blend in with the original performance — make sure that you choose something that allows you to create multiple velocity layers.
With my fresh samples in hand I returned to the mix and set about trying to get the drums where I wanted them. With the provided overheads I ended up removing everything below 150Hz with a high-pass filter, and a further 3-4dB broad cut at around 350Hz with a broad parametric EQ. That's an awful lot, and I was only able to do so because of the way in which I had prepared my samples with this in mind.
For the close mics I then spent a while getting a blend of sampled and recorded sounds until I felt like I had the trashy but more focused and punchy kick and snare that I wanted. It was then a case of experimenting with balancing my close mics, overheads and single room mic. The room mic was my link back to the 'reality' of the original sound and it seemed that adding more or less of this gave me a nice amount of control over how far I wanted to push the sampled drum sound against the original.
The bass seemed to require little attention but signalled the start of my adventures with outboard processing for this mix. I had a clean DI and a fuzz channel that I checked the polarity of and then blended to a single track before sending it out to a Neve 1073LB preamp, driven hard for extra colour. This then went onto a Retro Doublewide valve compressor I had on test at the time, which levelled the whole thing out nicely. As a final touch the bass then went into my Lindell EQ which I used to add a little 60Hz, providing a nice filling out of the bottom end, which can be lost with the introduction of fuzz or distortion.
It shows just how far things have come with digital signal processing that we can now talk about using outboard in terms of an optional or subjective thing. The debate still rages on audio forums about whether analogue still sounds better but the truth is that many mixers working at very high levels in the industry are now working either 100 percent in the box or using just a few choice pieces of analogue gear in a hybrid-type setup.
One thing that is often not appreciated about using outboard, however, is the effect it has on the mixing process itself. There's nothing fun about recalling a full mix in the analogue domain. But sending a few things out to some external processing, as described in this mix, can provide a nice change of angle to what you're doing and can just be more fun! In my personal opinion you can get a bit more heavy handed with compression and EQ with good-quality outboard, if needed, and seemingly get away with it. You don't need walls of equipment to introduce this element to your mixing set up and the 500-series format, in particular, has brought the option of adding high-quality analogue processing into the realms of reality for non-professionals.
With the drums and bass sitting nicely together I needed to get the vocals right before developing the mix any further. I had a lead vocal, double track and some simple chorus backing vocals to play with and I listened closely to see how good the sync'ing was on the double-tracked vocal before going any further. 'Not very', was the answer! And when the tracked vocal was given anything remotely like parity with the lead vocal it sounded pretty messy, with frequent and obvious sync'ing issues. I started to suspect this might have something to do with the generous amount of vocal effects in the band's mix, as I hadn't noticed obvious sync'ing issues when I first heard the track.
Where it was sync'ed though, it gave a nice thickening effect on the vocal, and if I wasn't going to use much of the double track I needed to thicken up the lead part somehow. This is one area that outboard can really excel at, so I set about finding a nice vocal chain for my all-important lead vocal track. As I mentioned, I was lucky enough to have a few extra bits of outboard available to me for this mix so I had a few options. I initially had the vocal going through just the Retro Doublewide compressor that I'd used on the bass. The Doublewide is a vari-mu compressor, and it does a lovely job of smoothing things out and adding some extra thickness, which worked a treat with this particular vocal. To then try and add some extra attack and excitement to the vocal I sent it on to a UA 1176 with a medium/fast attack and a very fast release.
It was interesting to observe the effect of the vocal being passed through two compressors of very different styles, as you could actually hear (and see, with the VU meters) the compressors reacting on very different parts of the sound. I was enjoying the effect of this a lot as the vocal felt more 'fixed' and solid, but also more energetic at the same time. This classic fast/slow dual-compressor set up can really work well on certain vocals and is very easy to re-create in the box as well. It's easy to get carried away with this stuff, however, so to cover myself I recorded back in two passes of the vocal with the outboard compression — one with lots of it and a more conservative option if it seemed too much in the cold light of day!
Feeling pretty enthused by my new vocal sound I spent a while editing the double track so I could bring it up at certain points of the track for effect. This mainly involved lining up the end of certain words and doing what I could to improve the sync'ing. Once I'd got this to a usable point I ran it out to the same outboard chain as the lead vocal and recorded the track back in to use as appropriate in certain parts of the track.
Whilst I was in the outboard mood, and I had everything setup, I had some fun playing with some outboard effects to see if I could find something that my plug-ins couldn't give me. I have a 'Great British Spring' reverb at my disposal that, while it looks very much like a dusty piece of drainpipe sitting in the corner of my room, provides a very dirty-sounding spring reverb sound. It's the definition of a one-trick pony, but it certainly provides some character, so I recorded a track of lead vocal through this, as well as the simple backing vocal parts. I repeated this process with an Electro Harmonix Memory Man guitar pedal, which gave a nice lo-fi delayed sound. This is how I like to work when using these kinds of effects, quickly finding a few options and then recording them in to play with and move on. The fact that these outboard effects are mono as well means they're easy to creatively place in a mix and blend with other stereo effects created by plug-ins.
Increasingly, I find myself putting together a mix for a guitar band by first getting the vocals to sit nicely on top of the drums and bass before positioning the guitars around these elements. With the guitar track count at a manageable level, thanks to my initial pruning, I spent a while playing about finding places to put things, panning-wise, and developing a feel for what the balance of the different parts should be. The clean guitar that starts the track and runs throughout was the first part I delved into properly and I continued in my outboard mood by sending it out to my 1073LB preamp driven quite hard to the point where it was just beginning to break up. I then used just a few dBs of gain reduction with the same fast/slow compressor combo used on the vocal. This seemed to help hold the guitar in place and 'firm it up' as I wanted this guitar to hold it's own a bit more than in the original mix. In a nod to the band's own mix I found that a similar chorus echo effect helped create the right feel for the clean part and I topped it off with an instance of the Slate Digital Virtual Tape Machine plug-in.
Enjoying what this had done to the clean guitar I repeated the process with the clean picking guitar that comes in on the chorus, adding a little EQ at 6kHz with my Lindell PEX500 EQ, to bring out a little sparkle. To help this track stand out even more I added a short delay with the Echoboy plug-in, which just helped blur the edges and fit in with the vocal effects more. Mindful of the band's intended 'wall of sound' approach to the heavier parts of the track, I generally tried to resist shaping parts too much to stand out individually, so it was just a case of experimenting with panning and levels with the remaining guitar parts until the blend felt right.
With the mix in place, and things feeling pretty much in shape, I entered the final stages of the mixing process, which involved some simple level riding of guitars, drums, vocals and effects to add interest and excitement within the track. Just a dB or two at certain points can really be effective at this stage, and it's perhaps even more important if you've used a lot of compression to help bring back a little of the dynamics. I should stress that I find it really is helpful to do this stage of the process at low monitoring levels as you get a true picture of what is happening, rather than basking in the glory of loud monitoring levels. I heard an excellent quote from Chris Lord-Alge recently where he referred to mixing at low monitoring levels as like a dentist telling us to brush and floss our teeth: "We don't really want to do it, we want to crank it up, but you're just fooling yourself, and deep down we know it's a sensible thing to do.”
It was pretty late in the day now but I referred back to the original mix to check where I'd taken things. I found this quite useful, as on returning to my mix I actually became aware that the distorted rhythm guitar needed to be a bit louder compared to the clean one. I also cranked up the whole level of the vocal effects as I liked the boldness of this element in the band's version. I took the creative decision to remove some simple tuning I'd done to the lead vocal and also nudge up the level of the room mic on the drums. I was very mindful of the raw spirit I liked on my very first listen of the track and felt I'd gone too far in this respect, so pulled back some of the polish just a touch.
I felt like my work here was done at this point so I applied some basic mastering-type processing to my master output, via a slight shelving EQ beginning at 6kHz and some limiting via Voxengo's excellent 'Elephant' limiter, and it was off to the band to see what they thought. Feedback was good from the band and their only request was for me to put a touch more effect on the vocal and just bring it down in level a touch. I was more than happy to oblige, and after a quick explanation that my mix was also a dB or so quieter on purpose we were done.
For this Mix Rescue I've tried to be as faithful as I could to the band's original mix but also try and add a touch of definition and fullness that will help the track stand up on a variety of playback systems. I had to keep reminding myself of my original plan of not taking things too far away from the original mix and it was a good exercise in self-discipline and challenging your own normal way of working. With fresh ears I felt perhaps I could have done a little more with the vocal, but using effects in a bold way can be a very subjective thing indeed, depending on your mood on the day and so on. I was pleased with how I managed to alter the drum sound in a natural way with my own samples and I've made a personal note to myself to try and take the time to build up my own library of these samples.
It's very easy to get obsessed with the technical side of mixing, especially if you're still developing your skills and learning the effect that techniques can have on your work. It's important to develop these skills, but arguably the real art is in creating the right aesthetic for the song you're working on and playing your part in creating a believable and exciting sound that draws the listener in.
Tim Lloyd-Kinnings: "As the mics we were using weren't of very high quality, we masked this by layering a lot of takes. Whilst the effect sounded pretty cool, it did tend to clutter the mix somewhat. This, coupled with large amounts of reverb and delay (again to hide poor mics and rushed positioning), meant it could be quite hard to pick out individual parts, unless listening especially closely. We were so pleased when we got the mix back from Neil. He's remained faithful to our stylistic DIY ethic, while adding just the right level of clarity and punch to the overall sound, especially the drums. We can't wait to find out how he did it!”