Mixing 29266x
This document was ed by and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this report form. Report 445h4w
Overview 1s532p
& View Mixing as PDF for free.
More details 6h715l
- Words: 7,687
- Pages: 11
The beginners’ guide to
mixing From balance and pan to EQ and compression, we take you through everything you need to know to get your tracks sounding great
24 / Computer Music / June 2010
the beginners’ guide to mixing / make music now <
Mixing is possibly the most important part of the music production process and certainly the most skilled – but don’t let that intimidate you! It’s the stage at which you bring all your individually recorded parts and samples together into one coherent sound picture that resonates with the listener. Comparable in many ways to the art of sculpture, the mixing process involves shaping the sound by setting levels and the positions of parts within the stereo field (pan), and using equalisation (EQ) and compression. Many other tools may also come into play in the shape of effects plug-ins. EQ works on the frequencies of a sound to shape the tonal balance of treble, mid-range and bass using a much more precise version of the tone controls on a domestic hi-fi. Compressors control the volume of a part, automatically scaling the dynamic range (the range of volume, from quietest to loudest) throughout the length of the song. At its most basic, this helps to keep a part at a consistent volume level in the mix, but compression can also be used creatively, to give energy to individual parts and the track as a whole, manipulating its envelope shape to create that ‘in ya face’ pumping, breathing, punching sound. It’s fair to say that dance, pop and rock music wouldn’t exist as we know it without the compressor.
Another dimension
oN tHe DVD
There are plenty of audio examples to accompany the walkthroughs in the Tutorial Files folder on the disc, and if any of the words or in this feature leave you scratching your head, check out our A-Z of Computer Music in the Beginners folder
Another not-so-essential but still important mixing tool is reverb. This ubiquitous effect places the sound sent through it in a virtual space, creating a sense of three-dimensional depth. With the current trend for very dry mixes, reverb has taken a bit of a back seat recently and is often not used at all – but it won’t be gone for long. A reverb space can be an imitation of anything from the inside of a teapot to a cavernous cathedral, and by using reverb selectively and subtly, with some sounds kept dry and up-front and others set further back in varying amounts of ambience, you can create a rich, ‘natural’ soundstage. Ultimately, though, mixing is all about the set of balances and how they change through the track – dynamic development, in other words – and as you’ll read later on, it’s not about getting the perfect balance, it’s about getting the right balance, the exciting balance. It’s also important to that, in a mix, you don’t necessarily have to use everything you’ve recorded – what you leave out can be just as effective as what you put in. And lastly, before we get started, you’re not heading for some kind of definitive ‘finish line’: a mix is never finished, it’s abandoned, and quite often, the sooner you abandon it and call it finished, the better. Incidentally, although we’re primarily using Propellerhead Record throughout this guide for illustrative purposes, all the principles discussed are universal and apply equally to any software mixer in any DAW. Now, let’s get mixing… June 2010 / Computer musiC / 25
getting ready to mix Once upon a time, when music was recorded to tape, you would usually start a mix from scratch: a flat mixing desk (level faders all at zero and no EQ active) with no effects inserted into any channels or being applied from auxiliary sends, and no rough monitor balance. Now, with the ‘total recall’ of software DAW mixers doing away with the need to painstakingly set everything up specifically for one project at a time, we tend to add compression, EQ and effects as the track develops, essentially mixing as we go, making it seem like a backward step to strip out all the effects and start again. That might indeed be so, but a half measure of bying all the effects and muting everything back to the ‘first’ part (be it drums, vocal or whatever else takes your fancy – a suitable starting point, essentially) is certainly a good and refreshing practice and needn’t destroy all the work you’ve done, since you can just save a separate version of your project file to return to if necessary. Also, it’s worth rendering a version of the final monitor mix as a stereo audio file, so you can quickly refer to it as you build the mix back up.
sound structure
A few small measures can help you psychologically prepare to get into mix mode, some of which you might have done already during the recording process. One is to break your song down into sections using markers, assuming your DAW offers such functionality. Being able to visually distinguish between verses, bridges, choruses, etc, helps you get a sense of the structure of the song and how the energy and dynamics of the mix need to develop. Looking at the structure of the track in list form might also suggest ideas for introducing drops or edits. Mixing can be an epic, confusing process at the best of times, so it’s worth
Zeroing the mixer is a good idea, even in these software-dominated days of mixing as you go
doing everything you can to make it as easy and painless as possible. A relatively complex mix might comprise over a hundred tracks, and you can be constantly scrolling up and down the mixer and/or arrange page to find the track you’re after. It pays to organise your tracks into a logical order, using colour coding if possible, so that you can confidently get straight at it. Put all the vocals together, all the drums and percussion, all the synths, etc, and by adopting an order that runs through all your mixes, if you ever have to revisit one, you should be able to quickly find what you’re after. That organisation naturally extends to naming all the tracks as well – but you’ve done that already, surely… Some DAWS (Pro Tools, most famously)
What makes a good mix? The question of what makes a good mix is hard to answer in under a million words, but there are a certainly few salient points to bear in mind. A good mix feels like an excitable dog on a lead – it wants to leap out of the speakers at you. For that there has to be a certain level of ‘punch’, and that comes from the skilful use of compressors on both individual parts and the mix as a whole. There also has to be an excitability, and that comes from precision equalising, particularly of the treble frequencies, creating clarity and brightness. There also has to be something for the ear to focus on: one dominant part at all times to lead you through the song. That might be a lead vocal interspersed with a riff or tune of some description, but never should two unconnected parts compete for listener attention. That’s prevented by using clever balancing and riding the volumes of the
26 / Computer musiC / June 2010
separate parts. There should also be something to keep you interested from beginning to end, and that’s also where dynamics come in – a charming combination of varying levels through different sections of the song: quiet and loud, light and shade, intensity and release. A good mix also has a warmth, weight and power to it that comes from the distinct working of the bass frequencies with EQ and compression. And last but by no means least, a good mix has a middle frequency balance that makes musical sense of the arrangement. To illustrate this, turn the music on at any point in your track and you should be able to understand what’s going on immediately. So, that’s a good checklist to run through as you approach the end of your mix – just ask yourself: does my mix fulfil all these criteria?
enable you to customise mixer and arrange page layouts, so you can create different windows for different instrument groups (drums, vocals, etc), which can again speed up access time and make you feel more in touch with the mix.
bussing about
Sub-grouping the audio outputs of instrument groups also makes mixing easier. For example, routing all your drum tracks to a single stereo bus channel before they reach the main output means that you can not only quickly adjust the overall level of the drums as one entity (which, generally speaking, they should be) but also process the whole group with insert effects. If you have time, it’s well worth going through the audio tracks in your project and checking all crossfades for clicks and glitches, especially on vocals. And while you’re on the vocal, check for clipped breaths and other sounds that will become exaggerated by the compression and EQ you’ll soon be applying. Cutting out noises from amps or headphone spill beforehand is also a good idea.
Audio junkie
Some people like to render plug-in (and, indeed, external hardware) synth and sampler parts as audio tracks prior to mixing. This is good practise, because MIDI can be erratic in of timing and note lengths, with notes hanging or not firing, synth patches changing unintentionally, etc. It also means you have a reliable audio archive of the part should the software you used to make it ever become redundant. Finally, get your head into the right space by assembling a few aspirational commercial tracks for A/Bing with your mix. Don’t be disheartened by the initial gulf in quality between yours and the pro mixes – with practice, it will gradually narrow.
the beginners’ guide to mixing / make music now <
Mixing it up Most tracks in your mix will need some degree of EQ to make them fit with everything else. Many will also need compression to push the energy and ‘size’ of the mix. These processors are added as inserts into the mixer channels. The order of the two can vary, but as a general rule, EQ should come after compression, because the latter can affect the tone. If the part needs extreme EQing, however, such as rolling off the low end, the other way round may well prove the better option. As you go through each track, listen to it in solo and apply EQ to bring out its best qualities. Then drop it into the track and listen to what you need to do to make it stand out in the mix and avoid interference with other parts. For example, an organ sound might be rich, warm and full on its own, but in the track, its low end could interfere with the bassline itself, turning
the bass end to mud. In that case, a reduction of the low frequency band of the organ from 150Hz down (known as a shelf roll-off) should do the trick, separating the bassline from the organ; and a slight boost of the mid-range, between 800Hz and 1kHz, might help articulation.
Physical presence
Always be careful not to overdo the ‘attractive’ frequencies – the upper mids and highs. If you EQ everything in isolation to have more bite in the presence (upper-mid) range (around 2kHz), they’ll all be competing for the same space when you drop them into the mix, getting in each other’s way and focusing all the sound into a narrow frequency range, which makes the mix sound small and thin. Our hearing perceives music as louder the more frequency ranges are occupied: a simple example is a jet plane and a
pure tone. If both are played back at the same measured decibel level the jet will sound much louder as it’s a broadband noise ranging across all audible frequencies, whereas the pure tone is a single-frequency sine wave occupying just one band. By EQing parts into different tonal areas, not only will you increase the size of the mix, you’ll also increase the clarity and separation. If a track is coming and going in a mix, or notes are tailing off, leaving a hole, or if it feels limp and weak then compression is the answer. A good compression setting start point is a 4:1 Ratio (literally the ratio of input level:output level) with a medium Attack time and fast-to-medium Release. From there, lower the Threshold to set the point at which the compressor kicks it into action and experiment to get the result you’re after. More on compression later…
> Step by step EQ and compression basics
1
4
A very useful tool for rolling off troublesome low frequencies is the high- filter. Guitars and keyboards often generate sub-harmonics, and vocals can be boomy, interfering with the bassline and creating a cloudy bottom end. Here (and on the DVD), a steep filter on the acoustic guitar gets rid of all frequencies below 130Hz to give the bass clarity.
The bandwidth is the size of the area around the centre EQ frequency that’s affected by your adjustments. A reliable old hip-hop production trick is to use big boosts of very narrow bandwidths on very specific frequencies. This helps the bass and kick drum in our track – a couple of narrow boosts shape the kick well.
2
5
Scooping out mid-range frequencies can open a sound up, make it softer and less woody, and help to unclog layered instruments. Here, a bongo part is opened up with a mid-range cut at around 600Hz. Precision boosting of mids can help the articulation of a musical part in the mix.
Attack determines how quickly the compressor kicks in. Set the Ratio to 4:1 (generally a good starting point) and lower the Threshold until you have 3-6dB of gain reduction. Adjust the Attack time as you listen to the percussive start of the sound. You’re looking for bite, clarity and a natural-sounding envelope. The guitar in our example on the DVD is clean when set right and squashed with a fast attack.
3
6
From 1.5-6kHz, the upper-mid range (aka the ‘presence range’) deals with clarity, edge and intelligibility. Use this to make a part cut through the mix, but be careful not to overdo it. Here the vocal is boosted around 2kHz to help it dominate.
The compressor’s Release time is another key setting, controlling the sense of energy in a sound, as well as the smoothness of the compression. A fast Release energises a sound, but also brings up noises and makes compression obvious, which can be a good thing, as with the sidestick in the audio file on the DVD. This Bomb Factory plug-in offers fast Release times.
June 2010 / Computer Music / 27
Lend us your ears You’ve recorded the song of your life: the arrangement’s perfect, the instrumentation is sublime and the performances are out of this world. Now all you have to do is balance and blend the parts into a mix that concisely expresses the emotion and energy of the song, and that stands up beside the countless other excellent mixes out there. Mixing is all about your ears. You might know what an EQ or reverb is or how a compressor works, but if you don’t serve your ears with quality monitoring, it really is all for nought. To use your ears effectively, you have to have a pair of decent monitor speakers in a room that you’re very used to. By that, we mean a room in which you listen to other music and like what you hear and therefore know how a mix of your own music should sound in that environment. If other music sounds boomy or overly bright in your monitoring system, it’s fair to say that your mixes will come out sounding the opposite of that, as you’ll automatically compensate for such problems when mixing. None of this necessarily means that you have to buy an expensive pair of monitors – most hits of the last 25 years have been partly done on Yamaha NS-10s, which were relatively cheap. You will probably, however, need to use dedicated monitors of some kind, rather than hi-fi speakers, the difference being that the former are designed to give you an even representation of the overall balance of frequencies, while the latter are designed to flatter the sound by taking out the mid-range frequencies and enhancing the bass and treble. The midrange is key to getting a good balance, because the fundamental frequencies of most instruments are in this range, so it helps to be able to hear them clearly and evenly. Being able to hear clearly doesn’t mean that you have to spend thousands of pounds on acoustic treatment either, although some can help. Most music is heard in living room environments and sounds fine, so it should be possible to furnish your room and position your monitors and listening point in a way that makes sense of what you’re hearing.
28 / Computer Music / June 2010
> Step by step Using reverb and delay
1
3
5
Rather than use reverb on every channel as an insert effect, which could bring our computer to its knees, we use a shared channel (aka, auxiliary) into which we can send a bit of signal from any other channels we like via their Send controls. Here we send a bit of vocal via Send 1 to a reverb on FX channel 1. You can use other channels’ Send 1 knobs to apply the same reverb to part of their signal.
The most important parameter of a reverb patch is the reverb time (often referred to as RT). This dictates the size of the space and generally sits somewhere between 1 and 2 seconds. Big spaces – churches and cathedrals – will have RTs up to 5 seconds and beyond, while rooms might have RTs below 1 second. This vocal reverb has an RT of just over 1 second.
As you’d expect, any instrument generally sounds good put into its natural environment. Chances are you won’t have a hall to record in, so an artificial model of one is a good second best. Here, the acoustic guitar is placed in a hall reverb on send 3, which enriches the harmonics considerably.
2
4
6
We want a general purpose reverb for our vocal and instrument channels, and in this case, nothing works better than a plate setting. The plate reverb type tends to add subtle space and size to a sound, particularly vocals. An alternative to this, however, could be a hall or chamber setting. Try them all out and get to know the character of each one.
Long reverbs on drums are OK for slow tracks, but things can easily get too washy. Room reverbs tend to work better as a general rule, and the key parameter after the reverb time is the high-frequency damping. Stone rooms are brighter than wood rooms, which are brighter than furnished rooms, for example. Here, a drum room reverb plug-in is set up on auxiliary send 2.
With a long delay on FX Channel 4, we can send a bit of vocal into it for a classic echo effect. Try a quarter-note delay synchronised to the tempo of your song. We’ve inserted a reverb after the delay with a tweaked wet/dry mix on the former, so that the delay goes to outer space. We’ve also added an envelopecontrolled filter after that for even more atmosphere. There’s no limit!
the beginners’ guide to mixing / make music now < > Step by step Adding interest with plug-ins
1
3
When mixing, you’re always looking to add character to your track. Filters are great for this, and an oft-used trick is to filter the drums into a narrow band in the intro before opening them up as you go into the main song. The example on the DVD uses a vintage-style filter, and an automated By button switches the tone back to normal for the verse.
Synth sounds are often too clean and tidy out of the box and benefit greatly from a bit of roughening and distressing with distortion. Any number of plug-ins are available for this and you’ll develop favourites through experimentation. Here, Audio Damage Kombinat is dirtying up our very clean flute sound, not to the point of nastiness, but just enough to give it some edge.
2
4
Getting to know you
You should be striving to add size to the mix, and one way to do that is to broaden the stereo width. Here, the acoustic guitar sounds big and wide, which was done by putting it through a stereofier: Air Width in Pro Tools. However, a stereo pitchshifter set 5 cents up on one channel, panned left, and 5 cents down on the other, panned right, is good for widening if you don’t have a stereoiser.
Always look to up the energy of the mix if things start sounding pedestrian. A synth pad of held chords that seemed like a good idea during recording/writing might need pepping up. Making it pulse in time with the track on eighth- or 16th-notes, either on its own or mixed in with the original synth, will add drive to the part. Here and on the DVD, our Wurlitzer is getting the pulsing treatment.
POWER TIP
>Automating parameters
5
You’re also looking to add specific moments of interest to your mix. Any good dance track will employ a sliced beat or two and Audio Damage’s Replicant is a randomisable slicer that’s perfect for livening up a tired drum fill in a mix. It’s so random, in fact, that it’s best to record it and edit the audio afterwards. Here, it’s livening up our intro drums.
Very obvious effects get boring after a very short time but can bring welcome momentary change to a mix – for example, a long delay on a specific word or a big reverb on the last drum beat. Using the power of effects automation, your mix need never have a dull moment. Riding up the delay send on the vocal channel for one word or increasing the level can all be recorded into your DAW’s automation system. Similarly, filters on drum loops can be programmed to open up for instant fills.
The world of effects plug-ins is massive, and with so many at your fingertips, the danger is that you won’t get to grips with the fundamentals of any one in particular. While presets are great to get you started, time spent getting to know the ins and outs of an effect can enable you to be more precise with your intentions and more original with your results. Here are a few starting points to consider.
Time delay effects This covers a wide range of processors beyond the plain old echo, including chorus, flanging, phasing, doubling and slapback delay. The first three use short, varying delay times to create a sweeping sound with enriched harmonics, which can be useful in a mix to thicken sounds. Chorus is especially rich, but can tend to make things overly soft if used too liberally. Short delays can be useful to widen a mono sound into stereo. Doubling involves a slightly longer delay time and is often used on vocals to thicken them up. Slap-back is longer still, and is another great vocal effect to consider instead of reverb. It was used a lot in 50s rock ’n’ roll and by John Lennon. It’s usually best to run all time delay effects on auxiliary sends, because they require a mix of the dry and effected signal.
Filters You’ll no doubt be familiar with filters from having played around with soft synths, and they’re basically extreme EQs. As insert effects, they can radically change the tone of something if it’s getting in the way or just doesn’t sound exciting. You can add some interesting movement to a mix by modulating the filter cutoff frequency, either cyclically with an LFO, perhaps in time with the session tempo, or according to level with an envelope filter.
Distortion One of the most useful processors when it comes to bringing energy to a flat or overly clean sound. Usually used as insert effects, distortion plug-ins come in all shapes and sizes: amp simulators, stompboxes, bit crushers and more. If a mix is sounding too light and polite, a few judiciously applied distortion plug-ins can bring the edge and darkness you’re looking for. June 2010 / Computer Music / 29
The song remains the same Vocals are the most important feature of any song-based mix. They have to sit on top of the mix at all times, at a level loud enough that they can carry the tune but not so loud that the backing track sounds small. It’s worth making a comparison with reference tracks to get a good working level. You’ll notice that in a professional mix, the backing vocals allow the lead vocal to be more prominent without appearing isolated, and effects are used to glue the vocal to the track. The tone should be bright but not harsh, and the vocal should be clear, with a full body. A touch of delay or reverb will make it sound bigger and sit it in the track.
These days, people aren’t as tolerant of wayward tuning, so a pitch correction plug-in is useful to have to hand in case any off-pitching rears its head. Any backing vocals are there to the lead and accentuate the importance of specific sections. They don’t need to be as bright as the lead vocal, but they do need to be fully legible. You can probably take some of the mid-range out, especially if your backing vocals are multitracked, to avoid them getting too thick. Listen for out-of-sync ‘esses’ or extraneous noises in the backing vocals and cut them in line with the lead vocal to make them less distracting.
Vocals are the most important part of a track, so they need to be very prominent in the mix
> Step by step Sitting the vocal
1
A high- filter is often useful as the first insert effect on a vocal channel. Set to 80Hz, it rolls off general rumble from the room without affecting the warmth of the vocal. Vocal ‘pops’ can be blitzed by automating the cutoff frequency up to 250Hz wherever they occur, but listen carefully to make sure it doesn’t sound too thin.
2
No vocal should be mixed without compression, and saturation can be a real boon. A compressor with medium Attack and Release (auto release can be smoother if your plug-in offers it) and 4:1 Ratio holds the level, and something like PSP’s amazing VintageWarmer plug-in can drive the signal to saturation, upping the energy and intensity while holding it at the top of the mix.
3
EQ is essential. Boost around 2kHz for intelligibility and to cut through the mix, but don’t overdo it. A boost from 8kHz upwards will add air and sparkle to a dull voice, and if the voice sounds thin, add a touch at around 500-800Hz. If it sounds overly hard, lose some around 1kHz. If the ‘esses’ are sounding overly sibilant, insert a de-esser after the EQ, but be careful not to give the singer a lisp!
POWER TIP
>Mic check Before doing any processing or balancing, check the vocal thoroughly in solo for smooth edits and noises. It might seem like a waste of time, but this can actually save time in the long run. Unnatural sounds and spurious noises will be exaggerated by the compression and balance of the voice, so check that all edits are crossfaded. Listen for a natural, smooth delivery, as if it was one complete take, including natural breaths and pauses. Clipped breaths can be particularly problematic if left unrepaired.
30 / Computer Music / June 2010
4
There may be occasions when you need to automate the EQ for the tone to remain consistent throughout the song. If the singer moves in to the mic for a quiet verse, the tone will become bassier, so roll off low frequencies from 150Hz for those sections. Similarly, in very loud sections the vocal might sound too hard, so a 1kHz dip can help. Here the low frequencies are being ridden in the bridge.
5
The vocal level in the mix is critical, and even half-decibel changes make all the difference. Work through the song, recording fader level automation. Real-time fader rides (ie, raising and lowering) with a mouse or MIDI controller can be tweaked with on-screen changes for fine adjustment. Ride the lead vocal, then balance the backing vocals with it. This is one of the last things to check.
Drum compression The drum kit is truly transformed with judicious compression. You always want it to be punchy and powerful. Individual drums like compression but it’s easy to get it wrong. There’s a balance to be had between the attack of the hit at the front of the sound and the weight of the drum behind it. The key is in getting the compressor attack time sufficiently slow that enough of the transient hit gets through for impact, setting the threshold level so that it punches but isn’t strangled, and getting the release time fast enough that the decay pumps back at you with power. It’s a fact of psychoacoustics that we hear shorter sounds as quieter than
longer sounds of the same decibel level, so effectively, the longer the drum decay, the louder it will seem. With miked drums, compression can cause problems, because the spill from other drums and cymbals (especially the hi-hat) is increased, creating nasty, confusing tones. At the same time, though, this can be a good thing with overhead mics and room mics, as the compressor tends to ‘glue’ the kit together – the kick drum pumps the cymbals, for example. This isn’t an issue with samples and drum machines, which is one reason why it’s hard to unify the sounds of a sampler. As demonstrated below, sub-grouping drums and routing
The drum kit is one instrument that really benefits from judicious compression
them to a flat and compressed bus (parallel compression) gives you the best of both worlds, combining punchy compression with the
dynamics of the uncompressed sound. A good technique is to bring the compressed group up gradually as the mix develops.
> Step by step Mixing drums
1
4
The kick drum has to be punchy, weighty and tight. Some compression with medium Attack and fast Release settings gives it punch. To give the kick more front or click to cut through the mix, we boost at 2kHz. We also boost at 70Hz for weight and 120Hz for power. Notching out 250Hz helps with the clarity.
The level of the hi-hats is crucial, because it helps motor the rhythm track. Depending on the material, it can be good to add some spitting aggression to hi-hats using fast compression. EQ is often used to roll off low-frequency spill from the kick and other drums, and to add some edge around 6kHz. On ballads and ‘nicer’ tracks, you can also bring out the more extreme top end over 12kHz.
2
5
Hopefully your snare was recorded using two mics – top and bottom. Top is for the power of the drum, bottom is for snare buzz. Snare compression is similar to that of the kick, although the bottom channel can have a fast Attack. Balance the two, EQ in some presence around 2-3kHz and maybe open the top end with a shelving boost at 6kHz. A boost around 200Hz helps with fullness and weight.
When mixing a drum kit, treat it as a whole single instrument, with the room mics and overhead mics as the most important elements, and the close spot mics as in-fills. With each of these we’re trying to get the whole kit sounding good in a space. Extreme compression on the ‘room’ can energise the whole track.
3
6
Because toms are hit so infrequently and can ring on in the background, edit around them. Compress them like the snare drum with fast release, EQ up the sound of the skins between 2 and 6kHz, and boost between 80 and 140Hz for weight. Notching out around 400Hz can open toms up and make them less woody.
If your DAW enables it (most do), parallel group compression of the whole kit can be very useful for keeping the punch of the drums together in the mix. Here we’re using a send rather than sub-groups, which isn’t ideal because of the stereo image, but it works.
June 2010 / Computer Music / 31
> make music now / the beginners’ guide to mixing > Step by step Mixing bass
1
It’s really important to keep an even and constant weight to your bassline. Use compression or even full-on limiting to maintain a ceiling level. A medium Attack will keep some transient impact in the playing, and a medium/fast Release will maintain a natural feel.
2
It’s hard to find space in a busy mix for clarity at the low end. An amp sim drive with a touch of distortion will impart some bite. Of course you can go over the top with distortion if you like, too. The bass in our example was recorded with drive but we’re adding more with an amp sim.
3
We’re boosting everything below 120Hz with a shelf EQ for more weight. To make it less cloudy, we notch out a narrow area around 200Hz. For definition, we boost around 1kHz. It might seem odd, but a filter roll-off below 50Hz can tighten the sound and raise the perceived volume.
> Step by step Mixing guitar
1
With electric guitar, always record the DI signal along with any amp mics, so that you can re-amp afterwards. This session was just recorded with the DI signal straight into our DAW, so it needs to be treated through an amp sim, as shown. We’re using a classic Marshall emulation.
2
The old amp spring reverbs were good for character. Here we’ve mixed a bit of spring reverb in with a touch of slapback echo, set to around 120ms delay time and a single repeat. The echo works better when the playing is more sparse – when things get busy, it can get confusing.
3
It is possible to get the right tone just using an amp, but usually EQ is necessary to fit it into the mix. In this case, we’ve added some bite in the presence range (around 2kHz) for cut and edge, and cut a touch of lower mid-range to open the sound up.
> Step by step Mixing keyboards and synths
1
Pad sounds are generally legato synth parts that fill the mix out, usually in the lower mid-range. They don’t get in the way or draw attention, but can be more prominent if it suits the track. Here and on the DVD, a Wurlitzer part does the job of a pad, warming and broadening the sound.
32 / Computer Music / June 2010
2
A lead line should take over from the vocal at the forefront of the mix as the focal point. Getting the level right is key, and the automation should be ridden in the same way as with the vocal, to keep it in the right place. In our example, we ride the piano solo out of the mix.
3
Incidental and subliminal noises are used to add interest and mystery. Generally they’ll be at quite a low level, but occasionally it’s good to have an odd noise jump out of the mix to surprise the listener. In our example, the reverse effects do that job.
> make music now / the beginners’ guide to mixing
Getting the balance right Once all your sounds have been processed individually, they need to sit together. You might have balanced them to some extent as you worked the solo parts, but you may not have been really listening to the big picture as you did it, so it could well be skewed. It’s worth muting everything and throwing up a fresh balance quite quickly so that you don’t over-focus – thinking too hard about the mix can turn it into more of a cerebral project than an aural one, which absolutely isn’t the way it should be. Any mix balance starts by establishing the foundation, usually the rhythm track – drums and bass. Get a good groove and the mix will look after itself to a certain extent. One by one, introduce the harmonic elements, the bits that articulate the musicality and mostly occupy the mid-range. At a reasonable monitoring level, set them against the groove, consider where they should sit in the stereo picture and pan accordingly. , you can automate the panning if an instrument needs to take centre stage for a section, and things can move around, too. Every now and then, drop the vocal in to make sure there’s still space for it.
“Imagine that your mix is leading the listener, someone who doesn’t know either you or the song” Check that the frequency balance is good – not too dull or too thin – by referring to other tracks. A common fault is for the mix to be too dull, so you might have to push the levels of the brighter instruments: hi-hats, acoustic guitars, etc. You may need to tweak individual EQs along the way, too. Imagine that your mix is leading the listener, someone who doesn’t know either you or the song. It needs to be exciting and unpredictable, but not jarring, and you need to provide a focus throughout, be it a vocal or lead instrument. To achieve this, use mix automation to ride the relative levels. Try riding levels around section changes for dynamic excitement, too, and make sure the change builds or falls in the right way.
34 / Computer Music / June 2010
> Step by step
1
3
5
Balancing and automation
Mute everything, then, part by part, starting with the rhythm track, build a mix with no dynamic automation. This static setup should give a good representation of the song with an even tonal balance. Check it against reference tracks to make sure it’s bright and full enough, and do any tonal tweaks to individual EQs as you hear fit.
When musical changes are approaching (the transition from verse to chorus, for example), consider how you can telegraph them and/or build them up to create a dynamic moment and, after the change, ensure the mix lives up to that build. Bridge sections before a chorus should do as the name suggests, serving as a musical link between it and the verse.
Another final job is to check the beginning and end of the mix. Listen to the front for noises and leave a bit of silence before the start for later mastering work. Check the end of the mix for any fade or ‘hangover’ noises you don’t want. You can always leave the fade-out until later if you’re not certain.
2
4
6
Exploit the stereo soundstage as much as possible to create width. Bass instruments and main drums tend to stay in the middle of the mix, but there are no set rules and odd panning can work as a pleasant surprise sometimes. Automated panning is an option, too, for swirling instruments or bringing featured moments into the centre.
One of the last things to do before committing to a mix is listen through to the vocal level and any other principal parts. Listen at a normal level and refer to other tracks to make sure the vocal is dominant enough and not too imposing. If necessary, apply manual automation changes.
Most DAWs have a ‘bounce to disk’ option. Set the start and end point, and bounce away. Check that the mix level doesn’t go into the red at any point. If it does, you’ll have to work to bring everything down proportionally – save a new version of the project before you do that, though. Do a final check of any mix compressors or EQs and you’re done. See you in the charts!
> make music now / the beginners’ guide to mixing
mixing tips and pitfalls to avoid REST YOUR EARS After only a few hours of concentrated listening, your perception shifts and wrong decisions get made. The beauty of instant recall with software mixers is that you can work for short periods of time on a mix and come back to it fresh, even if you work on other tracks in between. The best work on a mix is usually done in the first hour.
DON’T KEEP RAISING LEVELS It’s all too easy to get into a vicious circle of turning one thing up after another as parts compete to be heard. Instead, try turning things down and use EQ and panning to create separation and space.
OVER–EQING Similarly, there can be a tendency to keep boosting the presence range (around 2kHz) on the EQ of everything because that also increases the subjective loudness. This just leads to a harsh, thin mix, though. You’re looking for parts to be spread evenly across the audible frequency range. That way, elements don’t compete and the overall sound gets cumulatively bigger.
If your EQ curves look like this, you’re very likely overdoing it! Boosting the 2kHz presence range might increase perceived volume but it will grate over time, so go easy with those knobs
A/B YOUR MIX Have Winamp, iTunes, Spotify or your other media player of choice running in the background with a playlist of reference tracks open – ie, tracks that you ultimately want your mix to sound like. Constantly referring to them can make certain you’re keeping to the path. If you think your mix is better than the reference, though, you’re almost certainly wrong: you’ve probably not matched the volumes, or worse, your monitoring is set up poorly.
LISTEN AWAY FROM YOUR ROOM It’s always good to try and get a broad perspective on the mix, so regularly make a copy and listen in another room and/or your car. Listening with other people also helps you to hear it differently.
YOU CAN’T MIX ON HEADPHONES It’s nigh on impossible to mix entirely on headphones and get good results: it’s just not a natural way to hear music. They’re good for checking the finer details of a mix, in
particular panning and reverb, and for hearing clicks and noises, but they’re tiring after a short time and damage to your hearing is always a danger.
LIMITING Some people mix through a stereo compressor from the off. It can make balancing easier by controlling the relative levels but we’d recommend getting a good natural balance and leaving the compression or limiting to the mastering process, when you can really focus on getting your mix loud. That said, it can be useful to have an inserted mix limiter available for quick comparison with your reference tracks, which will, of course, have been professionally mastered.
SAVE AS YOU GO Red herrings are common when you’re mixing and you’ll sometimes come to the realisation that it sounded better a few hours/days/weeks/months ago. You should regularly and sequentially save new versions of your project as you mix. If you bounce a rough mix to check outside your studio, always give it a name and save a session file with exactly the same name so that you can go back to it easily.
THE BASS ISSUE The hardest thing to get right in a mix is the bottom end, which is key to a good overall sound. It’s also very hard to hear on a small monitoring system, but you can usually trust car stereos and hi-fis to tell you if there’s too much or too little bass.
PHASE CANCELLATION Keep your media player like Spotify open so that you can refer your mix to commercial tracks in your genre
36 / Computer musiC / June 2010
Be aware that what might sound really wide and ‘outside’ the speakers in stereo could disappear altogether when collapsed to
mono because of phase cancellation. Although mono isn’t used much these days, it’s a good idea to maintain compatibility anyway, so that your mix is always true, no matter what it’s played back on.
RADIO WAVES Ask yourself: ‘Could I hear this track played on the radio?’. This is a good way of detaching yourself from the mix and hearing it objectively. It’s got to excite other people who aren’t connected to it in any way, and it has to have more fizz and excitement than you might think.
DON’T OVER-MIX Mixing isn’t about getting the perfect balance. Spend too much time on a mix and you risk killing it by smoothing off all the edges. As you near the end of mix, see how dangerous you can be with the levels of featured parts.
MONITOR VOLUME All but the worst mix will sound good and balanced at high volumes because your ears compress it into shape, so don’t kid yourself by monitoring loud all the time. Louder listening may be appropriate when toning individual sounds, but monitor at a reasonable level when balancing: it’s truer and less tiring. If you feel an urge to turn it up all the time, the balance or brightness level could be wrong.
ON THE LEVEL If you find yourself pulling your master fader back a lot, you could be overloading the mix bus. Start your mix with the channel faders at around -12dB to give you headroom as the mix builds, and be prepared to scale them all back as a group if they creep up too far.
mixing From balance and pan to EQ and compression, we take you through everything you need to know to get your tracks sounding great
24 / Computer Music / June 2010
the beginners’ guide to mixing / make music now <
Mixing is possibly the most important part of the music production process and certainly the most skilled – but don’t let that intimidate you! It’s the stage at which you bring all your individually recorded parts and samples together into one coherent sound picture that resonates with the listener. Comparable in many ways to the art of sculpture, the mixing process involves shaping the sound by setting levels and the positions of parts within the stereo field (pan), and using equalisation (EQ) and compression. Many other tools may also come into play in the shape of effects plug-ins. EQ works on the frequencies of a sound to shape the tonal balance of treble, mid-range and bass using a much more precise version of the tone controls on a domestic hi-fi. Compressors control the volume of a part, automatically scaling the dynamic range (the range of volume, from quietest to loudest) throughout the length of the song. At its most basic, this helps to keep a part at a consistent volume level in the mix, but compression can also be used creatively, to give energy to individual parts and the track as a whole, manipulating its envelope shape to create that ‘in ya face’ pumping, breathing, punching sound. It’s fair to say that dance, pop and rock music wouldn’t exist as we know it without the compressor.
Another dimension
oN tHe DVD
There are plenty of audio examples to accompany the walkthroughs in the Tutorial Files folder on the disc, and if any of the words or in this feature leave you scratching your head, check out our A-Z of Computer Music in the Beginners folder
Another not-so-essential but still important mixing tool is reverb. This ubiquitous effect places the sound sent through it in a virtual space, creating a sense of three-dimensional depth. With the current trend for very dry mixes, reverb has taken a bit of a back seat recently and is often not used at all – but it won’t be gone for long. A reverb space can be an imitation of anything from the inside of a teapot to a cavernous cathedral, and by using reverb selectively and subtly, with some sounds kept dry and up-front and others set further back in varying amounts of ambience, you can create a rich, ‘natural’ soundstage. Ultimately, though, mixing is all about the set of balances and how they change through the track – dynamic development, in other words – and as you’ll read later on, it’s not about getting the perfect balance, it’s about getting the right balance, the exciting balance. It’s also important to that, in a mix, you don’t necessarily have to use everything you’ve recorded – what you leave out can be just as effective as what you put in. And lastly, before we get started, you’re not heading for some kind of definitive ‘finish line’: a mix is never finished, it’s abandoned, and quite often, the sooner you abandon it and call it finished, the better. Incidentally, although we’re primarily using Propellerhead Record throughout this guide for illustrative purposes, all the principles discussed are universal and apply equally to any software mixer in any DAW. Now, let’s get mixing… June 2010 / Computer musiC / 25
getting ready to mix Once upon a time, when music was recorded to tape, you would usually start a mix from scratch: a flat mixing desk (level faders all at zero and no EQ active) with no effects inserted into any channels or being applied from auxiliary sends, and no rough monitor balance. Now, with the ‘total recall’ of software DAW mixers doing away with the need to painstakingly set everything up specifically for one project at a time, we tend to add compression, EQ and effects as the track develops, essentially mixing as we go, making it seem like a backward step to strip out all the effects and start again. That might indeed be so, but a half measure of bying all the effects and muting everything back to the ‘first’ part (be it drums, vocal or whatever else takes your fancy – a suitable starting point, essentially) is certainly a good and refreshing practice and needn’t destroy all the work you’ve done, since you can just save a separate version of your project file to return to if necessary. Also, it’s worth rendering a version of the final monitor mix as a stereo audio file, so you can quickly refer to it as you build the mix back up.
sound structure
A few small measures can help you psychologically prepare to get into mix mode, some of which you might have done already during the recording process. One is to break your song down into sections using markers, assuming your DAW offers such functionality. Being able to visually distinguish between verses, bridges, choruses, etc, helps you get a sense of the structure of the song and how the energy and dynamics of the mix need to develop. Looking at the structure of the track in list form might also suggest ideas for introducing drops or edits. Mixing can be an epic, confusing process at the best of times, so it’s worth
Zeroing the mixer is a good idea, even in these software-dominated days of mixing as you go
doing everything you can to make it as easy and painless as possible. A relatively complex mix might comprise over a hundred tracks, and you can be constantly scrolling up and down the mixer and/or arrange page to find the track you’re after. It pays to organise your tracks into a logical order, using colour coding if possible, so that you can confidently get straight at it. Put all the vocals together, all the drums and percussion, all the synths, etc, and by adopting an order that runs through all your mixes, if you ever have to revisit one, you should be able to quickly find what you’re after. That organisation naturally extends to naming all the tracks as well – but you’ve done that already, surely… Some DAWS (Pro Tools, most famously)
What makes a good mix? The question of what makes a good mix is hard to answer in under a million words, but there are a certainly few salient points to bear in mind. A good mix feels like an excitable dog on a lead – it wants to leap out of the speakers at you. For that there has to be a certain level of ‘punch’, and that comes from the skilful use of compressors on both individual parts and the mix as a whole. There also has to be an excitability, and that comes from precision equalising, particularly of the treble frequencies, creating clarity and brightness. There also has to be something for the ear to focus on: one dominant part at all times to lead you through the song. That might be a lead vocal interspersed with a riff or tune of some description, but never should two unconnected parts compete for listener attention. That’s prevented by using clever balancing and riding the volumes of the
26 / Computer musiC / June 2010
separate parts. There should also be something to keep you interested from beginning to end, and that’s also where dynamics come in – a charming combination of varying levels through different sections of the song: quiet and loud, light and shade, intensity and release. A good mix also has a warmth, weight and power to it that comes from the distinct working of the bass frequencies with EQ and compression. And last but by no means least, a good mix has a middle frequency balance that makes musical sense of the arrangement. To illustrate this, turn the music on at any point in your track and you should be able to understand what’s going on immediately. So, that’s a good checklist to run through as you approach the end of your mix – just ask yourself: does my mix fulfil all these criteria?
enable you to customise mixer and arrange page layouts, so you can create different windows for different instrument groups (drums, vocals, etc), which can again speed up access time and make you feel more in touch with the mix.
bussing about
Sub-grouping the audio outputs of instrument groups also makes mixing easier. For example, routing all your drum tracks to a single stereo bus channel before they reach the main output means that you can not only quickly adjust the overall level of the drums as one entity (which, generally speaking, they should be) but also process the whole group with insert effects. If you have time, it’s well worth going through the audio tracks in your project and checking all crossfades for clicks and glitches, especially on vocals. And while you’re on the vocal, check for clipped breaths and other sounds that will become exaggerated by the compression and EQ you’ll soon be applying. Cutting out noises from amps or headphone spill beforehand is also a good idea.
Audio junkie
Some people like to render plug-in (and, indeed, external hardware) synth and sampler parts as audio tracks prior to mixing. This is good practise, because MIDI can be erratic in of timing and note lengths, with notes hanging or not firing, synth patches changing unintentionally, etc. It also means you have a reliable audio archive of the part should the software you used to make it ever become redundant. Finally, get your head into the right space by assembling a few aspirational commercial tracks for A/Bing with your mix. Don’t be disheartened by the initial gulf in quality between yours and the pro mixes – with practice, it will gradually narrow.
the beginners’ guide to mixing / make music now <
Mixing it up Most tracks in your mix will need some degree of EQ to make them fit with everything else. Many will also need compression to push the energy and ‘size’ of the mix. These processors are added as inserts into the mixer channels. The order of the two can vary, but as a general rule, EQ should come after compression, because the latter can affect the tone. If the part needs extreme EQing, however, such as rolling off the low end, the other way round may well prove the better option. As you go through each track, listen to it in solo and apply EQ to bring out its best qualities. Then drop it into the track and listen to what you need to do to make it stand out in the mix and avoid interference with other parts. For example, an organ sound might be rich, warm and full on its own, but in the track, its low end could interfere with the bassline itself, turning
the bass end to mud. In that case, a reduction of the low frequency band of the organ from 150Hz down (known as a shelf roll-off) should do the trick, separating the bassline from the organ; and a slight boost of the mid-range, between 800Hz and 1kHz, might help articulation.
Physical presence
Always be careful not to overdo the ‘attractive’ frequencies – the upper mids and highs. If you EQ everything in isolation to have more bite in the presence (upper-mid) range (around 2kHz), they’ll all be competing for the same space when you drop them into the mix, getting in each other’s way and focusing all the sound into a narrow frequency range, which makes the mix sound small and thin. Our hearing perceives music as louder the more frequency ranges are occupied: a simple example is a jet plane and a
pure tone. If both are played back at the same measured decibel level the jet will sound much louder as it’s a broadband noise ranging across all audible frequencies, whereas the pure tone is a single-frequency sine wave occupying just one band. By EQing parts into different tonal areas, not only will you increase the size of the mix, you’ll also increase the clarity and separation. If a track is coming and going in a mix, or notes are tailing off, leaving a hole, or if it feels limp and weak then compression is the answer. A good compression setting start point is a 4:1 Ratio (literally the ratio of input level:output level) with a medium Attack time and fast-to-medium Release. From there, lower the Threshold to set the point at which the compressor kicks it into action and experiment to get the result you’re after. More on compression later…
> Step by step EQ and compression basics
1
4
A very useful tool for rolling off troublesome low frequencies is the high- filter. Guitars and keyboards often generate sub-harmonics, and vocals can be boomy, interfering with the bassline and creating a cloudy bottom end. Here (and on the DVD), a steep filter on the acoustic guitar gets rid of all frequencies below 130Hz to give the bass clarity.
The bandwidth is the size of the area around the centre EQ frequency that’s affected by your adjustments. A reliable old hip-hop production trick is to use big boosts of very narrow bandwidths on very specific frequencies. This helps the bass and kick drum in our track – a couple of narrow boosts shape the kick well.
2
5
Scooping out mid-range frequencies can open a sound up, make it softer and less woody, and help to unclog layered instruments. Here, a bongo part is opened up with a mid-range cut at around 600Hz. Precision boosting of mids can help the articulation of a musical part in the mix.
Attack determines how quickly the compressor kicks in. Set the Ratio to 4:1 (generally a good starting point) and lower the Threshold until you have 3-6dB of gain reduction. Adjust the Attack time as you listen to the percussive start of the sound. You’re looking for bite, clarity and a natural-sounding envelope. The guitar in our example on the DVD is clean when set right and squashed with a fast attack.
3
6
From 1.5-6kHz, the upper-mid range (aka the ‘presence range’) deals with clarity, edge and intelligibility. Use this to make a part cut through the mix, but be careful not to overdo it. Here the vocal is boosted around 2kHz to help it dominate.
The compressor’s Release time is another key setting, controlling the sense of energy in a sound, as well as the smoothness of the compression. A fast Release energises a sound, but also brings up noises and makes compression obvious, which can be a good thing, as with the sidestick in the audio file on the DVD. This Bomb Factory plug-in offers fast Release times.
June 2010 / Computer Music / 27
Lend us your ears You’ve recorded the song of your life: the arrangement’s perfect, the instrumentation is sublime and the performances are out of this world. Now all you have to do is balance and blend the parts into a mix that concisely expresses the emotion and energy of the song, and that stands up beside the countless other excellent mixes out there. Mixing is all about your ears. You might know what an EQ or reverb is or how a compressor works, but if you don’t serve your ears with quality monitoring, it really is all for nought. To use your ears effectively, you have to have a pair of decent monitor speakers in a room that you’re very used to. By that, we mean a room in which you listen to other music and like what you hear and therefore know how a mix of your own music should sound in that environment. If other music sounds boomy or overly bright in your monitoring system, it’s fair to say that your mixes will come out sounding the opposite of that, as you’ll automatically compensate for such problems when mixing. None of this necessarily means that you have to buy an expensive pair of monitors – most hits of the last 25 years have been partly done on Yamaha NS-10s, which were relatively cheap. You will probably, however, need to use dedicated monitors of some kind, rather than hi-fi speakers, the difference being that the former are designed to give you an even representation of the overall balance of frequencies, while the latter are designed to flatter the sound by taking out the mid-range frequencies and enhancing the bass and treble. The midrange is key to getting a good balance, because the fundamental frequencies of most instruments are in this range, so it helps to be able to hear them clearly and evenly. Being able to hear clearly doesn’t mean that you have to spend thousands of pounds on acoustic treatment either, although some can help. Most music is heard in living room environments and sounds fine, so it should be possible to furnish your room and position your monitors and listening point in a way that makes sense of what you’re hearing.
28 / Computer Music / June 2010
> Step by step Using reverb and delay
1
3
5
Rather than use reverb on every channel as an insert effect, which could bring our computer to its knees, we use a shared channel (aka, auxiliary) into which we can send a bit of signal from any other channels we like via their Send controls. Here we send a bit of vocal via Send 1 to a reverb on FX channel 1. You can use other channels’ Send 1 knobs to apply the same reverb to part of their signal.
The most important parameter of a reverb patch is the reverb time (often referred to as RT). This dictates the size of the space and generally sits somewhere between 1 and 2 seconds. Big spaces – churches and cathedrals – will have RTs up to 5 seconds and beyond, while rooms might have RTs below 1 second. This vocal reverb has an RT of just over 1 second.
As you’d expect, any instrument generally sounds good put into its natural environment. Chances are you won’t have a hall to record in, so an artificial model of one is a good second best. Here, the acoustic guitar is placed in a hall reverb on send 3, which enriches the harmonics considerably.
2
4
6
We want a general purpose reverb for our vocal and instrument channels, and in this case, nothing works better than a plate setting. The plate reverb type tends to add subtle space and size to a sound, particularly vocals. An alternative to this, however, could be a hall or chamber setting. Try them all out and get to know the character of each one.
Long reverbs on drums are OK for slow tracks, but things can easily get too washy. Room reverbs tend to work better as a general rule, and the key parameter after the reverb time is the high-frequency damping. Stone rooms are brighter than wood rooms, which are brighter than furnished rooms, for example. Here, a drum room reverb plug-in is set up on auxiliary send 2.
With a long delay on FX Channel 4, we can send a bit of vocal into it for a classic echo effect. Try a quarter-note delay synchronised to the tempo of your song. We’ve inserted a reverb after the delay with a tweaked wet/dry mix on the former, so that the delay goes to outer space. We’ve also added an envelopecontrolled filter after that for even more atmosphere. There’s no limit!
the beginners’ guide to mixing / make music now < > Step by step Adding interest with plug-ins
1
3
When mixing, you’re always looking to add character to your track. Filters are great for this, and an oft-used trick is to filter the drums into a narrow band in the intro before opening them up as you go into the main song. The example on the DVD uses a vintage-style filter, and an automated By button switches the tone back to normal for the verse.
Synth sounds are often too clean and tidy out of the box and benefit greatly from a bit of roughening and distressing with distortion. Any number of plug-ins are available for this and you’ll develop favourites through experimentation. Here, Audio Damage Kombinat is dirtying up our very clean flute sound, not to the point of nastiness, but just enough to give it some edge.
2
4
Getting to know you
You should be striving to add size to the mix, and one way to do that is to broaden the stereo width. Here, the acoustic guitar sounds big and wide, which was done by putting it through a stereofier: Air Width in Pro Tools. However, a stereo pitchshifter set 5 cents up on one channel, panned left, and 5 cents down on the other, panned right, is good for widening if you don’t have a stereoiser.
Always look to up the energy of the mix if things start sounding pedestrian. A synth pad of held chords that seemed like a good idea during recording/writing might need pepping up. Making it pulse in time with the track on eighth- or 16th-notes, either on its own or mixed in with the original synth, will add drive to the part. Here and on the DVD, our Wurlitzer is getting the pulsing treatment.
POWER TIP
>Automating parameters
5
You’re also looking to add specific moments of interest to your mix. Any good dance track will employ a sliced beat or two and Audio Damage’s Replicant is a randomisable slicer that’s perfect for livening up a tired drum fill in a mix. It’s so random, in fact, that it’s best to record it and edit the audio afterwards. Here, it’s livening up our intro drums.
Very obvious effects get boring after a very short time but can bring welcome momentary change to a mix – for example, a long delay on a specific word or a big reverb on the last drum beat. Using the power of effects automation, your mix need never have a dull moment. Riding up the delay send on the vocal channel for one word or increasing the level can all be recorded into your DAW’s automation system. Similarly, filters on drum loops can be programmed to open up for instant fills.
The world of effects plug-ins is massive, and with so many at your fingertips, the danger is that you won’t get to grips with the fundamentals of any one in particular. While presets are great to get you started, time spent getting to know the ins and outs of an effect can enable you to be more precise with your intentions and more original with your results. Here are a few starting points to consider.
Time delay effects This covers a wide range of processors beyond the plain old echo, including chorus, flanging, phasing, doubling and slapback delay. The first three use short, varying delay times to create a sweeping sound with enriched harmonics, which can be useful in a mix to thicken sounds. Chorus is especially rich, but can tend to make things overly soft if used too liberally. Short delays can be useful to widen a mono sound into stereo. Doubling involves a slightly longer delay time and is often used on vocals to thicken them up. Slap-back is longer still, and is another great vocal effect to consider instead of reverb. It was used a lot in 50s rock ’n’ roll and by John Lennon. It’s usually best to run all time delay effects on auxiliary sends, because they require a mix of the dry and effected signal.
Filters You’ll no doubt be familiar with filters from having played around with soft synths, and they’re basically extreme EQs. As insert effects, they can radically change the tone of something if it’s getting in the way or just doesn’t sound exciting. You can add some interesting movement to a mix by modulating the filter cutoff frequency, either cyclically with an LFO, perhaps in time with the session tempo, or according to level with an envelope filter.
Distortion One of the most useful processors when it comes to bringing energy to a flat or overly clean sound. Usually used as insert effects, distortion plug-ins come in all shapes and sizes: amp simulators, stompboxes, bit crushers and more. If a mix is sounding too light and polite, a few judiciously applied distortion plug-ins can bring the edge and darkness you’re looking for. June 2010 / Computer Music / 29
The song remains the same Vocals are the most important feature of any song-based mix. They have to sit on top of the mix at all times, at a level loud enough that they can carry the tune but not so loud that the backing track sounds small. It’s worth making a comparison with reference tracks to get a good working level. You’ll notice that in a professional mix, the backing vocals allow the lead vocal to be more prominent without appearing isolated, and effects are used to glue the vocal to the track. The tone should be bright but not harsh, and the vocal should be clear, with a full body. A touch of delay or reverb will make it sound bigger and sit it in the track.
These days, people aren’t as tolerant of wayward tuning, so a pitch correction plug-in is useful to have to hand in case any off-pitching rears its head. Any backing vocals are there to the lead and accentuate the importance of specific sections. They don’t need to be as bright as the lead vocal, but they do need to be fully legible. You can probably take some of the mid-range out, especially if your backing vocals are multitracked, to avoid them getting too thick. Listen for out-of-sync ‘esses’ or extraneous noises in the backing vocals and cut them in line with the lead vocal to make them less distracting.
Vocals are the most important part of a track, so they need to be very prominent in the mix
> Step by step Sitting the vocal
1
A high- filter is often useful as the first insert effect on a vocal channel. Set to 80Hz, it rolls off general rumble from the room without affecting the warmth of the vocal. Vocal ‘pops’ can be blitzed by automating the cutoff frequency up to 250Hz wherever they occur, but listen carefully to make sure it doesn’t sound too thin.
2
No vocal should be mixed without compression, and saturation can be a real boon. A compressor with medium Attack and Release (auto release can be smoother if your plug-in offers it) and 4:1 Ratio holds the level, and something like PSP’s amazing VintageWarmer plug-in can drive the signal to saturation, upping the energy and intensity while holding it at the top of the mix.
3
EQ is essential. Boost around 2kHz for intelligibility and to cut through the mix, but don’t overdo it. A boost from 8kHz upwards will add air and sparkle to a dull voice, and if the voice sounds thin, add a touch at around 500-800Hz. If it sounds overly hard, lose some around 1kHz. If the ‘esses’ are sounding overly sibilant, insert a de-esser after the EQ, but be careful not to give the singer a lisp!
POWER TIP
>Mic check Before doing any processing or balancing, check the vocal thoroughly in solo for smooth edits and noises. It might seem like a waste of time, but this can actually save time in the long run. Unnatural sounds and spurious noises will be exaggerated by the compression and balance of the voice, so check that all edits are crossfaded. Listen for a natural, smooth delivery, as if it was one complete take, including natural breaths and pauses. Clipped breaths can be particularly problematic if left unrepaired.
30 / Computer Music / June 2010
4
There may be occasions when you need to automate the EQ for the tone to remain consistent throughout the song. If the singer moves in to the mic for a quiet verse, the tone will become bassier, so roll off low frequencies from 150Hz for those sections. Similarly, in very loud sections the vocal might sound too hard, so a 1kHz dip can help. Here the low frequencies are being ridden in the bridge.
5
The vocal level in the mix is critical, and even half-decibel changes make all the difference. Work through the song, recording fader level automation. Real-time fader rides (ie, raising and lowering) with a mouse or MIDI controller can be tweaked with on-screen changes for fine adjustment. Ride the lead vocal, then balance the backing vocals with it. This is one of the last things to check.
Drum compression The drum kit is truly transformed with judicious compression. You always want it to be punchy and powerful. Individual drums like compression but it’s easy to get it wrong. There’s a balance to be had between the attack of the hit at the front of the sound and the weight of the drum behind it. The key is in getting the compressor attack time sufficiently slow that enough of the transient hit gets through for impact, setting the threshold level so that it punches but isn’t strangled, and getting the release time fast enough that the decay pumps back at you with power. It’s a fact of psychoacoustics that we hear shorter sounds as quieter than
longer sounds of the same decibel level, so effectively, the longer the drum decay, the louder it will seem. With miked drums, compression can cause problems, because the spill from other drums and cymbals (especially the hi-hat) is increased, creating nasty, confusing tones. At the same time, though, this can be a good thing with overhead mics and room mics, as the compressor tends to ‘glue’ the kit together – the kick drum pumps the cymbals, for example. This isn’t an issue with samples and drum machines, which is one reason why it’s hard to unify the sounds of a sampler. As demonstrated below, sub-grouping drums and routing
The drum kit is one instrument that really benefits from judicious compression
them to a flat and compressed bus (parallel compression) gives you the best of both worlds, combining punchy compression with the
dynamics of the uncompressed sound. A good technique is to bring the compressed group up gradually as the mix develops.
> Step by step Mixing drums
1
4
The kick drum has to be punchy, weighty and tight. Some compression with medium Attack and fast Release settings gives it punch. To give the kick more front or click to cut through the mix, we boost at 2kHz. We also boost at 70Hz for weight and 120Hz for power. Notching out 250Hz helps with the clarity.
The level of the hi-hats is crucial, because it helps motor the rhythm track. Depending on the material, it can be good to add some spitting aggression to hi-hats using fast compression. EQ is often used to roll off low-frequency spill from the kick and other drums, and to add some edge around 6kHz. On ballads and ‘nicer’ tracks, you can also bring out the more extreme top end over 12kHz.
2
5
Hopefully your snare was recorded using two mics – top and bottom. Top is for the power of the drum, bottom is for snare buzz. Snare compression is similar to that of the kick, although the bottom channel can have a fast Attack. Balance the two, EQ in some presence around 2-3kHz and maybe open the top end with a shelving boost at 6kHz. A boost around 200Hz helps with fullness and weight.
When mixing a drum kit, treat it as a whole single instrument, with the room mics and overhead mics as the most important elements, and the close spot mics as in-fills. With each of these we’re trying to get the whole kit sounding good in a space. Extreme compression on the ‘room’ can energise the whole track.
3
6
Because toms are hit so infrequently and can ring on in the background, edit around them. Compress them like the snare drum with fast release, EQ up the sound of the skins between 2 and 6kHz, and boost between 80 and 140Hz for weight. Notching out around 400Hz can open toms up and make them less woody.
If your DAW enables it (most do), parallel group compression of the whole kit can be very useful for keeping the punch of the drums together in the mix. Here we’re using a send rather than sub-groups, which isn’t ideal because of the stereo image, but it works.
June 2010 / Computer Music / 31
> make music now / the beginners’ guide to mixing > Step by step Mixing bass
1
It’s really important to keep an even and constant weight to your bassline. Use compression or even full-on limiting to maintain a ceiling level. A medium Attack will keep some transient impact in the playing, and a medium/fast Release will maintain a natural feel.
2
It’s hard to find space in a busy mix for clarity at the low end. An amp sim drive with a touch of distortion will impart some bite. Of course you can go over the top with distortion if you like, too. The bass in our example was recorded with drive but we’re adding more with an amp sim.
3
We’re boosting everything below 120Hz with a shelf EQ for more weight. To make it less cloudy, we notch out a narrow area around 200Hz. For definition, we boost around 1kHz. It might seem odd, but a filter roll-off below 50Hz can tighten the sound and raise the perceived volume.
> Step by step Mixing guitar
1
With electric guitar, always record the DI signal along with any amp mics, so that you can re-amp afterwards. This session was just recorded with the DI signal straight into our DAW, so it needs to be treated through an amp sim, as shown. We’re using a classic Marshall emulation.
2
The old amp spring reverbs were good for character. Here we’ve mixed a bit of spring reverb in with a touch of slapback echo, set to around 120ms delay time and a single repeat. The echo works better when the playing is more sparse – when things get busy, it can get confusing.
3
It is possible to get the right tone just using an amp, but usually EQ is necessary to fit it into the mix. In this case, we’ve added some bite in the presence range (around 2kHz) for cut and edge, and cut a touch of lower mid-range to open the sound up.
> Step by step Mixing keyboards and synths
1
Pad sounds are generally legato synth parts that fill the mix out, usually in the lower mid-range. They don’t get in the way or draw attention, but can be more prominent if it suits the track. Here and on the DVD, a Wurlitzer part does the job of a pad, warming and broadening the sound.
32 / Computer Music / June 2010
2
A lead line should take over from the vocal at the forefront of the mix as the focal point. Getting the level right is key, and the automation should be ridden in the same way as with the vocal, to keep it in the right place. In our example, we ride the piano solo out of the mix.
3
Incidental and subliminal noises are used to add interest and mystery. Generally they’ll be at quite a low level, but occasionally it’s good to have an odd noise jump out of the mix to surprise the listener. In our example, the reverse effects do that job.
> make music now / the beginners’ guide to mixing
Getting the balance right Once all your sounds have been processed individually, they need to sit together. You might have balanced them to some extent as you worked the solo parts, but you may not have been really listening to the big picture as you did it, so it could well be skewed. It’s worth muting everything and throwing up a fresh balance quite quickly so that you don’t over-focus – thinking too hard about the mix can turn it into more of a cerebral project than an aural one, which absolutely isn’t the way it should be. Any mix balance starts by establishing the foundation, usually the rhythm track – drums and bass. Get a good groove and the mix will look after itself to a certain extent. One by one, introduce the harmonic elements, the bits that articulate the musicality and mostly occupy the mid-range. At a reasonable monitoring level, set them against the groove, consider where they should sit in the stereo picture and pan accordingly. , you can automate the panning if an instrument needs to take centre stage for a section, and things can move around, too. Every now and then, drop the vocal in to make sure there’s still space for it.
“Imagine that your mix is leading the listener, someone who doesn’t know either you or the song” Check that the frequency balance is good – not too dull or too thin – by referring to other tracks. A common fault is for the mix to be too dull, so you might have to push the levels of the brighter instruments: hi-hats, acoustic guitars, etc. You may need to tweak individual EQs along the way, too. Imagine that your mix is leading the listener, someone who doesn’t know either you or the song. It needs to be exciting and unpredictable, but not jarring, and you need to provide a focus throughout, be it a vocal or lead instrument. To achieve this, use mix automation to ride the relative levels. Try riding levels around section changes for dynamic excitement, too, and make sure the change builds or falls in the right way.
34 / Computer Music / June 2010
> Step by step
1
3
5
Balancing and automation
Mute everything, then, part by part, starting with the rhythm track, build a mix with no dynamic automation. This static setup should give a good representation of the song with an even tonal balance. Check it against reference tracks to make sure it’s bright and full enough, and do any tonal tweaks to individual EQs as you hear fit.
When musical changes are approaching (the transition from verse to chorus, for example), consider how you can telegraph them and/or build them up to create a dynamic moment and, after the change, ensure the mix lives up to that build. Bridge sections before a chorus should do as the name suggests, serving as a musical link between it and the verse.
Another final job is to check the beginning and end of the mix. Listen to the front for noises and leave a bit of silence before the start for later mastering work. Check the end of the mix for any fade or ‘hangover’ noises you don’t want. You can always leave the fade-out until later if you’re not certain.
2
4
6
Exploit the stereo soundstage as much as possible to create width. Bass instruments and main drums tend to stay in the middle of the mix, but there are no set rules and odd panning can work as a pleasant surprise sometimes. Automated panning is an option, too, for swirling instruments or bringing featured moments into the centre.
One of the last things to do before committing to a mix is listen through to the vocal level and any other principal parts. Listen at a normal level and refer to other tracks to make sure the vocal is dominant enough and not too imposing. If necessary, apply manual automation changes.
Most DAWs have a ‘bounce to disk’ option. Set the start and end point, and bounce away. Check that the mix level doesn’t go into the red at any point. If it does, you’ll have to work to bring everything down proportionally – save a new version of the project before you do that, though. Do a final check of any mix compressors or EQs and you’re done. See you in the charts!
> make music now / the beginners’ guide to mixing
mixing tips and pitfalls to avoid REST YOUR EARS After only a few hours of concentrated listening, your perception shifts and wrong decisions get made. The beauty of instant recall with software mixers is that you can work for short periods of time on a mix and come back to it fresh, even if you work on other tracks in between. The best work on a mix is usually done in the first hour.
DON’T KEEP RAISING LEVELS It’s all too easy to get into a vicious circle of turning one thing up after another as parts compete to be heard. Instead, try turning things down and use EQ and panning to create separation and space.
OVER–EQING Similarly, there can be a tendency to keep boosting the presence range (around 2kHz) on the EQ of everything because that also increases the subjective loudness. This just leads to a harsh, thin mix, though. You’re looking for parts to be spread evenly across the audible frequency range. That way, elements don’t compete and the overall sound gets cumulatively bigger.
If your EQ curves look like this, you’re very likely overdoing it! Boosting the 2kHz presence range might increase perceived volume but it will grate over time, so go easy with those knobs
A/B YOUR MIX Have Winamp, iTunes, Spotify or your other media player of choice running in the background with a playlist of reference tracks open – ie, tracks that you ultimately want your mix to sound like. Constantly referring to them can make certain you’re keeping to the path. If you think your mix is better than the reference, though, you’re almost certainly wrong: you’ve probably not matched the volumes, or worse, your monitoring is set up poorly.
LISTEN AWAY FROM YOUR ROOM It’s always good to try and get a broad perspective on the mix, so regularly make a copy and listen in another room and/or your car. Listening with other people also helps you to hear it differently.
YOU CAN’T MIX ON HEADPHONES It’s nigh on impossible to mix entirely on headphones and get good results: it’s just not a natural way to hear music. They’re good for checking the finer details of a mix, in
particular panning and reverb, and for hearing clicks and noises, but they’re tiring after a short time and damage to your hearing is always a danger.
LIMITING Some people mix through a stereo compressor from the off. It can make balancing easier by controlling the relative levels but we’d recommend getting a good natural balance and leaving the compression or limiting to the mastering process, when you can really focus on getting your mix loud. That said, it can be useful to have an inserted mix limiter available for quick comparison with your reference tracks, which will, of course, have been professionally mastered.
SAVE AS YOU GO Red herrings are common when you’re mixing and you’ll sometimes come to the realisation that it sounded better a few hours/days/weeks/months ago. You should regularly and sequentially save new versions of your project as you mix. If you bounce a rough mix to check outside your studio, always give it a name and save a session file with exactly the same name so that you can go back to it easily.
THE BASS ISSUE The hardest thing to get right in a mix is the bottom end, which is key to a good overall sound. It’s also very hard to hear on a small monitoring system, but you can usually trust car stereos and hi-fis to tell you if there’s too much or too little bass.
PHASE CANCELLATION Keep your media player like Spotify open so that you can refer your mix to commercial tracks in your genre
36 / Computer musiC / June 2010
Be aware that what might sound really wide and ‘outside’ the speakers in stereo could disappear altogether when collapsed to
mono because of phase cancellation. Although mono isn’t used much these days, it’s a good idea to maintain compatibility anyway, so that your mix is always true, no matter what it’s played back on.
RADIO WAVES Ask yourself: ‘Could I hear this track played on the radio?’. This is a good way of detaching yourself from the mix and hearing it objectively. It’s got to excite other people who aren’t connected to it in any way, and it has to have more fizz and excitement than you might think.
DON’T OVER-MIX Mixing isn’t about getting the perfect balance. Spend too much time on a mix and you risk killing it by smoothing off all the edges. As you near the end of mix, see how dangerous you can be with the levels of featured parts.
MONITOR VOLUME All but the worst mix will sound good and balanced at high volumes because your ears compress it into shape, so don’t kid yourself by monitoring loud all the time. Louder listening may be appropriate when toning individual sounds, but monitor at a reasonable level when balancing: it’s truer and less tiring. If you feel an urge to turn it up all the time, the balance or brightness level could be wrong.
ON THE LEVEL If you find yourself pulling your master fader back a lot, you could be overloading the mix bus. Start your mix with the channel faders at around -12dB to give you headroom as the mix builds, and be prepared to scale them all back as a group if they creep up too far.
Related Documents 2w1qw
Mixing 29266x
February 2021 0
Mixing 29266x
October 2020 0
Mixing Ppt 4m451x
November 2019 42
Makalah Mixing 5y2ok
April 2020 20
Internal Mixing x2657
May 2020 4