Thursday, 28 April 2011

The Benefits of Referencing Your Mix

Time to think about referencing your tunes. Ever find that when you're getting your own tracks in the mix, or playing them out in a club, somehow they just don't sound quite right when you put them up against released tracks? Even though they sounded fine back at home? You're not alone! Let's dive in and look at how you can make sure you're getting the right balance...


When I say 'referencing', I'm going to be talking about two different things; there's listening to your track on different speakers and in different environments, and then there's comparing your track with others.

First though, before you even get to that stage, you should take a break. If you've been mixing down all evening, your ears will have gotten used to the track as it stands; anything you do will likely just make it sound 'wrong'. So go for a walk, listen to some completely different music, watch a bit of TV. Better still, leave it and come back tomorrow. Then, load a bounce of the track into iTunes or Winamp, and sit back. You'll be better able to appreciate what the track is actually doing if you can't see the mixer or the arrange window of your DAW; the brain has a funny way of rationalising everything, and it really helps just to listen to the file in a different program than usual.

Once you're happy with your mixdown so far, then you need to fire up some tunes that you think your track sounds similar to - or that you want it to sound similar to! Make sure you've got the volumes matched up so that they all sound the same level, and flick back and forth to compare the differences in the mix (you might find it's easiest to do this in your DAW so you can mute and unmute channels). There are some obvious things to listen out for, like the basic issues of a mixdown; is your snare much louder or quieter than another track? The vocals? On a more subtle note, you should listen out for the overall balance of the track. It could be that although everything sounds good individually, the higher frequency elements of your track are quieter, relative to the bass - which might mean they don't come through so well in a club. Or vice versa, of course.

If yours sounds somehow 'harsher' perhaps, but you can't identify which element is causing it, then it could be that you've been boosting high frequencies on several channels (again, even if they sound fine on their own), and would be well advised to go back in and tweak them.

Pay attention also to whether the other tracks use any cunning techniques with the arrangement - often you'll notice that a track may create space for a vocal by dropping out some lead lines or high frequency percussion; or will program the bassline so that it's not playing at the same time as the kick. This can be a useful way to avoid cluttering the mix, or avoid it sounding 'top heavy' for instance.

Many people find it helpful to use some kind of graphical frequency analyser when mixing down and referencing, and it can certainly help - if you look at one track which seems pretty flat, and yours has a strange spike at a certain frequency, then you may need to consider what is causing it. But equally, you don't necessarily have to go after every little peak in the spectrum - your track has to have some kind of sonic focus after all! You don't want to be mixing all the character out of your beats. And you should always try to trust your ears, rather than what you can see on the screen.

Once you're confident that you've got a balanced and comparable sound, you then need to go and take your new beat, and your reference tracks, and listen to them on other speakers. This is vitally important, as so far you're going to be working around the imperfections of your room and speaker set up. So there could be some issues you've not spotted yet.

If you can, try and listen on a small setup which doesn't have much bass, a bigger system which has plenty, and maybe in the car - car stereos can vary enormously. Some common things to look out for would be the low end - especially on the small system. Can you still hear the presence of the kick or sub on some tiny speakers? At the top end, does your lead or vocal still stand up clearly amongst the tinny rattle of the rest of the track? Is it the same with the released tracks? Similarly, on some boomy speakers, you want to make sure that the low end still has clarity, and doesn't drown out everything around it.

It's a bit of a long and often quite boring process, and one that will mean you get through plenty of blank CDs. But it's necessary if you want your tracks to hold up well against the best out there, so get a notepad and pen, and start listening closely. You may find your mixes start to improve quite quickly!

Tuesday, 19 April 2011

Be Inspired by "Found Sounds"!

Any producer will know the feeling. You sit down in front of your sequencer, fire up a template, and sit there staring vacantly at the screen, as all the ideas disappear from your head. You're stuck for inspiration. What to do? Here's one angle that could get you moving again...

Before you start producing a track - in fact, it's best to do this kind of thing when you're not feeling inspired or creative, so that you can get on the case more quickly in times of creation - why not try amassing an army of 'found sounds'?

When people mention 'found sounds' it's often easy to imagine that we're talking about avant-garde, experimental noise stuff. But it doesn't have to be like that - they can provide a brilliant source of ideas, details, glitches, rhythms and more, which may give you the starting point you're looking for.

The most common way is to get hold of a cheap sound recorder and take it out and about. This is a great way of getting atmospheres, background textures, incidental effects that you wouldn't otherwise find. Take it to the subway station, walk up a busy street with it, go into the fields and run a stick along a fence. This kind of technique is something that Eskmo used to awesome effect on his recent Ninja Tune album - the result being that it has a very distinctive atmosphere.

If you don't have the cash for a decent field recorder, then just try using your phone - or even the webcam on your laptop. It maybe lo-fi, but what we're after is ideas more than sound quality. So get recording! Back in the studio, time to get the mic out - if you have one. Get anything you can to make noise - hit things with sticks, slap the cat, make weird noises with your voice. If you don't have a mic - make one! Have a search on line for 'contact microphones' - these things can be made for literally pennies from your local electronics store. You can stick them on anything and they'll record the vibrations; put one on the side of a balloon and blow it up to make yourself sound like Darth Vader, put on on the side of a beer can and rip it up to make sounds like a giant car crusher. You can also just stick one on the wall and record yourself making noises like it was a normal microphone. The world is your oyster.

Once you've got plenty of sounds - lets say 15 or 20 minutes worth - then it's time to get busy making them into something a bit more useful. Some, like an atmosphere from a street, can be dropped into a track straight off. Others may want processing, and luckily there's a whole host of plugins to help you out. Try some extreme timestretching and reverb to create sci-fi atmospheres. Or, try using an 'audio quantising' process, such as Logic's Flextime on sounds that have a clicky, glitchy, percussive aspect, to make some crazy rhythms that you can then build a beat around.

Made a crazy metallic weird sound with your contact mic? Stick an autotune plugin on it, to get a melodic riff you'd never have thought about making otherwise! If you have some glitches, try leaving them unquantised and just drop them into a beat, to give a random swing and groove. Any of these tricks can give you ideas to get moving on a track. Or, if you've got something going already, can help you lift it up from the standard beats and bass vibe, into something else entirely. And the best part is - they're unique to you! You can bolster your samples with sounds that no-one else has, and create a signature style.

So, next time you're sitting at the computer wondering what to do, don't just give up and load facebook - get started making your own inspiration. You never know where it could lead...

Wednesday, 13 April 2011

Let's Get Deep! Using Sub Bass in Music Production

Sub bass these days is one of the most essential aspects to a dance music track - and now that we have decent sound systems and quality rigs in clubs, it's ever more important to get it right. But what's the best way to do it? And what makes a good sub anyway?


We all know what sub-bass is - it's that bedrock of your track, ranging in the 40 - 100 Hz region; either underpinning your bassline or kick (or both). But there are plenty of misconceptions about it, and how you should treat it, so lets clear a few of these up now.

First, a simple sub-bass is a sine wave. That's all. It makes things nice and easy, because there's no need to worry about what would be a good synth to use for sub bass; just load up a sampler with no patch, and use a sine from that. Done! We could go into the maths of it, but essentially any other type of wave (lets say a square wave) - has higher frequency harmonics in it; on a square wave these appear to be the 'corners' of the square. Low pass filter a square wave and you'll be left with a sine again. If people talk about a 'fat' sub, then it's probably because there are higher frequencies in the sound. You might want some of these harmonics though - we'll touch on that shortly.

Because sub bass is such an 'extreme' frequency, there are plenty of technical issues that come into play. A lot of them arise from the limitations of vinyl or club systems. You need to remember that as frequencies get lower, the wavelength gets longer; this means a speaker (or needle) has to move twice as far to create the sound, and will need much more power to push all that air. As such, it's a good idea to try and keep your lowest frequencies above about 40Hz, which is around the notes of E1 or F1. Below this, at the kind of volumes dance music requires, it starts getting increasingly difficult to cut to vinyl (the grooves need to be further apart, limiting the length of the track) - and not all club systems can handle the extreme lows anyway. Your author has first hand experience of playing one of his own tracks in a busy club, and finding that it sounded like there was no sub-bass on the track whatsoever. The sub (on the note of D1) was too low for the club system to handle; it sounded terrible!

Another issue you'll see discussed is that of panning. The accepted wisdom is that you need to have your basses in mono, and this is a sensible option to go for. Vinyl can skip and jump if the deep bass grooves are significantly off to one side; and anyway sub-bass frequencies are below the level that you can reliably tell where they're coming from (the wavelength is longer than the width of your head, for a start). A mastering engineer or vinyl cutter may well decide to mono a mix anyway below around 300Hz if there are any problems with stereo bass - so you'll need to make sure (as always) that your mix sounds good in mono too. And again there can be club problems - many club soundsystems are mono anyway, but even if not, a sub panned hard will only be coming from one set of bass speakers. And that's going to sound weak.

Besides the technical, engineering aspects, you also need to write your subline sympathetically to the rest of your track. This means, for a start, have it monophonic - clashing notes will sound muddy and use up headroom. So make sure the 'release' on your ADSR envelope is tight, to ensure that notes don't overlap. If you're using a subby kick, then try to make sure that your sub isn't playing at the same time - or if it must, sidechain it off the kick to duck the volume, again for headroom purposes.

To return to the subject of harmonics, this is something that can be useful for a number of reasons. You can add a sense of warmth, or 'fatness' by adding in frequencies above the standard sub region - we're talking 150 to 400Hz here - that will give things more of a glow. But, arguably more importantly, they'll help the sub come through on smaller speakers. A radio or an iPod dock will struggle to reproduce frequencies right at the bottom end; not so much of an issue if the top end of your bass is a raging wobble dubstep affair, but if you're writing more of a roller, then it may sounds like there is no bass at all. Get some higher frequencies on your sub and people will be able to hear what it's doing, even if they can't feel the weight. A simple way of doing this is to set up a send to a buss, and then distort it and bandpass filter it around 200 or 300 Hz. This way it will copy the original sound, shouldn't interfere with the low end (as you've high passed it) and will give extra interest for the listener.

Sub-bass is a tricky subject, and it's something that requires practise to get right. But learn the basics and you're halfway there. So, jump straight in - fire up your speakers and get writing those big basslines!

Friday, 8 April 2011

It's a Take! Prime Loops Guide to Vocal Production

With the ongoing love affair between pop, hip-hop and dance music, there has never been a time when the vocal track has been so popular in underground dance music. So read on for some production tips on how to fit a vocal into your own beats...

From dubstep to deep house, from layering an acapella on your future garage track to recording a full session, everyone's looking to vocal music in 2011, and it's not hard to see why. With dubsteppers like Rusko and Benga troubling the charts, and producers like L-Vis1990 grafting R'n'B vibes onto house tracks, we may even have a genuine trend on our hands.

So what are the main points to make your vocal stand out? There are a few things that need to be addressed, but first off, the mistake that your author sees most often is that the vocal is simply too quiet. Pay attention next time you're in a busy shop, for instance, and a Rihanna or Beyonce tune comes on to the crappy house system. Over the background noise, what can you hear? Oftentimes, all that will be audible is a vocal, a hihat, maybe a little string line. The brightest elements, all pushed to the fore. If you're writing a vocal track, then the vocal will be what people latch onto, the bit that gets stuck in their brain, they'll be remembering the words. So you need to make it easy for them to pick out the vocal from the rest of the track. Err on the side of volume!

On a related note, you need to make sure the vocals are crisp and bright; plenty of top on there. This is less essential for backing vocals and harmonies, but in general you need to remember that your lead vocal will be jostling for space amongst the hihats and lead lines; so it needs to have plenty of high frequency content. In addition to vocals you've recorded yourself, I often find that acappellas downloaded from the internet can be recorded from vinyl, and lose some clarity in the process, so even with ready-recorded vocals you might need to add some top back on. Similarly, if you're working in strongly electronic fields, you'll need to compress the vocal fairly heavily. Of course, sometimes you want a natural feel to the track; but in this age of slamming basslines, searing leads, vocoders and overdubs, 'natural' tends to be restricted to Norah Jones records. You need the vocal not just to sit up front, but stay there, so get that compressor fired up. Fast attacks, medium release, hard knee, taking off anything up to 10 or even 15dB.

If you've recorded several takes, then it's good to layer them up to get a thicker, bigger sound. Pan a couple centrally, then have one to the left and one to the right for a really full atmosphere. But - and this is a big, tedious, important but - you need to make sure they're fully in time with each other. That means down to the level of individual words; even syllables - because if they're not, the result will just be a mess. Like having your drums slightly out of time, the impact will be lost. So this means you may need to go right in and edit your vocals word by word to line them up. It's a long and drawn out process, and possibly one of the most boring procedures in music production, but it's necessary.

Effects can be very useful on vocals, and used well can really bring a track to life. A good example is Magnetic Man's track 'I Need Air'. A fairly straightforward vocal, mainly on one note, is given a metallic texture in the verses with heavy use of a vocoder, and in the choruses with reverb and delay; while variety is provided with flangers, phasers, glitched edits, extreme reverb settings and so on. Without all this processing the vocal would be in danger of becoming dull; especially as it's a fairly wispy sounding recording with no backing tracks and few adlibs. Luckily, clever production techniques keep it engaging until the end.

Effects can also be used to 'rescue' a vocal somewhat - subtle use of an autotune can correct out of tune singers, chorus can beef up a thin sounding vocal track, and one of the most popular recovery tools is a good de-esser. Use these to cut out sections when your vocalist is overemphasising 'ss' sounds to create a harsh sibilance. A de-esser is essentially a frequency sensitive compressor; and getting good results is rather an inexact science.

The Sony Oxford de-esser comes highly recommended, but if you can't afford one of those then try your DAW's native option. Sibilance is usually around the 6-8kHz region, but you'll need to do some fine tuning. If it doesn't get the required result, then you can fall back on the long way - draw in some volume or EQ automation by hand. You can even try a standard compressor, set up to sidechain off a buss, where the buss contains just the vocal track tightly bandpassed around the offending frequency range. It's a convoluted solution, but if it gets the result you need, it's as valid as any other!

Production of vocals is a big subject to tackle in one article, but it's important to get to grips with if you're going to write the next chart smash. So don't wait around - dig out that microphone and get moving!

Wednesday, 6 April 2011

The Benefits of Referencing Your Mix

Time to think about referencing your tunes. Ever find that when you're getting your own tracks in the mix, or playing them out in a club, somehow they just don't sound quite right when you put them up against released tracks? Even though they sounded fine back at home? You're not alone! Let's dive in and look at how you can make sure you're getting the right balance...


When I say 'referencing', I'm going to be talking about two different things; there's listening to your track on different speakers and in different environments, and then there's comparing your track with others.

First though, before you even get to that stage, you should take a break. If you've been mixing down all evening, your ears will have gotten used to the track as it stands; anything you do will likely just make it sound 'wrong'. So go for a walk, listen to some completely different music, watch a bit of TV. Better still, leave it and come back tomorrow. Then, load a bounce of the track into iTunes or Winamp, and sit back. You'll be better able to appreciate what the track is actually doing if you can't see the mixer or the arrange window of your DAW; the brain has a funny way of rationalising everything, and it really helps just to listen to the file in a different program than usual.

Once you're happy with your mixdown so far, then you need to fire up some tunes that you think your track sounds similar to - or that you want it to sound similar to! Make sure you've got the volumes matched up so that they all sound the same level, and flick back and forth to compare the differences in the mix (you might find it's easiest to do this in your DAW so you can mute and unmute channels). There are some obvious things to listen out for, like the basic issues of a mixdown; is your snare much louder or quieter than another track? The vocals? On a more subtle note, you should listen out for the overall balance of the track. It could be that although everything sounds good individually, the higher frequency elements of your track are quieter, relative to the bass - which might mean they don't come through so well in a club. Or vice versa, of course.

If yours sounds somehow 'harsher' perhaps, but you can't identify which element is causing it, then it could be that you've been boosting high frequencies on several channels (again, even if they sound fine on their own), and would be well advised to go back in and tweak them.

Pay attention also to whether the other tracks use any cunning techniques with the arrangement - often you'll notice that a track may create space for a vocal by dropping out some lead lines or high frequency percussion; or will program the bassline so that it's not playing at the same time as the kick. This can be a useful way to avoid cluttering the mix, or avoid it sounding 'top heavy' for instance.

Many people find it helpful to use some kind of graphical frequency analyser when mixing down and referencing, and it can certainly help - if you look at one track which seems pretty flat, and yours has a strange spike at a certain frequency, then you may need to consider what is causing it. But equally, you don't necessarily have to go after every little peak in the spectrum - your track has to have some kind of sonic focus after all! You don't want to be mixing all the character out of your beats. And you should always try to trust your ears, rather than what you can see on the screen.

Once you're confident that you've got a balanced and comparable sound, you then need to go and take your new beat, and your reference tracks, and listen to them on other speakers. This is vitally important, as so far you're going to be working around the imperfections of your room and speaker set up. So there could be some issues you've not spotted yet.

If you can, try and listen on a small setup which doesn't have much bass, a bigger system which has plenty, and maybe in the car - car stereos can vary enormously. Some common things to look out for would be the low end - especially on the small system. Can you still hear the presence of the kick or sub on some tiny speakers? At the top end, does your lead or vocal still stand up clearly amongst the tinny rattle of the rest of the track? Is it the same with the released tracks? Similarly, on some boomy speakers, you want to make sure that the low end still has clarity, and doesn't drown out everything around it.

It's a bit of a long and often quite boring process, and one that will mean you get through plenty of blank CDs. But it's necessary if you want your tracks to hold up well against the best out there, so get a notepad and pen, and start listening closely. You may find your mixes start to improve quite quickly!

Wednesday, 30 March 2011

Top 5 Reasons to use a MIDI Controller

For those of you brought up on a strict diet of mice and keyboards, the allure of a MIDI controller may not have entered your thoughts just yet. But watch out - this blog is here to entice you into the joys of hardware manipulation, without losing the convenience of software.

Cue the over-dramatic music to accompany the upcoming over-dramatic sentence: Here are 5 reasons why MIDI controllers could be exactly what you've been searching for your entire lifetime...

1. Speed up musical note input

Still typing in your notes like a sucker? Still clicking in every kick and clap? A MIDI keyboard or Pad-based input device could be exactly what you've been looking for.

Even if you aren't amazing on the keys or a pro on the pads, taking the time to watch a few free tutorials to get yourself up to a reasonable standard could pay itself back in kind over the rest of your lifetime in the amount of time saved putting in melodies and beats. You don't have to be a maestro - that's what "quantizing" is for! It puts all your midi inputted data in perfect time with the music.


2. More natural sounding instruments

Do you find your pianos end up too plinky-plonky, like an over-enthusiastic pre-schooler keyed them in? Rediscover the beauty of manual velocity control - if you play your notes in, you'll get a more human, natural sound since the velocities of each note will vary minutely.

This, combined with tiny variations in note triggering times, can sometimes be a positive thing when you're trying to make music that doesn't sound like it was made on an electronic device!

3. Improve your skills for live performance

If you push yourself to play in the more difficult phrases in your music, you can transpose those skills to the stage! More and more DJs are finding themselves dissatisfied with simply triggering their tracks - MIDI devices are the way to go to bring some talent back into the live music arena where electronic music is concerned!

Just think - would you rather watch your favourite artist stand in front of some spinning decks or see them play aspects of their tunes live?

4. Manipulate other parameters

Not only can you get control of the notes, but things like filter sweeps, envelope manipulation and wobble control become so much easier when you have some knobs to tweak!

Your sounds are infinitely more malleable when you have a hands on approach, and after some simple routing you can make your warm synth pads evolve over time, your lead synths modulate in or your basses wobble at different rates, all at the touch of a button, turn of a knob or twist of a dial.

5. Speed up your workflow

It's not just the note input it speeds up. Mixing with a MIDI controller could save you hours of time as you dont have to keep switching focus between the virtual mixer and the sequencer to adjust your levels. You can be applying an effect and EQing an audio track at the same time.

So if you're serious about music production, it's vital to make sure you are getting your ideas down as fast as possible so you can maintain that glorious creative flow. It's so hard to keep that up when you're too busy worrying about your virtual interface to think about that perfect rhythm or change those synth parameters just right. So streamline yourself - investing in MIDI controllers might be the best decision you made since deciding to make computer-based music in the first place.

Friday, 25 March 2011

The Science of Beatmaking - Part 3


In part three of our short series on building a quality drum break, we'll be considering what to do once you get to the latter stages of the writing process - mixing down, editing and adding the final touches to your drum track.

First, to make things easy in the mixdown process, you should consider bussing your drum sounds. If you're using a lot of different elements - like with layered hits, or different types of percussion, this will help you control the overall level a lot more easily. It is often useful also, to have several 'pre' drum busses; perhaps one each for your kicks, the snares, the high end, and 'other' (bongos maybe) - which then feed into one 'main' drum buss. This will mean it's easier to change the level of the entire drum track when mixing, or quickly apply effects to the whole group.

So, once you've bussed your drums together, what to do? Oftentimes people talk about compressing everything together to help things 'gel'. The reason for this is because a busy drum beat will have to interact with the compressor - a hihat on it's own will sound loud; a hihat playing at the same time as a kick will be reduced in volume, as the compressor turns down the gain. So the various sounds in your beat will shift about in the mix, giving a movement that wasn't previously there.

Compression should be applied sparingly though; it is the quickest way to make a punchy drum track sound flat and lifeless. A fast attack and low threshold will smear all the transients from your beat, and lose all that funky feel. An envelope shaper or transient designer, placed on the buss, can help add in a little 'snap' back to your beat, but it shouldn't be used to make up for a badly used compressor. Keep that gain reduction down!

Reverb is another tool to help things sound 'together'. The point of reverb is to make something sound like it's 'in a space' and so by putting everything in the same space, it can sounds more coherent. Try applying a touch of reverb across your whole drum buss; something small, a reverb time of around 0.3s, mixed very quietly. It shouldn't really be very noticeable as reverb except when you mute and unmute it, but it will help give a cohesive sound.

Reverb can also be a creative tool - consider the huge dubstep snare sounds (long reverb, large pre-delay, just on the snare channel), or you could try experimenting with the pre-delay setting on your reverb plug to give a rhythmic effect (set it to equal one quaver, to give a pumping, off-beat sound, for instance). Remember to EQ your reverb returns though; nobody needs to hear 100Hz in a reverb tail, it doesn't add anything to the effect, but it can add muddiness. So cut things down to only that which is necessary for your track!

We did a whole article on parallel processing recently, so look out for that, but it can be most useful with drums - you can solve the issue of losing your transients through compression by sending your buss to another buss with a compressor (or limiter, tape saturator or distortion) and mixing that in quietly with your original signal. This way you can compress as hard as you like to get that smack from the drums, without losing the dynamics of your main drum track. It's an excellent tactic, when used carefully, to get more power and volume from your beats. Remember also, when parallel processing, that if you're going to EQ the new channel, you should use a linear phase EQ. The science is too complex for a short article like this, but basically, EQ's change the phase of a signal, and if you're having two drum channels playing at the same time, you could end up with phasing problems. A linear phase EQ, while not being a perfect solution, leaves the phase intact - so use these in preference on parallel processing.

Once you've got your beat sounding pretty crisp, you may even want to do a spot of resampling. This is more of a creative process, but can add interest to your overall track; bounce the drums down, and reintroduce them to your arrangement. Then you can chop things up for short fills, or go crazy with the FX, reverse a whole bar, and so on. IDM fans may want to get busy with a glitch plugin - dBlue's cunningly titled 'Glitch' is one of the industry standards - just set it up on your newly imported file, bounce the whole drum track with the plugin on it so you get 5 minutes of solid glitching, then chop that up and pick out the best bits. Easy!
Finally, then, how to mix? Well, this can't be answered from here - it depends on the kind of track you're writing. If it's a jungle track, then you want the drums to be centre stage, right up front. If you've got a vocal, poppy track on the go, then you need the drums to sit behind the vocal, while still having enough clout at the low end. So, it's hard to advise, but the main issue is that you need to consider the desired effect of the track, and mix accordingly.

As mentioned in part one, we could write for weeks on this subject and still not cover every angle. But hopefully this three-part overview has given you some inspiration to start firing up the sampler and getting busy with the beats. May the funk be with you...