Lockjaw's Den

Production tips and thoughts on bass music. http://soundcloud.com/lockjawdnb

[lockjawdnb]

Equalization as a Mindset

It’s been a while since I've done one of these. I've thought for a while about something important and worthwhile to post about, something that could benefit all producers, from drum and bass to pop. In fact, I think this post could be useful to anyone involved in any sort of music production or recording.

Something I think about from time to time is the difference in the type of producer I am now and the type of producer I was 1 or 2 years ago. I still use the exact same plugins as before and the exact same setup but I know that overall, I keep slowly improving overall. I think this is due to a culmination of different things but one of the most important ones is one that I will relate to EQing. One of the most important things to realise and conceptualise when it comes to producing electronic music (especially on a completely digital format, such as myelf) is that we have a finite amount of space to work with as excessive frequency doubling and overlap is always bad news. In other words, you have to think about everything you put into your production and sonic landscape as if it were a puzzle.

Most people are familiar with using EQ to treat and improve the sound of single tracks and/or samples. What I propose in this article is the adoption of applying this mindset to the entirety of production and using it to gain a deeper and more effective knowledge of production as a whole and to use it to gain a more holistic view of one’s productions, therefore resulting in a clearer, more coherent end product. What I suggest is that one realises that the concept of equalisation is effectively one and the same as sample selection, plugin use and arrangement/songwriting and basically all components of production in that they all accomplish the same thing EQing does; select and attenuate or emphasise different frequencies. I will illustrate this with some examples.

Imagine you have just started a new project and are designing a drum loop. From the start, you are faced with some options and should be asking yourself; will my kick be round and subby? Or high and clicky? How punchy will it be? Will my snare be high pitched and woody? Or low and thumpy? Will my hats be digital and crunchy? Or more natural and washy? This method of seeing components of a track and their important characteristics as separate frequencies and choosing/building them accordingly based on what you know you will need is critical. Imagine you are now making your bass synth. What kind of midrange does it need to fit your drums? Your kick is subby so you want something with a more lilting subbass and defined midrange. Are your hi hats are super washy and big? Maybe the treble of the bass shouldn’t be the focus then and clash with them. You’re trying to make something cold and digital for the bass, so how much midrange should you keep in or scoop out to get that desired sound? Trying to make something more “old-school”? How much low mid energy should there be to make it sound fat but not muddy? Is the distortion plugin on your bass unknowingly adding far too many harmonics in this area? Maybe a duplicate bass instance with a bitcrusher on it and the low end scooped out will result in the sound you need? Now imagine making a pad to go under all this. What sort of waveforms should I use? Saws? Squares? How will filtering this pad affect where it sits in the mix? Is it taking up space where the bass is? I’m going to send the pad to a reverb send, how should I EQ the reverb? Should I sidechain it to the kick? Is the rhythm of what my pad plays fitting in with the rhythm of the bass? Does it distract from the important frequencies of the bass? What I’m trying to illustrate here with these basic examples is the idea of keeping the entirety of your mix in your head whilst you are producing sounds to fit into your song. Every decision you make weather it be deciding what samples to choose, how to use every plugin or doing work on a synth and even deciding how to arrange and write the actual music should be related to this idea of “global EQing” because ultimately, you are manipulating sound and that means manipulating frequencies which is precisely what equalisation is, weather it be through changing a sound by distorting it, EQing it, using a filter, using volume or anything else. I’m not sure if I’ve explained myself thoroughly here but this concept is clear in my head and I know that it is the central tool that helps my productions become clearer, more powerful and effective as time goes on. Things start to click when you realise and can identify what different frequencies in a sound mean and how they work for you in a track. You will become more acquainted with constructing sounds and know from the start what frequency characteristics to aim for, resulting in sounds which sound more related to each other, which contrast each other better and which, overall, combine better (like a jigsaw puzzle) to create a well-rounded production.

[lockjawdnb]

Drum Processing - Drum Sends

So you’ve spent hours and hours carefully processing and developing some drum hits. You’ve processed all your individual hits and have a nice beat rolling but it doesn’t sound quite there yet; what to do...?

Here are some drum group send techniques I’ve developed over the last few years that can squeeze a bit more vibe from your drums. First, I’ll assume you have your drums set up similar to this picture below. I have my kicks, snares and hats all bussed to their own individual groups like so. From here, these submixes go to a whole drum group channel...

...as seen in this picture, the group track entitled “drums”. This is where all the drums meet. On here, you can do more processing/compressing/distorting and it’s from this bus that we will be making some sends.

The first idea we will play with is to make a drum group-specific reverb, as seen here. The idea is to make a send with an EQ first in the chain and then the reverb. It’s not visible here, but I’ve sent about -20db or so of volume from the drum group to this reverb send. From here, it’s a good idea to pre-EQ out lots of the low end of all the drums on the send channel and maybe some of your snare fundamental as you don’t want duplicates of these floating around the mix. The idea of this send is to give the drums some space and ambience so they don’t sound like samples sitting in an empty project. As this is drum and bass, quite short reverb times are good, anywhere from 10ms to 50ms usually works, depending on the tune. Fiddle with the character controls and EQ settings of your reverb as well as the amount of signal being sent to the send to make the reverb fit in with the drum sound and the rest of the track. Listen to how it sounds with the rest of the track playing as well as solo.

Another send I often like to make for my drums is a stereo widening FX send, as seen below. The set up here is identical to the previous send except here, I’ve sent full volume from the drum group to this send and used this Sound Delay plugin on the send (after similar pre-EQing as on the previous reverb send to get rid of low frequencies and snare fundamentals) to slightly delay the left and right channels, creating a wide stereo duplication of the drum group signal. I send the entire drum group signal here and then use the send channel volume to mix it in with the rest of the drums. Be careful here with having too much high end or creating anything that will screw up your dry drum transients (EQing out lots of things is good for this). This will add some subtle width and size to your now reverb’d drums.

Finally, a more specific drum send, but this time only for the snare. Below is a picture of another send that I’ve sent only my snare submix group to, not the full drum group.

This channel is set up with a pre-EQ to only let in high frequencies, a reverb and another Sound Delay plugin. The idea here is to create a wide and sharp high end attack for the snare only. A very short reverb time and a small pre delay mimics a nice white noise tail for the snare. Listen carefully and tweak carefully to assure your artificial snare tail mixes in properly with the snare and the rest of the drums.

These are just a few ideas. Experiment and see what sounds good. I believe these techniques and others are what can separate average drums from really nice sounding drums and can add that extra few % of vibe to your track.

[lockjawdnb]

Drum and Bass Zen

I've been producing electronic music almost daily now for about four years and each tune I complete, I learn something new. I'm still using almost the exact same tools and plugins as I was about four years ago but the way I produce changes subtly with each project, as does my perception of the genres I'm involved in.

Recently (the last 6ish months) I've been thinking more and more about the difference between a "good" tune and a "bad" tune, specifically in drum and bass but also in the range of music I listen to regularly. Over the period of couple of years of listening to drum and bass, patterns seem to emerge and it's becoming clearer to me now why certain tunes "work", others don't and some just obliterate a dancefloor, which is what I want to talk about here.

For the sake of this article, I will use the term "vibe" to describe what, generally, makes a track "good", first in the context of drum and bass. To me, a great vibe is created in a track when all of it's separate elements combine to create something which is more than the sum of their parts. Vibe is the feeling you get when a track sounds like its challenging you, creating a visual for you to examine and presenting an image of what it sounds like, presenting itself in an almost intimidating way. When a track has a great vibe, it really comes to life. Here are some examples in DNB:

https://www.youtube.com/watch?v=Z6KJwxuY_U0

https://www.youtube.com/watch?v=zyMUbiR-Pro

I think Micro Organism is self explanatory; the entire track is shaped around a single idea; whatever that idea is differs from person to person but every part of the track is dedicated toward creating that image and developing it.

Lifespan is similar but I think focuses more on creating a certain mood, or emotional response. The track is simple but uses some ideas such as repetition and carefully structured atmosphere to convey some kind of sense of being.

I think this idea is key all music for that matter but let's look at DNB more closely.

The problem I face every time I produce a track is the task of glueing the different elements of the track (the drums, the basses, the other sounds) together in such a way, whilst creating them, to create something worth listening to that moves you. The idea that every aspect of one's track must be flawless (to the extent that one can make it) is one I've had from the very beginning but the idea that not only must everything be perfect, all elements of your track need to serve a clear purpose and ideally, combine to create something really cool, something with a definite vibe. This is the hardest part of producing now. I know that if I simply spend a certain amount of hours producing a sound, it will be satisfactory eventually. Making the vibe is much harder.

So, how do we approach this? There are a range of methods, I guess. Some practical ones include:

Use fewer elements, but make them good. This means making your snare the best damn snare you've ever heard. Cut out extraneous basses and synths if they're just there because you spent 17 hours on them. Make the ones you have really good.

Make intros lead up properly to drops. This can be tough and is something I struggle with but remember that the purpose of an intro is to ready a listener for the "drop", whatever that may be.

Figure out what you're trying to convey. Have you stumbled upon a whack rhythm in a bass that you were trying to make a pad out of? Run with it and build a track around it. Cool tracks almost always have a central "idea", which usually takes form in some kind of rhythmic shape, which the rest of the track is built around. Also, dancefloor vs. non-dancefloor psychology in tracks is something you have to pay attention to. Just listen to Noisia's recent "Purpose" EP; the tunes are all great but they're obviously all intended for dancefloor consumption and thus, get boring after a while.

Develop ideas. This builds on the first point. If you have a really good sound, develop it over time, add to it, but don't over do it. Make every section of your song build upon the last.

These are just a few ideas. These ideas are obviously part of what constitutes being a musician so one could spend their life studying these things. The problem with producing is making the sounds at the same time as figuring out how to try to use them effectively.

There have been countless times where I have been in the very early stages of a project and made that breakthrough moment where my crappy, robotic loop has suddenly seemed to take on a life of itself and squandered it by over thinking and adding too many things into the mix. The temptation to over-produce is real, but often times, a track will be more effective with less. Ideally, in a dream world, production value is as high as you can make it without degrading the vibe or cohesion of the track.

I'm sure there's more I could say on this topic but I'm not so sure it would make any sort of practical sense. Essentially, the point I want to try and get across is that, on a dancefloor, nobody cares or can even necessarily hear your painstaking detail in percussion. If you're making a dancefloor tune, make a tune for the dancefloor and figure out what that means. If you're making something deeper, make something deeper and make it as beautiful and detailed as you can whilst still getting the idea of the track across.

[lockjawdnb]

Distortion and Saturation

I've decided to write some stuff on this topic because I spend half my life distorting, saturating and mangling sounds and it's a pretty integral thing in the world of making big, fat, and (sometimes) disgusting sounds. 

There are a multitude of ways you can distort saturate things. Doing this to your sounds increases perceived loudness, compresses and adds new material (harmonics, frequencies) to current sounds. I think this is really important as DAWs are, inherently "cold" and "empty" places to make music in; if you don't make something, it's not there and everything you want in your music has to be added there by you. The nature of soft instruments too means that there it takes a lot of processing to make a sound interesting and pleasing to the ear because harsh, simple and one-dimensional sounds (like the ones that come straight out of sample packs and synths) don't usually sound that great. Saturating things can bring them to life and add fatness, richness, size and sparkle to elements which otherwise sounded dead and lifeless.

There are a few plugins I use all the time in every project so I'll give a rundown of them and how I use them.

PSP N2O

This is a ridiculously good plugin, by far my most used. It does a bunch of stuff but I'll just focus on the distortion part of it. One of the modules is a distortion section with a bunch of different modes; 3 "symmetrical" distortions, 3 "asymmetrical" ones and a clipping mode. The symmetrical modes provide a softer, gentle distortion, the asymmetrical ones give weird harmonics and overtones to things and clipping is just brutal.

I like to use the clip function on tonnes of stuff. It's especially great for fattening drum hits (kicks, snares), adding grit to synths as well as just making stuff louder/compressing as it drastically boots level. It only takes a few dB of clipping till something starts to break up nastily though.

The asymmetrical mode gives weird harmonics and has a unique sound. A cool side effect of this mode is the way it can break up treble sounds when being affected by bass. What I like to do is bus an 808 kick with a hi hat, put N2O on the bus with the asymmetrical setting, and then EQ the 808 kick out of the group on the bus channel (the kick should be basically just a low sine wave) after the distortion. Because the bassy sound triggers the distortion, the hi hat gets broken up really nicely. The EQ after the distortion means you can just have the distorted hi hat by itself.

PSP VintageWarmer 2

Another sick, sick PSP plugin. I've had this one for years but only kind of recently figured out how to use it properly because it's designed in a really strange way and the manual is astonishingly scant on details. It's essentially a tape-modelled limiter/compressor.

Cranking the drive knob makes it work kind of just like a limiter but the interesting parts are the speed and knee knobs. The knee doesn't really refer to a threshold cutoff angle like it would in a compressor but more like a threshold control with built-in make up gain (this isn't even vaguely mentioned in the manual...). The speed knob refers to tape speed, 0 being the "slowest" and 100 the "fastest". I'm not sure of the specifics but the "slower" the setting, the more coloured and "tape"-ey the sound gets; harmonics start popping up, especially in bassier sounds and stuff gets fatter.

Generally, the distortion you get from this is pretty subtle but if you rape the drive/knee controls it gets pretty gnarly. It's important to use the mix control as it just gets too extreem at 100%. I like using about 10-30% mix and then cranking the knee and speed controls to taste to add grit to synths and basses. Just cranking the drive control and tweaking the in-built EQ can also level out sub lines.

Ohmicide

The only real "distortion-only" plugin I use, and probably the most erratic and unpredictable too. It's a 4-band type deal that lets you set bands of frequencies and distort each one individually with a tonne of different distortion algorithms. It's not the most well-designed plugin but you can get a TONNE of sound shaping out of it.

You can completely change a synth sound with this thing or you can just beef it up a bit. As it's so extreme, it's easy to completely wreck and mangle sounds but with some tweaking you can bring stuff to life. Generally, I just split stuff into subs, mids and highs and distort like that, on a whole sound. I don't usually get a shitty sound then try to use this to make it sound great, I usually bring this in to kind of "finish off" and fatten a sound which is kind of lacking a bit.

There so many obscure features in this plug in (mid/side encoding, envelopes for each band, the "morph" (???) feature) that I tend to just use it in a basic way. Aside from fattening stuff up and adding grit to things, I think the most logical way to use it is just to abuse things with it because you can get some seriously whack sounds out of it. Works great on turning drum loops into demonic cries and transforming vocals into distorted upright bass sounds.

 As well as plugins, in-built FX bits and pieces in synths are integral to making a sound fatter and more interesting.

The sine shaper, hardclipper, T-tube and phase modulations in this patch are all contributing subtly to creating this particular sound. Getting to know what each section does and how it sounds is important. It's also great to put modulations on dry/wets and amounts of distortion in all the individual sections to add more subtle inflections and effects. Generally speaking, it's always better to get the sound good at each step of the process so it's good to make stuff sound great inside the synth and then use plugins like one's I've mentioned to further enhance the sound.

As great as distortion and saturation is for creating a warmer, deeper and livelier sound for your entire mix, too much will wreck stuff. It's easy to hear when good distortion turns ugly but pay attention to how squashed things are getting; some sounds need to be squashed and mangled whereas other sounds that you were gonna combine and rip to shreds may sound better just slightly fattened and left open to sit above or behind other sounds. That being said, if you want your tracks to sound huge, loud, aggressive and "pushed", tactical distortion is the way to do it. Couple it with careful compression and you can get tracks sounding as loud as Noisia's or Skrillex's.

[lockjawdnb]

Workflow

So I'm gonna write some stuff about the lifespan of tracks I work on.

I generally only work on one solo tune at a time. I've got a bunch of internet collabs going on which, due to internet collabing restraints, take ages so I work on multiple tunes at once in that respect but with my own solo tunes I can only work on one at a time as I feel that a) if I don't put all my energy into one at a time it's being spread too thin and b) when I start one I can't bear to let it sit unfinished. So this is how it usually goes...

I have a basic template set up in Cubase which looks like this. I'm not a fan of using stock sounds for every tune and I make all my sounds from scratch for each tune but for organisation sake, I make a little set up like this with folders and some basic plugins which are gonna be in every project, regardless.

I always start a project with making some drums, usually with a combination of synthesis and samples.

This pic shows a group for kick and snare; there's an FM8 and a Superior Drummer combining to make a snare and a Massive being used to make a basic kick. I always try to make at least the kick and snare as good as possible early on and then bounce them to audio as running stuff in synths/VST's like this introduces a little latency and isn't nearly as accurate or neat as having a single, consolidated snare/kick sample. If I have multiple layers making up a single hit, I'll spend time lining them all up properly and making the perfect hit then bouncing it to audio and deleting all the plugins.  Doing this is good because it forces you to make some decisions; I'll do a save-as after every major step in a tune so I don't lose anything but it just feels way better having just two tracks with a kick and snare single audio file which will always sound the same.

Once I've got a basic groove rolling I'll bust out the synths and attempt to start making some workable sounds. I pretty much always have at least some kind of vague idea of what kind of vibe or feel of tune I'm aiming for and the construction of the drums helps with this (acoustic or synthetic sounding drums etc).

This picture shows some synths I'm working on in the early stages of a tune. 

In this case, after I had a drum loop going, I started trying to make some stuff in Massive. All these instances are derivatives of a single patch I made. After making a good patch, I save it and continue to mess with it; each time I find another cool aspect or add another sick modulation to the patch I'll do a save-as of the patch and duplicate the track so I have a separate instance of the synth for every slightly different patch.

This is important as moving a knob or two only a tiny bit can literally ruin a sound and make it lost forever (not exaggerating). It's really important to just save stuff as soon as you get a cool sound.

This stage is really challenging as you're trying to a) make good sounds and then b) create a groove or riff out of it. In this case, I created an initial patch with a very basic riff/pattern and knew that I wanted the core sound/rhythm to stay the same but I wanted to evolve the sound with subtle modulations and differences, hence all the duplications.

After toiling on the core bass sounds of a tune, I'll start adding in more drums and other sounds. Assuming the core elements of the tune are good and under control, the rest of the tune will usually fall into place much easier. Making the main drum "kit" and bass synths is usually the majority of intensive work for me anyway. I tend to see everything else as sort of peripheral (but not less important). It's hard to keep track of what your tune actually sounds like during the sound design process so try to hear everything you add in context of the whole tune.

After all sound design, synth construction and arrangement is complete in a track, I'll start bouncing out stems of things currently running in synths.

This pic shows the all the bass stems from a finished project.

This is done as rendering something to audio makes it about as concrete and final as you can get. Obviously, save-as's ensure nothing is lost so I can go back to previous projects to change things if I need to but there comes a point where everything can be consolidated down. From this point, the only further processing I do is bussing all relevant stems together, and slightly processing them all in bigger groups.

Plugin delays are compensated for in Cubase (like all DAW's) but I'm not 100% on exactly how it works but I do know for certain that rendering tracks whilst they're running in real-time from a VST introduces about 5-10ms of latency. Rendering stuff to audio like this not only solidifies everything but also lets you get all sounds exactly where you want them (more important for synths that are making transient/drum sounds because you don't want snares being a few miliseconds late). Doing this also makes stuff much easier to mix as when it comes to mixing time you essentially just have 80 audio tracks or so and all you have to worry about is a volume fader and not crap like this...

From then on, it's basically mixdown time (assuming all your arrangement and sound design is done).

[lockjawdnb]

Stereo Imaging

So this will be about everything to do with stereo image in my DNB experience.

Manipulating stereo image is a subtle thing but can subconsciously completely change the way you perceive a mix.

The basic concept, I think, of screwing with stereo image is deciding what the key elements of a tune are. Obviously, in DNB, these will be your kick, snare and bass. You want these things to be fair and square in your face, especially if you're talking about a tune which is gonna be played in a club settings.

I understand the stereo image in terms of panning and increasing the width/mid-side difference of a sound. Panning is self explanatory. I don't know the specific details of mid-side but I do know how to process stuff in mid-side mode. Basically, in the same way that you can split a sound into Left and Right channels, you can split it into mid and side mode, mid meaning everything that is mono (shared by both channels) and side (which I understand to be the "difference" between the two channels).

Listen to these two examples:

The bass in this tune is (I think) completely mono:

http://www.youtube.com/watch?v=-H2V8ZJtP58

Compare with this tune. Listen to how wide and "past the speakers" the main midrange bass sounds:

http://www.youtube.com/watch?v=4awlOQPgh3s

When you compare the two, they are radically different.

I guess the general "rule" for spreading things wide is that you can spread higher frequency sounds more. I completely mono everything (via the master) on my tracks in Ozone under 100hz. You can go to town with things above probably 750hz or so but I think it's always a good idea to keep the 30-500hz-ish range pretty mono'd.

This is how I've treated the bass from a tune I've been working on:

http://i411.photobucket.com/albums/pp198/Mossrat/bass1.jpg

As you can see, I've split a bass into tops, bottoms, "spread" and created a send for the whole lot too. I've completely mono'd the sub (which is cut off at 150hz) and almost completely mono'd the top layer too (which starts at 150hz). Together, these 2 channels just provide the dry, simple bass sound which I bounced out of my Virus with a whole bunch of processing and plugins but I made sure not to include much stereo-affecting stuff.

The "spread" channel has been EQ'd heavily in mid/side mode as you can see. I've cut out the majority of the mid (mono) part of the sound because I already have those frequencies in the other 2 channels. You can see then that I've boosted the side channels around the high mids. Just scooping out the mid part of the sound will make it sound fairly wide but I also applied some chorus for more stereo movement and sent it to a reverb (which you can't see).

These 3 channels are bussed to the group "main bass" where they are distorted and process more. This bus also has a send going to "m bass double". This is where it gets interesting because I went to town with crazy processing on this send track. There are a load of distortions, a reverb and chorus plugins on there, I basically just made it as wide and big as possible. Because it's a send, I can mix the level of this track in with the original split bass sound. All together, these 5 tracks completely fill out the stereo spectrum and I have complete control over them. I tried to keep the majority of the bass slap bang in the middle and just giving the midrange and fizz part of it some width.

Here's an example of what you can do with drums.

This is an image of my main hi hat channel plus it's overhead channel (from Superior Drummer):

http://i411.photobucket.com/albums/pp198/Mossrat/hats.jpg

The EQ here is working on a bounced out audio track of the overhead channel I've got from Superior Drummer but only of the hi hats. The main midi channel is spitting out a mono version of the hi hat and the bottom overhead channel has been EQ'd to accentuate the side's, giving a huge hi hat sound but still having the majority of it front and centre.

These are just two quick examples, mainly using EQ, of how to utilise some stereo trickery. It's easy to go overboard and make everything too wide; when this happens, you get a big sound but your tune will probably lack punch and weight in the middle, plus you run the risk of making stuff go out of phase and losing half your tune when it gets mono'd through a club system.

[lockjawdnb]

Reverb

Where do i even start on this topic?

If you're a producer of any skill level you know about reverb. Reverb is one of the few "make-sound-gooderizers" in the arsenal of production; everything sounds at least a LITTLE bit better smothered in reverb, right?

Reverb, IMO, is actually kind of underrated, especially in its ability to vastly change the overall sound and vibe of a mix. Careful use of reverb can put your mix in an ice-cave, a cathedral or somewhere in space.

When it comes to DNB, reverb can be especially tricky. Here's two examples where reverb really works to make a mix:

http://www.youtube.com/watch?v=NEkvTfMR0Sc

http://www.youtube.com/watch?v=jUj2J3FsaFw

Nickbee was the first DNB producer i heard who really uses reverb properly IMO. Note the very subtley reverbed snare in alien and basically everything in mangup that has a touch of a small-sounding reverb which gives the tune a very subtle but noticeable vibe.

Contrast these tunes with a tune like this:

http://www.youtube.com/watch?v=r6mHQtr3kW4

Nte the complete lack of reverb in the drop. Granted, the tune is far more "banger" than any nickbee tune but still, the tune would sound entirely different drenched in reverb. as such, it sounds full on and right up in your face (this isn't just because of the reverb but i'm trying to prove a point).

If you're trying to create a fast, rolling and in your face piece of DNB, reverb won't really be your friend. things like snares, hi hats, kicks and main midrangey bass noises will probably benefit from not having any reverb at all. hearing even the tiniest reverb tails on elements of a tune suddenly gives them "space" to your ear and detracts from the full impact of a sound. of course, this depends on what kind of sound you're going for ie. moody and atmospheric like a nickbee tune or sizzling and fast like a Phace tune or something.

The tough part is that using reverb to "cover up" in adequacies in a sound or using reverb as a quick fix to make a sound better usually isn't the best route; it's far too easy to whack on a reverb onto an average midrange bass sound to give it a bigger and reverby sound where what you really should be doing is making the sound better with no reverb (which will probably be better for your mix).

Using reverb in DNB (in my eyes):

So all this being said, obviously, reverb is still a great tool for use in DNB, if used with skill.

Firstly, the main thing to consider would be weather to use reverb as a send or as a direct effect. for those who don't know, a send is an auxiliary channel in which plugins may be inserted. Audio from another track (like a synth or whatever) can be "sent" to this send track and then that signal is mixed back into the original track. At what point the main track sends off audio to the send can be either pre- or post-fader. you can look it up if you don't know, all DAW's have some kind of sends compatibility.

The advantage of using a send over just plugging a reverb onto a track is that you get the reverbed signal but it is mixed along with the completely dry signal which is what i always prefer as i don't like having to compromise the dry signal of a sound by smothering it in reverb. The other advantages of sends are a) being able to send as many tracks as you like to a single reverbed send b) you get a more coherent and similar reverb sound in total if you only have 1 or 2 reverb sends as opposed to a reverb dropped on every single channel and c) you save a bit of CPU power (if that's a problem for you).

I almost exclusively use reverb on sends these days but using reverb as an effect itself can be sick. Fr example, the one of the built in reverbs in Cubase, Roomworks, has a very distinctive sound; most of the time, it sounds like whatever you apply it to sounds like it's in a metallic cave or something, which is cool, as it's tweakable but has a very distinct character. with short tail times and a "small" space setting and high wet/dry ratio, applying this type of reverb to a bass can give it a ridiculously large and spaced out sound (check my tunes "axis" and "internal" for over the top reverb not only on bass sounds but everywhere...).

Another note, experiment with the stereo width of reverb too; by default most reverb stereo controls will be set to "100" or max or whatever which means they are as spread across the stereo image as they can be. having some reverb signals more centered or completely mono with another, maybe longer reverb spread further will give a tune more depth.

On the other side of the spectrum, reverb can be key for genres with much more space than DNB ie. house, dubstep etc. check any of synkros tunes for reverb overload.

http://www.youtube.com/watch?v=M2Ni5vfvsx8

To summarise, I think it's very tricky deciding which things in your tune can benefit from reverb and how much reverb those elements really need and also what character of reverb they need. During the production of one of my own tunes, I apply reverb, being wary not to apply too much, then constantly keep turning it the reverb down as the tune progresses. when it comes to mixing stage and I start paying close attention to the mix, I usually realise I actually still have far too much and keep reducing. it's pretty common to be using far too much reverb in general as even the smallest amounts will often achieve what you are using reverb for in the first place. what I'm striving to do now with my tunes is use as little reverb as possible in order to achieve a more direct and upfront sound, not only in DNB but in slower genres too.

If in doubt, check tunes that are you similar to what you are producing and be as analytical as possible. What I try to visualise is if any reverb in my track is interfering with the dry signal of a specific sound or if it's detracting from the sound as i see the reverb of a sound as very separate from a dry sound. It's all about figuring out how revering a specific sound will affect a listener's perception and reverb can very, very easily radically change a mix. if you are using multiple reverbs in a track, keeping coherence between them is key.

In my opinion, less is definitely more with reverb.

[lockjawdnb]

Synths

so i'm sitting here in bed dosed up to the eyeballs on painkillers as i've just had all four of my wisdom teeth out. what better chance to write all about...synthesis!

this is gonna be a tricky post, there's a lot to say on the topic but i'm going to limit it to drum and bass synthesis for now.

i guess i'll start with what synthesis is and why it's so important in electronic music, especially instrumental music which dnb is a lot of the time.

if you listen to any "popular" radio stations or pay attention to any music which is present in the mainstream culture you'll realise soon that 99.9% of it contains vocals; humans are naturally drawn more to and can identify more easily with music that has vocals as the main element of the song, which is obvious.

as soon as vocals are removed from a piece of music, it loses a certain amount of interest. this is why instrumental music, in general, just doesn't have as wide an audience as "popular" music.

so the challenge faced by all producers of instrumental electronic music is how to fill that main "voice" which doesn't exist any more. this is a pretty sweeping generalisation but you get the idea.

this is where synths come in. i'm gonna assume anyone reading this knows what synths are and how they work. synths are, for me anyway, and the genre of music i'm involved in, the raw materials for all the sounds i produce.

so you've started a new tune; you've got some drums rolling and maybe some atmosphere or something. how do you turn that loop into something interesting and worth listening to? you need that main voice.

every time i start a tune and get to this point i go through a period of a few days of utter hopelessness and loss of self worth as i try my hardest to construct the main "voice" of my tune. most of the time, i'm trying to make a bass sound; this is, after all, drum and bass.

i'll give you an example of this idea.

http://www.youtube.com/watch?v=1JlEufxaJlw

the majority of this tune is simply that big, wobby bassline which changes constantly to keep the tune interesting; your ear is obviously drawn mainly towards that sound. that's an extreme example though, a more typical example might be a tune like this:

http://www.youtube.com/watch?v=tFo3E03_bcQ

the focus of attention is most definitely the huge bass sounds in the drop.

simply having a good main sound to focus on in your tune is a good start but it's not enough; finding other, smaller, more peripheral sounds (which are as high of a quality as your main sound) and creating a drum sound and pattern which emphasises and enhances your main focus are  key. here's a good example; you can easily identify the 2 or 3 main, big bass sounds in this but pay attention to the higher pitched and more subtle sounds going on in the background:

http://www.youtube.com/watch?v=ZnhM8NL0D8I

trying to explain how to create an interesting voice for your tunes is kind of a futile exercise. for me, it always involves hours and hours of experimentation and failure as i try to make something worth listening to. usually, there's a breakthrough moment when i'm working on a bass sound and then quickly start getting ideas down and making other synths to create the core groove of a tune.

the only thing i can really say about bass sounds though is that they need modulation and movement; this includes crazy filtering, automated effects like phasers and whatnot as well as simple things like volume changes and how the sound fits in with the kick and snare (this is actually really important). the hardest part though is getting the actual timbre of the sound to be good which is tough because synths by nature sound digital and nasty as hell.

as i said before, whilst doing all this you need to keep in mind all the other sounds which will enhance your main focus and fill the "sides" of the tune out and keep it interesting as well as your drums.

the more you do this and the better you get a synthesising sounds you want, easily, the better your tunes will get. the most important thing of all of this though, is ensuring every individual element in your tune, weather it be synths or drums or whatever is as high quality as you can possibly make it. it's totally reasonable to spend weeks or (god forbid) months on the core elements of your tune to get it sounding as good as it can. it's the equivalent of a singer rehearsing a part over and over until they can sing it properly, in tune with the correct timbre.

basically, the main idea of this post is to make sure that there's never any filler in your tunes and that every 4-8 bar section is interesting and captivates your listener. it's an incredibly hard task, co-ordinating tens and tens of different tracks to function in harmony and it is, essentially what production actually is; producing sounds that are worth listening to multiple times, for the duration of a tune.

if anyone wants to contact me and discuss anything i've talked about further, feel free to.

[lockjawdnb]

Mixdowns

mixdowns.

a mixdown is the second to last process in the production of a tune, for me anyway. a full mixdown will, if done right, fully bring out and compile all the good things about your tune and bring it a step closer to completion.

i really like doing mixdowns as i only allow myself to do them once the design and arrangement of a track is 99 or 100% complete and i always know my track will sound at least a little bit better after a complete mixdown.

i never used to do full mixdowns; i always used to just produce a track and get everything right as i was going. this was working out ok, but i decided to give my tune "elephant" a full mixdown after it was completed to see what all the fuss was about and decided it was definitely the right way to go.

the thing with doing a full mixdown is that it allows you to separate the sound-design aspect of the tune from the mix aspect of the tune which are kind of completely separate. it pays to stop worrying about the actual sound of every element in your tune and stop and think about what kind of mix your tune demands.

however, the longer i produce, it seems the better i get at getting my mixes right as i'm producing; my tracks usually have between 50-75 or so channels and if i take the limiter off my master, i only ever peak the master by a few dB or so.

i produce with limiting on my master. i know this is controversial to some people but i only limit by a few dB; not only does this stop stray peaks so i don't have to continually worry about distortion at the start of a track when i'm getting ideas down fast but compression/limiting also just makes things sound better. i guess i'm of the generation that was spawned fully into the loudness war; uncompressed DNB just doesn't sound "right" to me.

back on topic. i find that the majority of my tracks all individually peak at around -15 to -5db or so on the track meters (not on a spectrum). generally, there is no real reason for anything to be going over 0dB anywhere at any point in a plugin chain or track (it's also important to remember that sounds between about 2-5khz will sound louder than things higher or lower than that with the same dB readings as the human ear is attuned for this frequency range).

when the time comes to do a mixdown for a typical, bass-heavy DNB tune, i approach it like this:

1. render out the main bass elements of my tune to audio (once i'm happy with them...).

2. do a save-as of the project and rename it "*trackname* mixdown" (good habit to save-as regularly during production of a tune) in case you fuck up drastically (pretty unlikely) and need to revert to and old project.

3. import my rendered bass and whatnot into the new project and delete any now redundant bass synths (for cleanliness and less clutter).

4. zero the mixer. i open up the cubase mixer and bring as many faders down to -inf as possible (i mute tracks with volume automation). pro-tip: if you have group tracks, just zero the group track otherwise the submix of the other 1-10 tracks or whatever you have feeding into the bus will be ruined.

now, the fun part. i'll find a drop in the tune and start with the kick and snare (BTW, this is with some rough limiting/mastering processing going on to get an idea of the final product).

i think of DNB mixes like this; you have your kick + snare, subbass and your "mids" (usually bass mids but whatever else is the main "voice" in your tune). these are the three elements i focus on trying to be the main parts of the mix.

i imagine my kick and snare being the anchors of the tune, or being the "highest, loudest" point of the tune. it took me ages to realise the importance of the snare in the mix, most of the time, snares in tunes i like are mixed to be right up front, as a centrepiece of the mix.

i bring the kick and snare up. these two elements are the only points in my mixes that are ever "as loud as they can go"; i basically just try to get my kick and snare as loud, punchy and full as possible and base everything else around that.

i'll loop the first 8 bars or so of a drop. once my kick and snare are at good levels (use a spectrum on the master too, after limiting). now, i'll bring up the subbass.

i've written before about numbers regarding sub so i won't go into that. once i get that right (i might use some multiband compression on the 100hz and below region to control the sub) i'll move onto the mids.

generally, i end up with 3-10 or so main bass mid channels eg. the main basses of my tune without the sub on different channels so i can get them all sitting right. this part is tricky as it depends on your tune. from this point on, i try to keep everything just slightly "under" the kick and snare. it's an abstract thing to think about but in a lot of tunes where you might initially think the bass sounds are the loudest points in the mix, they're actually sitting just under the snare if you pay attention to it. don't lose track of the anchors of your tune!

i usually find i have a tendency to keep increasing the volume of my basses during the production of a tune; a lot of the time, they come down quite a few dB in the mixing stage. there's always a sweet spot where the reduction in dB doesn't make them lose much loudness per se but frees up a lot of headroom in the mix to let your drums come through properly.

once i get these three things at a decent level i'll go through the project 8 bars at a time, raising every track up to level as it appears in the tune (important: make sure you don't forget any tracks. keep checking the mixer to make sure you haven't left anything muted/zero'd before you finish).

i find everything else to be fairly peripheral. hi hats and cymbals probably come fourth in the mix-list but they can also usually afford to be fairly quiet. if your stereo placement and sound design/arrangement is good, the mixdown should be fairly straightforward from here but there will most likely be some things that will jump out as you pay close attention to them.

another important thing to consider during mixdowns is the dynamic range contrast of the mix ie. the difference in loudness between your intros/bridges etc and your drops. it depends on the tune really and how you want it to work out. i like having big differences in volume between intros and drops to give maximum impact to drops but it really depends on the construction of your tune.

another thing, you gotta mix/test mixes on different sound sources; headphones are good for getting really detailed with mixes but a lot of the time headphones will mask other things which become immediately apparent when played through monitors so check extensively on both and more sources.

after this, i'll start "mastering" the tune but that's a slightly different story.

once your initial mix is done, save and turn off. i don't recommend doing a full-track mixdown in different parts. how my ears hear the tune differ greatly depending on how long i've been listening etc. so i'll always do a mixdown in full then come back to it the next day or so and i'll usually find something that needs to be changed slightly. from there, test the tune on more speakers (eg. in a car, a friend's speakers, home hi-fi etc) and keep A/Bing during the entire mixdown process with pro tunes to see how your favourite producers have mixed certain elements of their tune.

final step: render mp3, upload and cry with joy. or cry with hate as you realise how underwhelmed you are with your tune :P

[lockjawdnb]

Generating and Controlling Subbass

Subbass, one of two key elements of DNB. Subbass is that earthshaking frequency range from about 100hz to 20hz. Many people will say you don't hear subbass, you feel it, which is partly true, but you do actually hear it and controlling it can be a tricky business.

First, let's check some examples of mussive subbass.

Check the drop on this bad boy, possibly my favourite DNB tune of all time. Massive bass:

http://www.youtube.com/watch?v=Wcoe6PxDI7U

And this tune. This one goes lowwww:

http://www.youtube.com/watch?v=6Y3GIxRlT6s

First up, I'll tell you straight up; if you're seriously listening or trying to appreciate DNB or other bass heavy  genres through iPod headphones or laptop speakers...you're doing it wrong. You need something capable of pushing a lot of air to get the full experience. Listening to DNB without full sub range is like trying to watch a movie without sound; it's retarded. "But Lockjaw," you say, "what about all the average masses who want to listen to your music on the shitty car radio?" Well, nobody gets into DNB for the money so it's not worth trying to compensate, cause if you don't got the sub, you're missing out.

So, assuming you're equipped with either some decent monitors or decent headphones, let's move on; how do you get subbass?

There are two ways in my book; either using a synth to generate a dedicated subass line, or splitting your main "bass" sound and treating the sub region of that. Neither is better, each serves it's own purpose; personally, I like the second option as it provides a much more natural, "organic" sound. Check this example;

http://i411.photobucket.com/albums/pp198/Mossrat/layers.png

This is what the basses from my tune "Deadweight" ended up looking like in Cubase (hear it here: http://soundcloud.com/lockjawdnb/deadweight). The tracks labelled "sub" and "spr sub" make up my 30-150ish-hz region. The tracks under the "bass" bus are everything else; each "bass #" track represents a different bass "riff" from about 150ish-hz upwards.

So basically, I record my fat bass sound, once it's more or less done, movement-wise, to audio. Then, I duplicate the audio, lowpass one and highpass the other, like so:

http://i411.photobucket.com/albums/pp198/Mossrat/EQ-1.png

The bottom EQ is my mids and treble track (forget the notches) and the top is the sub channel EQ. As you can see, when played together, the entire spectrum of the bass is essentially covered.

By splitting your basses up like this, you get a way more "natural" sounding overall "bass" sound; if I were to simply layer a fat sinewave coming straight out of a synth under my mids, it would sound off and not gelled together. I have a feeling this frequency splitting method is pretty popular. A good example I can find off the top of my head might be this track. Listen to how the subbass is clearly a part of the main mid part of the bass in the drop.

http://www.youtube.com/watch?v=DbiKsshD8Hs

Also check this tune. Note how the sub once again follows the dynamics and movement of the "wob wob" bass mids.

http://www.youtube.com/watch?v=EBuXxbo-hi8

As you can hear, the sub does not ALWAYS follow the main mids. Like I said before, I like using this method to generate sub for the entire track, not just acquiring the sub section of a bass to provide sub. For example, in my track "Internal" (http://soundcloud.com/lockjawdnb/internal-free-320kbs-download) the same bass patch that was used to generate the main, reverby stabs was also used to generate sub for the entire track. In this case, however, the mid-rangey stabs don't have a dedicated sub channel. Instead, I highpassed them to keep them out of sub rage and came up with a big riff using the same midrange bass patch and used it for the sub, lowpassing. What I find most important, given these methods is that you maintain consistency and flow.

So what do you do after you've cut your bass frequencies up nicely and have your sub content on a separate channel? Well, in my view, sub is one of the few things in DNB that can benefit from having a constant dynamic level; that is, it's always solid as hell. To do this, I limit the shit out of sub. This is the processing I applied to my main sub channel in "Deadweight", and it usually looks the same on most of my tunes:

http://i411.photobucket.com/albums/pp198/Mossrat/subFX.png

I've mono'd the bass (I mono everything below 100hz basically) as it had a lot of stereo movement down to the 30hz range, limited it pretty hard with VintageWarmer2 and applied the lowpass after. Some might argue that you should EQ first then limit but it doesn't really matter in this case; the EQ after limiting removes any weird distortion hard limiting might introduce.

I try not to limit it so that it starts breaking up nastily; the point of the limiting is to get it nice and constant and level the dynamics a bit (but not to overkill it) as well as bring up the volume. That's about all I do for my sub.

What I also like to do, which I find adds that little extra touch to your bass sounds, is create another duplicate of your original bass sound and bandpass it to only let in about 100-175ish-hz rang and then spread this really wide. This gives you those upper level harmonics which add crucial texture and movement to your sub region and the FX for it for "Deadweight" looks a little like this:

http://i411.photobucket.com/albums/pp198/Mossrat/sprsub.png

So, like I said before, controlling subass can be an insane job, and it's something I struggle with on every tune. I find the only way to get it "right" is by using a spectrum analyser. A while ago, when I started "mastering" my own tunes, I started throwing tracks in Ableton and looking at them through a spectrum analyser to see what levels their elements were hitting at (one of the most valuable things you can do). I found, after a while that ALL DNB tracks I analysed had their subbass regions peaking between the -9 and -4(ish)db regions on Ableton's spectrum; this is definitely the "magic number" for subbass, depending on the style of tune. Now, I have no idea how those spectrums work or what the numbers really mean, I just know how to read them and make use of them for my own work.

Now in cubase I use Voxengo SPAN which uses some other kind of measuring metric which I have no idea about. What I figured out soon though was that -28db on SPAN represents -6db on Ableton's spectrum, so that's what I always aim to get my sub peaking at AFTER limiting (this is important; pro tracks you analyse have obviously already been mastered).

http://i411.photobucket.com/albums/pp198/Mossrat/subspectrum.png

So you can see here, somewhere in the drop of "Deadweight" my sub area reaching about -28db on SPAN.

During the production of a track, I don't really worry about the sub level too much; I figure that out when I master (ie. whack a limiter on and pretend I'm Noisia). I just make sure my sub isn't so loud that it's obscuring anything (this happens when you have a limiter on your master and the power of the sub literally just smothers the other frequencies, causing them to duck in volume, since the limiter let's nothing over 0db...this puzzled me for ages...).

Lastly, if you're producing in a less than perfect environment, like me, sometimes the only way to really figure out if your sub is sounding right is to go by the numbers. If in doubt, I find a track which is similar to the one I'm working on and follow the sub levels closely on the analyser and try to match them. As a general rule though, I find that if my sub is peaking over -28db for any real reason, it's just too much and will be greatly exaggerated and noticeable on a system.

Don't underestimate the power of the sub. Just because you can't feel your ribcage rattling when you're listening to your tune at home doesn't mean the sub isn't there or powerful enough. If the sub is there, it's there and it's probably you're monitoring/room not doing you justice; when I played one of my ancient tunes "Faithless" in a club, I realised instantly there was an obscene excess of the sub as the whole club shuddered with sine waves. It's always better in my opinion to have a little less sub than too much. Check this track, for example:

http://www.youtube.com/watch?v=7vNP1wMx7rw

The drop sounds pretty heavy but you wouldn't guess there's actually very little sub content in it.

Lockjaw out lolololol.

[lockjawdnb]

Snare Philosophy

(Click image URLs for bigger images...)

So I thought I'd kick this little blog off with some words about drum and bass snares and some techniques I use to create them. There's way too much to talk about snares in just 1 article so I'll try and limit it and just give a basic introduction.

In DNB, your kick and snare form the anchors of your beat. Maybe you already know this, maybe you don't but the importance of having a strong kick and snare is paramount. How else can you stand hearing a snare almost 250 times in a tune unless you actually enjoy hearing it every time? Kicks too, but to a lesser extent. Snares dictate the overall rhythm of your track and can change the way your tune feels/sounds.

Snares come in many different shapes and sizes and what type of snare I use for a track depends on what vibe I'm going for with that track or sometimes, I'll base the vibe of a track off how my drums are sounding (more likely).

The anatomy of a (basic) snare consists of a big sine wave descending in pitch really quickly with white noise/treble on top. The sine wave gives the snare the punch and impact and everything else up in the higher regions give it snap and the timbre that makes you hear it as a sine. That's a real basic overview. Here's what my snare from my track "Internal" (you can find it on my SC) looks like through Izotope Alloy's EQ spectrum (BTW, if you don't have a decent spectrum analyzer, get one. Ableton's inbuilt one is really good, I use this one cause it's free and good: http://www.voxengo.com/product/span/ also, some EQ's have built in spectrums which are good too like Logic's one and FabFilter Pro-Q and Alloy's EQ which I use a lot).

http://i411.photobucket.com/albums/pp198/Mossrat/Tutes/snare.png

Check that big spike at around 200hz. That's what gives a snare impact. If you go and analyse some snares from your favourite tunes you might be surprised to see how many of them have this sort of frequency composition. Sure, you can have smaller, more high-endy snares but to get the right sort of power you want for any kind of DNB on the heavier side, you're looking for a snare with that kind of punch. If you pay attention to snares all day long, you'll come to recognise different sounding snares and even be able to tell their rough frequency composition by ear. Check out the snare in this tune:

http://www.youtube.com/watch?v=z87x9kKt5pY

That snare sound, to me, can only really be attained by pitching up what was once a big, fat 200hz-ey snare.

Now the hard part; making a good snare. If I had a sure-fire way of making excellent snares my life would be far easier.

To contrast, check out this snare from Hybris:

http://www.youtube.com/watch?v=keNMHLKC5A4

It's really tiny. By tiny I mean, it doesn't have a huge bassy attack and it's fairly buried in the mix (comparatively). 

Probably the most common method for creating snares (and other hits) is layering. This means you find a bunch of samples that each represent a part of your snare and combine them. For example, this is how I made the kick from my last tune "Deadweight":

http://i411.photobucket.com/albums/pp198/Mossrat/Tutes/kicklayer.png

You can see from the tracks that I have a bass kick channel, a high end one and a "spread" channel (more on that later). Basically, I found a kick with a nice, bassy part of it ("punch" in kicks comes from the 100hz-ish region) and then another one with a nice, cool sounding top part. I lowpassed the bassy one and highpassed the higher one to cut out the frequencies in each one which is provided by the other and when they are played together, they make a nice kick.

Although this is a kick, the principle is exactly the same for snares; find a snare with a nice, low end punch and then find a sample that has a nice "snap" to it to layer on top. More processing is required to get the two to gel together but that's another story. That's the basic premise though.

Another way to make your own snares is by synthesising them. I started experimenting with this a while ago using MASSIVE and had some sick results (the snares from my tunes "Axis", "The Native Bounce" and "Internal" were all generated partly from MASSIVE). I won't go into too much detail, but this is what the patch looks like:

http://i411.photobucket.com/albums/pp198/Mossrat/Tutes/massivesnare.png

Basically, it's just one oscillator generating a sine at anywhere between 200-350ish hertz with an envelope making it drop down like 36 semitones really quickly and then the noise generator adds the high-end. The key is tweaking the envelopes really precisely, otherwise it doesn't sound that great. Doing just this will get you a reasonable 808 snare but for huge DNB snares, a lot of processing and layering is required to get the full sound but more on that later.

So those are the two main methods that I've used for making my own snares. As long as you can figure out the rough frequencies of your favourite snares from tunes you like, you can go about making them yourself. And if any of you crazy people want to actually talk to me about anything to do with this, send me an email or SC inbox of FB message or whatever.

[lockjawdnb]

Intro

So, for anyone who finds this blog by their own volition and has no idea who I am, my name is Louis and I primarily produce drum n bass of the heavy variety. You can find my sounds here:

http://soundcloud.com/lockjawdnb

I do not profess to be any sort of production-mastermind, I'm just here to vent production ideas and knowledge as I gain it myself as a way to help other people who might be interested and to talk shit about drum and bass, the genre which I am so interested in.

Like I said, these tips and anything else I post should be taken with a pinch of scepticism as these are only the methods which work for me and which I have learned, mostly, by self experimentation.