minimal dub/techno

so i've been producing dnb all this year and feel like a change

i've always liked guys like monolake, pole, loscil, etc. but never made anything like that

now in dnb, you might spend 5 days solid cutting up and processing breaks - only to resample them, work them into an arrangement, cut them up again, re-EQ everything, layer them with more single hits and other breaks, and spend weeks on the arrangement and mixdown

it's anal and time consuming - which is why i always get bored of what i'm working on

now at the other extreme, i try a new experiment - load up an oscillator->carrier and make a simple bassdrum - throw in a synthesized mallet-type sound, a tom and an fm chord, and in about 2 minutes i seem to be at the finishing stages

i know this is maybe the hardest music of all to do well, and i've got a long time to go before i'm anywhere near accomplished at it - but can you justify calling a tune finished when it basically takes longer to listen to than it did to make? and when you know you've not done anything remotely original or different, or anything that anyone else couldn't do easily?

it's like finding a great synth preset, playing a chord and calling it an ambient tune... there just seems to be something so wrong about it - but rationally, if it works it works right?

post something. If it's any good, then you're a genius. If it's not, then you're doing it too damn fast. As for me, I suck at electronic music; so I'm regressing back to ambient techno. It's fun to make.

well this style you are talking about is indeed very difficult to make. the music itself , the notes rhythms, etc, is not that hard. actually id say the musical part of the production (except rhythm) is the easiest. as you said, a few chords...

what you should probably work alot more on is:

1. choice of sounds and sound design

if you listen to monolake, youl hear how the drums he made are very cohesive and they really form a kit. its not just taking and 808 set or whatever. you should synthesize every sound or record it yourself with a microphone. then eq, compress, and make the sounds go really well together.

2. structure, progressions

the art of subtility. adding elements and sustaining the interest just enough to not bore the listener. thats what minimal means to me. making the trakc progress and have different sections wit hdifferent elements in them, making the elements disappear... it really should take as long as a dnb track to make. in a dnb tune you usually use a few breaks and they doint usually suddely appear and disappear. well it does depend the type of dnb youre making, but im talking in general. for minimal youre going to need to put alot more time in how the track flows and it can be hard if youre used to a very energetic and propulsed sound like dnb.

3. effects/mixing

with minimal, you usually want to fill the whole spectrum but with just a few elements at a time. the easy solution is: lots of reverb. but if you want to be a bit more elaborate, you have to use alot of effects, some just in the back ground, some in the foreground, some really buried under the mix. delay is useful here. and finaly the progressive effects like flange, phase, etc... for these its useful to have a hardware knob to turn instead of drawing automation because you want to put alot of feel into these sweeps. oh yeah a compressor is nice to have too, you can make it pump, side chain.... useful stuff to keep the energy at a good level.

dont tell anyone, but if you can get your hand on an analog filter like a sherman bank, alot of these minimal guys put every single track through it. its all about the enveloppes and hp/lp/bp sweeps.... FFFFFSHHHIIIOOOUUUUU.

i remember you said you work in the digital realm only. you can really benefit from a mixing desk with send/returns, faders to make subtle and slow changes, a real phaser pedal. because its all about the flow and intuition, little changes.... analog hardware really helps for that. in digital, you have to program every parameter, its a really different way of working. except if, like monolake, you make your own environment in max msp. thats just the best thing to do, could even be better than having a whole rack of hardware stuff and mixer. you just need a controller with alot of knobs, map the knobs to the parameters... but thats a long long long road my friend :)

thanks, cool advice!

i'm actually 50:50 cos i've got 2 studios - 1 based around a laptop with csound, max, renoise, and 1 based around a mackie 8buss & toft, with all eventide, avalon, CS70, juno's, SX3, etc.

but yeah, it's really just the rawness of the production and arrangement which i'm finding it hard to get to grips with

i mean, do you ever play the first thing that comes into your head with a synth patch and call that an ambient tune? i'd have difficulties admitting to myself that that's a "real" track i think - but then you know there's no reason why it shouldn't be

yeah that's exactly it

i was realising this the other week... this dnb tune i've been working on solidly for about 2 months - when you get some distance and actually hear it, you're only ever focusing on 1 or 2 musical elements at any one time

the big mistake is to try and cram more in there - i have to consciously go back and strip things out - otherwise it sounds directionless and muddled

even with a full orchestra - you're actually only focusing on one or two elements at a time - it's not like every instrument is playing its own counterpoint... and if it were, we'd not make any sense of it

yet, with dnb, you spend 3 hours getting the phase and tuning between two snares perfect - you might re-process the same sound 15 times or more in the end, what with all the resampling and re-cutting

but you're still really left with something every bit as simple musically as an analog drum machine and a stylophone

the big drawback with dnb is that so many musical and rhythmic ideas won't work because of all the style constraints... but with a drum machine and a stylophone, you can actually use any melody or rhythm you want without getting self conscious

That makes me think. I have noticed that as well. I will try to cram in as many things as well but that never works, I usually have to manually subdue and bring out parts. I think that is the key, having a lot to listen to but not all at once. Little things in the background that go unnoticed are good especially if you can mix around them. I think that there is an art to doing this in a musical way. When i think about what is brilliant about a lot of older phex track (new ones too) i usually come to a similar conclusion.

yeah, i always thought the old prodigy experience stuff had a lot going on if you listen closely - i can't think of many other ppl who manage to cram so much in

but yeah i was thinking of old aphex tunes like analogue bubblebath and didgeridoo - i mean that one's just a drum machine loop and a single note playing a sequence... probably just looped and recorded live down to DAT controlling the desk

surely the sounds as original source material must exist before arrangement, I think once they are, the sounds guide you to how 'they' would like to be arranged.

So, we're dissing dub now?

  bowen said:

surely the sounds as original source material must exist before arrangement, I think once they are, the sounds guide you to how 'they' would like to be arranged.

sort of - but then if you're working with something like max (which almost all of this crowd seem to), a bass drum's usually only a sine wave and a pitch envelope - a hihats just some filtered noise - most the chords and instruments are just 2 operator FM sounds - so i don't know... i think it's more like the sounds shape themselves to the arrangement just as much

i've never worked like this before - but the real advantage seems to be that you can program your sounds so they're never the same twice... can be really subtle, but that's what allows you to repeat the same section over and over without it ever sounding monotonous

Ok i see what you're getting at sequencing in max is a little tricky to do, I've never particularly happy with 'timeline' 'seq~' and the like, i would really prefer to just midi the patch up and either record or rewire it.

I havn't looked too much into it, but have you seen the techno~ object it looks really nifty for sequency synth changes within max

i've still not got too into max myself - it amazes me that people have the time to do so much with it - i've collected a few patches and generative sequencing things, but really wouldn't consider myself accomplished or 100% comfortable with it yet... i'm kind of hesitant to go down that road of spending any more time thinking about interfaces and computers, than music - i can see that happening far too easily with me

but then again, i do like the flexibility of being able to manipulate all your sounds on the fly, and have tonnes of different length loops all triggering different parameters, so you never really hear the same thing twice

you can get that to an extent with other sequencers - i was actually looking for a really simple, modular synth engine i could use in renoise to do all of this

i've got reaktor, but it sounds too coloured to me, and uses too much CPU

i just want clean sine waves, noise, modulators, carriers, lfo's, filters, etc.

did you check out tassman? i haven't used it much, but it seems alright.

tassman looks almost ideal for me - but it's physical modelling based and the sound engine's tweaked like reaktor

if you could get exactly that, but with a really pure sound engine like Max/msp, csound or supercollider - i kind of prefer a computer to be a computer if you know what i mean

  Quote

i just want clean sine waves, noise, modulators, carriers, lfo's, filters, etc.

i've been experiencing some aliasing issues this past few months with some stuff, noticibly with anykind of wavetable object, sticking to the inbuit waveforms is fine, the sine is nice and clean. so the cycle~ saw~ etc..

also on the subject of filters have a look at 2up_svf~

http://www.2uptech.com/archive.html

its a bit nicer than the built in filters in my experience its driven a signal level so frequency changes are smoother. its osx only though

the moon: i exactly get what youre saying about reaktor and tassman coloring the sound. theres still something very toyish and emulated about them, and im really not interested in that. i mean, if you use digital synthesis, why try to make it sound like analog. id rather have a really sharp and clean, simple, digital sound and thats what max does really well. its a serious program. very good for abstraction and pure forms.

oh and very nice analog studio youve got there... a cs70!!!

so the discussion turned to psychoacoustics. thats very interesting. psychoacoustics is the study of the way we hear sound and the way we segregate elements. the basic koan you have to answer first is: how many sounds can you hear at anyone time? the answer is 2, because you have 2 ears. we only segregate (differentiate instruments, sound sources) psychoacoustically. and we do concentrate on one main element at a time, but it is possible to train yourself to follow more. the classical ideal of course is: one thing at a time. simplicity. repetition. thats why aphex twin was called the mozart of techno at one point.

but then when you get into schoenberg, berg ,webern, these guys had a very different aesthetic :)

remember: you are the listener as well as the composer. compose music the way you listen to music.

how about synthedit? i don't know anything about this, but i'm curious.

i would say synthesidt is not worth it. seriously, even if youre totally new to that kind of modular software, dont lose your time with reaktor or shit like that. print the max msp tutorial and start reading. you WILL learn max and also some very useful audio concepts that will follow you through your musical career.

  Duke Remington said:

dont tell anyone, but if you can get your hand on an analog filter like a sherman bank, alot of these minimal guys put every single track through it. its all about the enveloppes and hp/lp/bp sweeps.... FFFFFSHHHIIIOOOUUUUU.

DO NOT, i repeat, DO NOT BUY A SHERMAN FILTERBANK IF YOU WANT TO MAKE DUBBY MINIMAL TECHNO!!!!!! :))

my filterbank is great for slowly shifting, raw and in-your-face soundscapes. awesome, powerful sound but it distorts really easily. it is VERY hard to get subtle and smooth sounds out of it. it's like some kind of rabid animal, it feels real organic when playing it and it's pretty complex (you can do far-reaching CV stuff with it too).

if you want to do clean filtersweeps, you can do that with software (easily) or with almost any other random filter.

but not with a sherman.. :)

btw i honestly have no clue about this 'colouring' and 'toyish' of reaktor that you guys talk about.. post some examples? to compare a clean sine kick from reaktor with one from max...? i'm curious what this 'toyish'-sound is...

also, if you post a minimal, dubby techno track that you've made yourself, i might be able to give some advice, composition- or soundwise.. but (m)anyways, i'd say that attention to sonical detail and an arrangment that grabs and keeps your attention is the most important.. listen very closely to monolake's hongkong (one of the classics in this genre IMO) if you haven't done so already.. for me, that album is the essence of dubby techno..

bowen: using a PC :( i've just been looking at plogue bidule again - might be worth checking out...

duke: yeah, they do that with so many synths these days... it's silly, but just making a sine wave in soundforge and applying a pitch envelope to it produces a perfect bass drum - and you try exactly the same thing using some arturia moog modular, even operator or u-toniq, and it just sounds small, unbalanced and undefined

to be fair, i don't really get much better bass drums from the Z1 or Virus either... can only assume it must be something to do with the VA algorithm... but i'd get cleaner sounding bass drums and toms out of the CS-70, and that's saying something! even the case is warped by its own presense

attention in music is a tricky one... aparently, the brain really can only consciously track one pitch at a time - and that's usually the highest sounding pitch you lock on to - but then the bass counterpoint certainly changes the context of that melody... hearing the bass only ever relative to the lead line?

and then of course, all the added notes have this qualatitive effect on the mood and character of the harmony, but we're certainly not tracking them in the same way we would with the lead or bass line

i think i've read somewhere that even with 2 ears, we still only hear a single sound consciously at the time

there was some experiment where they'd play two different words at once, l & r, like "rainbow" and "spinach" - and very few people would be able to register both consciously... the brain would deliberately try to ignore one side

yeah, hong kong's one of my favorites

but i'm more interested in taking ideas from dub techno - like this concept of minimalism and simplicity, and just letting things roll, and applying it to my own work... there's something very egoless about allowing yourself to release work which has taken no effort - i mean a lot of herr henke's compositions are heavily generative... i think all the effort must go into the construction of these patches

maybe it's just a kickback from doing something as anal and production heavy as dnb for so long

  Duke Remington said:

i would say synthesidt is not worth it. seriously, even if youre totally new to that kind of modular software, dont lose your time with reaktor or shit like that. print the max msp tutorial and start reading. you WILL learn max and also some very useful audio concepts that will follow you through your musical career.

yeah i know max/msp, but like moon said it's a little too big for some stuff. i always make it into a programming project instead of doing something musical with it. i have to let it go for a month (or as long as it takes for me to forget how it works), to really get to use it as a musical device.

interesting stuff, about only hearing one sound at a time, I guess although at one moment in time it is only one sound, we can switch fast enough to hear more, kind of like how a film appears to be moving but is 24fps.

I wish i'd taken the psychoacoustics module in uni now, ahh well too late for regrets,

on the question of uni though, how many of the max users learnt it in a structured environment? I did, I'm pretty sure Forlon is studying something similar and also Plugexpert, and Duke did you say something about electroacoustic classes? did you learn max in a structured way? Not to put anyone off using it I think its superb but I know I wouldn't have got as far had I not had to do coursework with it.