Electronic Music 411 - reaktor: glitches and techniques

07/08/08 + mpg

ok, so playing the ignorant newbie right now (beginner's mind) i would like to talk about reaktor and it's functionality. more specifically, i would like to talk about it's ability to process signals in random ways and create new kinds of sounds. obviously, as the thread title shows, i'm talking about glitching types of processes.

personally, i am looking for ways to destroy samples for use in rhythmic ways, or destroyed in rhythmic ways (?). normally i use chains of vst instruments and occational some kind of reaktor module (grain delay). it seems to me, that in reaktory, the granular processing is the best way to completely obliterate audio signal and turn it into something completely different. this process could be a very long and slow process or a very and fast short process (time makes a difference? speed? ideas?)

this could be very simple or very complicated. for instance, one really simple way to get interesting sounds is to randomize a sinewave frequency in some kind of atonal scale send the audio output into something like DFX scrubby which then in turn goes into something like a dblue glitch, or some kind of processor that is "time based" "randomized" "edits"... or substitute glitch for something like livecut.. whatever, conceptually you see what i am getting at.

for this purpose, let's not talk about the practicality of this in a live situation, let's look at it in terms of recording and sound design. so then, say, you look at someone like richard devine who is clearly very well versed in sound design since that is his job that he gets paid for. and then bring it back to: ways to chew up samples, destroy sound, turn sounds into new abstract objects or even "gestures."

do not be shy here.. this is a thread about magic. i should also like to say, if you do things outside of reaktor (max/msp perhaps) and wish to share concepts, please by all means. this is about concept and thought, not so much execution (althought execution is equally as valid and important and if you have thoughts about this as well, express them).

07/08/08 + mpg

link

07/08/08 + eyesnine

check out my releases, made in reaktor (except the latest one) is that the kind of thing you're talking about?

in reaktor you'll have a lot more luck sampling than processing signals. unless you happen to know core inside out and feel like programming your own time stretching algorithms utilizing audio buffers.

get yourself familiar with the grain resynth sampler module. you can do a lot with it. grain cloud you can do more with, but with a huge hit to your cpu, so you'll be limited in your polyphony. tempo synced parameter changes will yield some good results. the rest is a matter of experimentation, like what parameter settings work for the samples you're using. also, you'll have to figure out a way to sequence everything... which is the real magic trick. shouldn't be too tough if you're familiar with reaktor, though.

Recent blogs: Reaktor Tutorial - Control Structures pt 2. Routers, Reaktor Tutorial - Control Structures pt 1. Separators, Reaktor Tutorial - Signal Flow

edited: Jul 08 2008 + mpg

thanks for response. i should clarify: this isn't about ME and whatever i know. this is about increasing the robustness of introductory knowledge, and possibly coming up with new theoretical knowledge, if the accident were. perhaps this kind of discussion is not a very good vehicle for new theoretical knowledge or abstract thinking to come about?

i am wondering..

07/08/08 + eyesnine

well, i don't think "glitch" is really that complicated. it seems like you summed it up pretty well in the thread introduction. thats pretty much what will make your music sound glitchy. the rest is really about stuff that works well in any genre of music. complimentary/contrasting processes, rhythm/tempo/shuffle/swing, velocity/accents, note/parameter sequencing, message/concept, pacing, managing silence, textures, instrumentation, harmony/melody, development, etc...

Recent blogs: Reaktor Tutorial - Control Structures pt 2. Routers, Reaktor Tutorial - Control Structures pt 1. Separators, Reaktor Tutorial - Signal Flow

07/08/08 + eyesnine

also, this thread probably isn't a good vehicle for new knowledge to come about. thats what practice and experimentation are for. it may be a better vehicle for hearing about something somebody else is already doing, if thats what you're into.

Recent blogs: Reaktor Tutorial - Control Structures pt 2. Routers, Reaktor Tutorial - Control Structures pt 1. Separators, Reaktor Tutorial - Signal Flow

edited: Jul 08 2008 + mpg

you're probably right... talking about quantum mechanics may not be the best vehicle for expanded understanding.

i would have to be more abstract probably.

analog vs digital is also involved here

edited: Jul 08 2008 + mpg

and my goal was to apply concept to glitching, so maybe i failed there in trying to bring forth new information with using old terminology.

glitch is obvtiously, really, a random process... how a cd skips in real life and whatnot.. can it be recreated... buffer skipping whatever terminology you wanna use

FREE JAZZ

07/08/08 + astroid

i can't resist the thread

one thing that i'd be trying, if my fucking g2 weren't broken, is progressive/algorithmic resampling. it could be done at the macro or micro level, with rules applied to the same. at one end (just what i think, i haven't tried this yet), it'd sound like a delay, at the other end it'd be something like a timbral blur. then, instead of just simple feedback circuits, you could put testing circuits in to steer the tendencies. the testing circuits, since they are one removed from the business of manipulation, would probably be a smoother way to gesturally control the manipulation. :D

Recent blogs: thank you, it WAS my birthday yesterday, wetf

07/08/08 + eyesnine

well... my take on the whole thing is that i hear too much stuff that doesn't utilize classic musical techniques that have been around for thousands of years. its like people seem to think that a beat is a brand new invention and they're just hammering it into the ground. as far as i'm concerned that horse died a long time ago and the people making good stuff now are the people that realize that the real trick is to make music, just how its always been made, but well, and with new instruments, to expand their sonicality and compositional freedom. not to say that a beat makes a piece of music worthless, just to say that its just one tool in the toolbox.

classical music isn't a good representation of how people made music hundreds of years ago, its a representation of the only means people had to record music. i think if audio recording technology was two thousand years old, you'd have found that there probably was a greek band that sounded just like kraftwerk, autechre or whoever, but with wood, animal hide, stone and voice instead of electronics or software. how do you write music like that down though?

in summary, its just the same old stuff in a fancy new package.

hopefully there's some expanded understanding for someone, somewhere in my last few posts.

Recent blogs: Reaktor Tutorial - Control Structures pt 2. Routers, Reaktor Tutorial - Control Structures pt 1. Separators, Reaktor Tutorial - Signal Flow

07/08/08 + lysdexic

astroid said: "i can't resist the thread

one thing that i'd be trying, if my fucking g2 weren't broken, is progressive/algorithmic resampling."

unbreak your g2 and post some audio plz.

lys.// np: toni braxton - unbreak my g2

Recent blogs: lys.mix//10.07, slsk 1 second massacre, myspace page

07/08/08 + mpg

astroid said: "i can't resist the thread

one thing that i'd be trying, if my fucking g2 weren't broken, is progressive/algorithmic resampling. it could be done at the macro or micro level, with rules applied to the same. at one end (just what i think, i haven't tried this yet), it'd sound like a delay, at the other end it'd be something like a timbral blur. then, instead of just simple feedback circuits, you could put testing circuits in to steer the tendencies. the testing circuits, since they are one removed from the business of manipulation, would probably be a smoother way to gesturally control the manipulation. :D"

THANK YOU!

you have layers of thought, you have looping... to me this question is like "why are things beautiful?" why does this work out, what is the aspects of aesthetic that allow you to accept a d use the output? when does one say: this is no good to use.?

07/08/08 + astroid

for me, what i want in my music, is a certain amount of transparency of structure, probably at the medium level, where it can dance just outside of the range of the short-term memory. i want to have those phrases become an object, and then the objects to create the form.

the structure could be anywhere, and i also have been enamored of the idea of mapping structure in a place where in doesn't usually belong. -deep structure at the phrase level. algorithms make it possible to have that structure at places and levels where you would have needed a few weeks with a calculator to render one second. with a little visual imagination, you can turn shapes into sound and then give them rules to propagate.

that fascinates me to no end, the idea of the composition of a song being the way in which it manipulates the material you add in real time. chains of fx or midi manglers-like an intelligent dblue or arpeggiator. on a modular system, you could extract the structure of development and control from what it actually does. this would be like using the logical action of an arpeggiator to control a resampler: pitch-ordering maybe translating into timbral filtering. that's a quite basic example. you could map a whole sonata form, or anything else, cellular automation maybe, into something as small as a timbre.

when will people learn how to do this well? who knows.

Recent blogs: thank you, it WAS my birthday yesterday, wetf

07/08/08 + eyesnine

^^ hey, thats what i'm working on! ;)

Recent blogs: Reaktor Tutorial - Control Structures pt 2. Routers, Reaktor Tutorial - Control Structures pt 1. Separators, Reaktor Tutorial - Signal Flow

07/08/08 + Roshi

I like doing wierd things with parameter automation to help delineate song structure.

The other day, I was playing with noise and used a tremolo to make peaks in the audio - and then sent it through a gate with a high threshold. By playing with phase and frequency of the tremolo, I could get some pretty strange rhythms out of it...

To me, it's about giving up a level of control or controlling the texture of a process at another level.

I am not quite at the astroid level of musical thinking yet, though.

Recent blogs: I heart Ondes Martenot, Summer Reading - The rest is noise, oogazer set