On Thu, Sep 30, 2004 at 01:46:10PM -0400, Peter Lutek wrote: > > Instead of keeping many different versions of > > the actual audio files, you would keep different iterations of the > > mixdown/effect settup scripts. > thanks, eric! yes, this meshes better with my workflow -- i'll look into > it. > but you now have me curious. can you describe your workflow, in a way > which would illustrate how it differs from what i've described? i am > NEVER interested in being stuck in a rut, and would like to learn from > an alternate viewpoint. Ok, but bear in mind that at present and probably for several more years I am not attempting to make music at all. I'm really just playing with sounds to hear what might be possible. I find it liberating to not be directly concerned with pitches, scales, harmonies and rhythms. Instead I am letting those happen as they will as a consequences of manipulating sounds with time shifts, reversal, looping and effects processing. Some idea of what I do might be understandable in these scripts and notes: http://zhevny.com/bin/ I need to do an rsync tonight to update that, but what's there should give a decent idea, if not a very clearly organized one. I have a huge pool of mono files, about 100GBs now. This pool includes previously completed works (split to mono if they were originally stereo), digital recordings of a wide variety of sounds I originally collected on cassette tapes over the past 8 or so years, recordings of myself playing guitar, piano and bass trombone, some recordings of myself speaking gibberish or reading written works of my own, sounds collected more recently with mini-DV camcorder, etc., etc. My scripts largely center around randomly selecting a random number of files from this large pool, selecting a random portion of the selected file, applying a random degree/direction of pitch shift, looping the result and mixing that together with various panning, volume and effects controlled either by randomly generated controllers or live by myself via midi. There are numerous variations on the above planned. They are in various stages ranging from rough ideas sketched out in plain text through hacked together mish-mashes of shell and python all the way up to a few that are actually in use daily. Eventually the random selection process will feed the generation of .beef banks for use in specimen, and perhaps soundfonts, to create midi playable instruments w/arbitrary scales/tunings. The ardour based projects I speak of will grow out of this as well. The sound materials will be again selected at random. But, I will spend more time with a chosen set, studying the sound of various autogenerated submixes, playing with and getting to know a set of sounds over a period of weeks or months. Then, eventually I'll sit down to lay out a more structured composition. For this final stage I agree that ardour will be a better suited tool. Perhaps this last type of work is most like other people's work flows. In addition to those sound manipulating type of work, I'm also using ecasound as a LADSPA host to control the paramaters of some sound generating plugins via midi CC messages. These session are primarily improvisatory in nature with now set plan of action ... I just slide and dial and tweak to hear what happens and revel in the exploration. In many ways I'm am simply and happily playing as a child would with a bucket full of building blocks. -Eric Rz.
