what would you do? I have no fucking clue. The recorder player seems to want me to have a sampler, which I don't. I have supercollider and trying to generate a sample loop is causing me mad timing problems and also, wtf is with this "live" stuff? who does things live? ok, aside from everybody else on earth. jesus god, i have like 5 days (practically) in which to pull something together and almost nothing in the way of a clue.
Last night, I dreamt that I had to pass an exam in order to get some sort of certificate from my school (have I mentioned that I got confused and missed the last class?). The exam booklet was like 100 pages long and nobody told me I only needed to do the first few pages, so I skipped over them to do the easier problems first. Also, doing the test required me to wear these headphones in a weird stretchy helmet which was squeezing my head with great force. Anyway, I got no credit whatsoever and they told me to leave. Also, the school doubled as a bike shop and Stephan, our tech guy was also a bike mechanic.
Hey, I have an idea. I could break up a recorded stream into long grains and scramble them! oh, no, wait, that sounds like shit with a recorder. no, i could learn an entirely new thing like bbcut of fft in record time only to have it (A) exhibit unexpected behavior which sounds like shit. (B) no, wait, A sums it up.
Fast solutions = fast hacking, right. I'll have something in the next 2. 5 hours, i'm sure. @#@#%%@#Eg3tt675rfvhjksvDhjkvsdfhjkgsdfhjkfsd
Tag: Celesteh
2 comments:
Well... before you switch the reverb on you could:
Add small and slow detunings à la Nono:
interval = 0.66666666;
input = In.ar(in);
signal = PitchShift.ar(input, 0.1, SinOsc.kr(Rand(0.01, 0.009), 0, interval).midiratio)
* amp * EnvGen.kr(Env.asr(1, 1, 1), gate, doneAction: 2 );
signal = input + signal;
Or make her play pppp and add lots of amplification.
And yes, "live stuff" sucks!
Be careful though, flute and live electronics have a long tradition of... "being hard to work with".
PitchShift.ar!!! Wow, I have got to look through the UGen list more carefully.
The recorder (and flute) is just so pure that the slightest disturbance shows through like dark underwear and light linen pants.
I can do any granularization, really. But pitchshifting might work.
What would be nice is if one of the PV ugens could pick out the N most strong frequencies and give them back to you, so you could use them as arguments to oscillators. Maybe one already does that. I haven't done anything thus far with FFTs. Anyway, that would be nice for pulling apart multiphonics, since you could quietly echo and then add in some slew.
SinOsc(freq_2arr + [slewUp, slewDown])
or something. That would be nice.
Post a Comment