Re: XORNOT set recording

Iain Duncan <[email protected]> Tue, 19 Jul 2005 12:20:26 -0700
Newsgroups gmane.comp.audio.csound.tekno
Message-ID <[email protected]>
> I have listened to your set and I like what I hear.  Maybe you could give me
> some advice about how you are working with the tools you mentioned to
> realise this in a live situation.
> 

Thanks very much for the compliments! We have a whole wack of code we've 
been working on for a few years now, only about half of which is 
actually being used right now for the set. There has been a lot of 
experimentation to try and find the best balance of what we should be 
doing as we play for fun and creativity vs how much we should 
pre-prepare to make it sound good.

Right now, the musical lines are all preprogrammed, and we are basically 
remixing on the fly. We make them in gvim using the python extensions I 
wrote to make gvim look/feel like a python tracker. Then we use my 
python-applet "makescore.py" to turn text arrangement files of the 
tracker files into big assed score files. The score files include all 
the actual notes along with patch snap-shots at the beginning of each 
tune section. The big synths and the mixer channels and fx are using 
tables to control any dynamically modifiable paramaters, so the 
patch/mixer snapshot instruments just slam a bunch of values into those 
tables. Preparing the tunes took only about 3 weeks in total, but that 
was largely by necessity!

We also wrote a bunch of midi input parsing instruments in csound, and 
that allows us to mute and unmute tracks and alter the synth and mixer 
paramaters using hardware controllers. So at the moment, that's all 
we're doing live. We have written a sequencer engine and sequencer input 
modules for doing real time improvised step sequencing as well, but at 
the moment, that is on hold till we think our abilities and amount of 
input hardware are up to adding it back into the set. Much as I love 
improvising, I'd rather improvise less to a full house than improvise 
the whole thing to an empty dance floor!

Csound5 has so many huge advances that the next step for me is to redo 
the input module in C++ interfacing with csound via the api, which will 
allow a lot better code control, and also to then reincorporate the 
sequencer engine for more improvisatory possibilities. The hard part is 
balancing code work with writing tunes and getting shows done! But we're 
learning...

Thanks
Iain

_______________________________________________
csoundtekno mailing list
[email protected]

Subscribe, unsubscribe, change mailing list options:
http://plot.bek.no/mailman/listinfo/csoundtekno