Kas,
Friendly hint; Prepare for a bit of disorientation after you played a bit and resume talking, in my experience it's a good idea to think of a first sentence to say after you stop playing right before you start.
This is really good advice, thanks. I hope I can pull it off. I mean, it's hard enough trying to get things to work properly, but then it's kind of a strange thing to play for people when they have no idea what you're doing (or whether or not you're just there checking your email while your music plays ;-). I think that this approach is more approachable when so much of what is happening is "under the hood."
... Of what I've seen that code-base is one of the most interesting structures build in ChucK so far (my own code for long-term projects is very, very boring on a code level, I fear) so if I can vouch for anything it must be that those in the NYC area would have a great chance of seeing something very interesting at your performance.
I can only hope that's true, Kas! Thanks for the kind words. The code-base has changed dramatically since you've seen it. I think that the original way of doing what I was doing sacrificed so much performance in order to gain the programmatic control I needed, that it lent itself much easier to a non-realtime process in the end. What I'm using now isn't quite as open-ended, but it much better suited for doing performance in realtime. I got a lot of mileage out of using CurveTables as envelopes. Dimitris and I were on about that not too long ago.
BTW, I hope you'll be writing on this list about those Spear files later as well. Word on the (rather imaginary) street has it that people are pushing for a official release to include the file IO. So far I only saw the design on the Wiki and if I'd have to go by that alone I see no chance for even parsing files in ChucK at all. File parsing should be exciting; if we can parse Spears files we should be able to parse (and remix) .ck files, then save and run them. For all it's pervesity (and quite likely because of it) I find that idea very hard to resist indeed.
I will definitely be putting that together soon. You have every right
to be totally underwhelmed by the file IO functions in CVS (at least,
the version I'm using anyway)--they are very primitive. I really wish
that ChucK handled strings better, but you can get away with using a
lot of Std.atof() and Std.atoi() functions, although I do wish that
they could fail with NaN rather than default to 0.0 if you feed them
something like "asdf12.97asdf."
What we have to look forward to are some file IO objects with various
read functions - readFloat(), readInt(), and readline() - a simple
tokenizer that splits on white space, and a way to write files too. I
don't know if I would try to parse .ck files with them... but it
can't hurt to try! It would be fun thing to mess around with.
For some of my projects, I've been using OSC as a communication tool
between ChucK programs and other running things. Now I'm really
excited about the prospect of using fifo pipe in Unix to communicate
lots of information between ChucK and something else--a simple
numbers-only protocol between Python and ChucK over a fifo pipe would
give a ChucK program access to Python's string-parsing abilities...
Anyway, the possibilities are endless and endlessly interesting. I
would love to see this functionality in the next release. (I would
love to see the next release!)
-Mike
2009/8/29 Kassen
Mike;
I'm planning on doing a quasi-participatory didactic thing with the audience to teach them about granular synthesis as I demonstrate the instruments I've built, so it's a little non-traditional as far as concert flow.
The times when I saw Michel Waisvisz play he performed like that, gradually shifting the focus from a talk to a concert. If done well it can solve a lot of the problems with non-traditional performance methods. I tried it a few times as well and while it's hard to come anywhere near Michel's level (I consider him to have been one of the great minds and performers in the history of electronic music) it's a underapreciated form.
Friendly hint; Prepare for a bit of disorientation after you played a bit and resume talking, in my experience it's a good idea to think of a first sentence to say after you stop playing right before you start.
Even if you won't be in the area, you might want to take a look through my (spaghetti) code. As Kassen can tell you, my attempts at doing granular synthesis in ChucK border on obsession and have spent a long time trying to squeeze out as much performance while keeping it as easy to program as possible.
I like obsessions (well, healthy obsessions). As I experienced that code the real obsession there isn't actually with the grains themselves but with your perspective on- and your control over them, which is of course a far more stimulating mental illness. Of what I've seen that code-base is one of the most interesting structures build in ChucK so far (my own code for long-term projects is very, very boring on a code level, I fear) so if I can vouch for anything it must be that those in the NYC area would have a great chance of seeing something very interesting at your performance.
There's a possibility that I'll document the show, so there might also be video at some point.
I hope you'll find the time and attention to do that; I know how hard it can be to have to mind both your own performance and record at the same time. I have of late been very bad at that.
BTW, I hope you'll be writing on this list about those Spear files later as well. Word on the (rather imaginary) street has it that people are pushing for a official release to include the file IO. So far I only saw the design on the Wiki and if I'd have to go by that alone I see no chance for even parsing files in ChucK at all. File parsing should be exciting; if we can parse Spears files we should be able to parse (and remix) .ck files, then save and run them. For all it's pervesity (and quite likely because of it) I find that idea very hard to resist indeed.
Yours, Kas.
_______________________________________________ chuck-users mailing list chuck-users@lists.cs.princeton.edu https://lists.cs.princeton.edu/mailman/listinfo/chuck-users