Yes, ID have to be unique. ID allows you to target actions on an specific object. For example, to move an object, you can type (in script) setPos 3 0 0 0 to place the object with ID 3 at the position (x,y,z)=(0,0,0).
However, there is a mistake in our scripts examples that will be fixed for next release. We don’t want to prepend ID with « 100 » or « 200 » but we want to make ID range (0->1000 for cursors, 1000->2000 for curves, 2000+ for triggers).

@vvvv1 wrote:

is it possible to have a look at your KSpluck synthdef aswell?

thanks a lot for the help! 😉

… sorry, quoted the wrong question first time …

There’s a version of KSpluck, along with more physically modeled SuperCollider patches, at http://www.informatics.sussex.ac.uk/use … lling.html

@Guest wrote:

Hey thanks dick for everything but i still got some problems

when i change my sawbass arg note to freq to have iannix’s triggerIndex control it’s pitch it just clips and crashes…
hmm don’t get much on how Iannix messages works and the one you wrote either
or perhaps it is just some stupid mistalkes i’ve made
it will be great if i could get some explanations for newbies like me!: )
thanks a lot ; )
(
SynthDef(
« sawbass »,
{
|note = 35, amp=1.2, cutoff=350, decay=1|
var env, audio;
env = EnvGen.kr(Env.new([0,1,0],[0.001,2],[3,-50]),doneAction:2);
audio = RLPF.ar(Saw.ar(note.midicps,amp),cutoff)*env;
audio = CombC.ar(audio,1,0.125,decay);
Out.ar([0,1],audio);
}
).send(s);
)
(
g = OSCresponderNode(nil, ‘/trigger’,
{ |time, responder, msg|
var pitch, octave, triggerIndex = msg[1];
var scale = [ 28, 29, 31, 33, 35, 36, 38 ].midicps;
var note = [ triggerIndex % 7 ];
octave = (triggerIndex / 7).floor;
Synth(« sawbass »/*, [ freq, note*(2**octave), amp, 1.5]*/);
}
).add(s);
)
NetAddr.langPort

well…

I tried your program out, found it made a few low grumbly noises and then quit on a memory allocation fail.

I’m probably not the guy to diagnose your SC troubles, just a beginner myself. A glance at your code above suggests you might be generating a lot of Synths that aren’t being removed fast enough. That would cause the memory fault. The pitch selection logic seems a bit odd. I tried your code with the sawbass decay a bit shorter, and pitch selection tweaked as follows, and it worked for me:

(
SynthDef(
« sawbass »,
{
|note = 35, amp=1.2, cutoff=350, decay=1|
var env, audio;
env = EnvGen.kr(Env.new([0,1,0],[0.001,0.4],[3,-5]),doneAction:2);
audio = RLPF.ar(Saw.ar(note.midicps,amp),cutoff)*env;
audio = CombC.ar(audio,1,0.125,decay);
Out.ar([0,1],audio);
}
).send(s);
)

(
g = OSCresponderNode(nil, ‘/trigger’,
{ |time, responder, msg|
var pitch, triggerIndex = msg[1];
var scale = [ 40, 41, 43, 45, 47, 48, 50 ];
var note = scale[ triggerIndex % 7 ];
Synth(« sawbass », [ note, note, amp, 1.5] );
}
).add(s);
)

thanks for your reply, it works better but still not great !!
sc obviously can’t take too many triggers at the same time,
i wonder if there is a way to work around that

all the best

Yau

@Guest wrote:

thanks for your reply, it works better but still not great !!
sc obviously can’t take too many triggers at the same time,
i wonder if there is a way to work around that

all the best

Yau

well … I’m pretty sure SuperCollider can receive and process plenty of OSC traffic, certainly as much as you can transmit from IanniX. The question is what your program does with the traffic.

Try this:
OSCresponderNode(nil, ‘/trigger’, { |time, responder, msg| msg.postln }).add;
OSCresponderNode(nil, ‘/cursor’, { |time, responder, msg| msg.postln }).add;

Then see how many cursors and triggers you can add before SC gives up or misbehaves. I think you’ll find IanniX hangs on the send and display side before SC stops handling traffic. Yes, you will hit trouble if you try to instantiate a thousand UGens at once. Is that a limitation of SC, or of your hardware config?

The biggest problem with your patch is the comb you have echoing each new note. I couldn’t figure out what you intend the relation between the echoes and the pace of triggers to be. I believe that fixing your note/frequency selection logic and removing the comb will make your patch perfectly well-behaved.

cheers

Thanks Bob for debugging some OpenGL bugs!