StereoConvolution2L / Convolution2L

View: New views
3 Messages — Rating Filter:   Alert me  

StereoConvolution2L / Convolution2L

by Sven Hahne-2 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

dear list,

i´m trying StereoConvolution2L for HRTF with the a sine tone for  
testing.
As soon as I start a new kernel (with another impulse response  
buffer) a click occurs...
Is there anything I have to take care of / preparing the buffers in
a special way to avoid these clicks ?

thanks,
sven


_______________________________________________
sc-users mailing list
sc-users@...
http://lists.create.ucsb.edu/mailman/listinfo/sc-users

Re: StereoConvolution2L / Convolution2L

by nescivi :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

On Wednesday 07 May 2008 16:59:45 Sven Hahne wrote:
> dear list,
>
> i´m trying StereoConvolution2L for HRTF with the a sine tone for
> testing.
> As soon as I start a new kernel (with another impulse response
> buffer) a click occurs...
> Is there anything I have to take care of / preparing the buffers in
> a special way to avoid these clicks ?

sine tones are the worst test material for this, as you will always get
artefacts, even when interpolating.

which version of SC are you using (last change to this UGen was 14th of
February) and what code are you using? please post!

sincerely,
Marije
_______________________________________________
sc-users mailing list
sc-users@...
http://lists.create.ucsb.edu/mailman/listinfo/sc-users

Re: StereoConvolution2L / Convolution2L

by Sven Hahne-2 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message


Am 08.05.2008 um 09:34 schrieb nescivi:
On Wednesday 07 May 2008 16:59:45 Sven Hahne wrote:
dear list,

i´m trying StereoConvolution2L for HRTF with the a sine tone for
testing.
As soon as I start a new kernel (with another impulse response
buffer) a click occurs...
Is there anything I have to take care of / preparing the buffers in
a special way to avoid these clicks ?

sine tones are the worst test material for this, as you will always get 
artefacts, even when interpolating.

I see...

which version of SC are you using (last change to this UGen was 14th of 
February) and what code are you using? please post!

i did a svn update and recompile today...
basically the clicks are not noticable with other sample material.
so if you think that this is a basic problem...

thanks for help :)

cheers, sven


d = d ? ();

d.leftBuffer = Buffer.alloc(s, 512, 1, bufnum: 0);
d.rightBuffer = Buffer.alloc(s, 512, 1, bufnum: 1);

d.fileL = File("pathToKemar/full/elev0/L0e090a.dat", "r");
d.leftBuffer.setn(0, d.fileL.read(Int16Array.newClear(d.fileL.length)) / 2.pow(15) );
d.fileL.close; 

d.fileR = File("pathToKemar/full/elev0/R0e090a.dat", "r");
d.rightBuffer.setn(0, d.fileR.read(Int16Array.newClear(d.fileR.length)) / 2.pow(15) );
d.fileR.close; 

d.leftBuffer.plot
d.rightBuffer.plot

SynthDef( "conv-test", { |t_trig = 1, kernel1 = 0, kernel2 = 0|
var input;
input = SinOsc.ar(440, 0, 0.5);
Out.ar(0, StereoConvolution2L.ar(input, kernel1, kernel2, t_trig, 512, 1, 1.0));
}).send(s);

x = Synth.new("conv-test",[\kernel1, d.leftBuffer, \kernel2, d.rightBuffer]);

// click occures
x.set(\kernel1, d.rightBuffer, \kernel2, d.leftBuffer, \t_trig, 1);


_______________________________________________
sc-users mailing list
sc-users@...
http://lists.create.ucsb.edu/mailman/listinfo/sc-users