[PD] reverse delay for live signal processing?

[Peter Plessas]

Hi Kevin,

i think you are getting dropouts when using arrays since the graphical 
data in the array object is being updated. Use the [table] object (which 
is the same as an array, but without the graphical display right 
visible. The waveform is accessible once you double-click it).

Using arrays/tables you can also loop backwards, which is not possible 
using delread/write.

So you would have to write into an array cyclically (phasor~ is your 
friend here), and read at the same position, feeding back the audio 
signal from the tabread to tabwrite, thus creating a loop.

You would have to force pd to do this one after the other ("order 
forcing") by putting the into a subpatch each, and connect those 
subpatches via audio patch cords. (there is also something on this topic 
in the help patches in you doc folder i think).

Once you got this working, don't forget to check out the "vasp" library!

Good luck,

Peter

Kevin McCoy wrote:
> Hello listy,
> 
> A lot of my friends play instruments I would like to process in 
> realtime.  I
> understand that one popular approach is with delay objects.  At first I was
> wondering if there was a special array that didn't give dropouts when it 
> was
> rewritten in realtime.  I don't think there is?
> 
> But similar things could be accomplished if there was a way to access a
> [delaywrite~] buffer as dynamically as you can an array.  One example would
> be reading the buffer backwards?
> 
> I'm eventually looking to build something similar to one of those Boss
> Loopstations where you can overdub, reverse, etc in realtime.
> 
> Granted, I know next to nothing about how these things are actually coded
> inside, but I thought I would pose the question and ask for ideas.
> 
> Thank you!
> Kevin
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> PD-list at iem.at mailing list
> UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list