[PD] send recieve efficientcy

Frank Barknecht fbar at footils.org
Fri Dec 26 21:44:56 CET 2003


Hallo,
Bradon Webb hat gesagt: // Bradon Webb wrote:
> The more efficient areas that I would like to have are
> in the recieve sections.  such as in [pd toggles]
> where there are recieves $0-step1 through $0-step16
> and also where the notes are recieved at $0-n48
> through $0-n84 in [pd notes].  I was wondering if
> there was a way to recieve all the revieves at one
> location that could carry a variable signifier or some
> kind of range.  this is mainly to eliminate the need
> for me to connect 40 recieve boxes to one inlet..., in
> the future I may want to do something with over 100
> recieves and connecting all those boxes seems like it
> could be done a better way. 
> 
> Thomas had the suggestion of using nqpoly abstractoin
> (see below) however I am not familiar with how to use
> that to do what I want.

I cannot say much about dyn or poly, but if you need to to often
repeating tasks, you should try to get comfortable with Pd's internal
messages. You can find the needed docs e.g. at pure-data.sf.net in the
CVS.

I often use little "builders" for these things, that write parts of a
patch into a subpatch as a scratchpad, and then I copy these over to
the real patch.

Another thing to consider is how to modularize your work. Do you
really need a 100-step-sequencer, or would 10 10-step-sequencer do it
as well? 

If yes, then write a 10-step-sequencer, and reuse it 10 times. If you
then need a 1000 step sequencer later, reuse your 10*10 step sequencer
10 more times.

Or even better: let the 10 step sequencer have a method for changing
state, so that you can have one 10-step thingie visible, but it
represents an arbitrary length of sequences.

ciao
-- 
 Frank Barknecht                               _ ______footils.org__




More information about the Pd-list mailing list