[PD] looking for a non-linear stereo pan object

Marco Scoffier marco4linux at earthlink.net
Mon Mar 3 00:30:26 CET 2003


I have made a little patch to do panning, but it is a linear
pan which leaves a great big hole between the two speakers.  I remember
a few years back at IRCAM a demo of using a table with a handdrawn sort
of logarithmic? curve to get a nice pan (in Max), but I am feeling a little
clueless and helpless right now, and any help is appreciated. Below is a
list of the things I have tried so far.

Searching the pure-data.org site comes up with nothing.
I cannot find a LADSPA plugin either.

I have tried the pan~ in johannes zmolnig's "HOWTO write an External"
http://iem.kug.ac.at/pd/externals-HOWTO/node6.html
but it mixes two input's to one output, and seems to be linear also.  I
want one in to two outs.

vbap in the externals collection from sourceforge isn't compiling for me
on debian unstable linux :
I keep getting errors like this:
1: warnings being treated as errors
vbap.c: In function `vbap':
vbap.c:137: warning: `winner_set' might be used uninitialized in this
function

If I initialise these variables to zero, I keep getting more
uninitailised variable errors, once I have initialised them all 

Error 2 in define_loudpeakers.c with no line number... nothing.  

I tried removing some of the warning flags in the makefile, but this is
a bit above my head.

the precompiled vbap.pd_linux is loaded by pd but there is no
precompiled define_loudspeakers.pd_linux so I can not use the object.

<rant> couldn't somebody put pan or spacialize or something with the
vbap name somewhere.  I mean there is not even a readme saying what vbap
does</rant>

I have also tried audience~ but I don't need a gui and the install
script from the makefile doesn't put the images speaker.gif and
wander.gif in a place where the audience~ object can find them.  I have
tried moving the images around from pd/externs/audience~/ to pd/externs
to pd/bin but I always get couldn't open "speaker.gif": no such file or
directory

Thanks for any pointers,

--Marco
-- 




More information about the Pd-list mailing list