From: "vanDongen/Gilcher" gml@xs4all.nl :
http://music.columbia.edu/~matt
From: "CK" x@meta.lo-res.org :
or
From: "Greg Rippin" gmr222@nyu.edu :
Amen. I've been trying to pipefit csound with pd for awhile now, and it
has
been quite difficult.
From: "Krzysztof Czaja" czaja@chopin.edu.pl : (about csound opcodes)
Hi,
Some points still remain unclear to me :
source code to port an opcode to Pd as an external ? (what exactly are those "licensing clashes" ?)
Krzysztof, are you talking about some 'csound-opcodes-plugin' object ?
Is someone here working on this kind of pd 'csound~' external project ?
Is it possible to build such an external for Pd ? (Greg, what are the main
difficulties you had to deal with ? )
solution in term of performance, features and useability ?
It seems someone recently did a "Call for a pd Csound Object/plugin~" on the csound ml. Any news ? (the csound list web archive is unreachable tonight). About porting the max/msp csound~ object to Pd, did you (CK) get any info or source code from Matt Ingalls ?
best regards nicolas
PS : any help or answer about my previous post ("templates...") would be greatly appreciated :)
On Fri, 2002-01-25 at 22:37, Nicolas Lhommet wrote:
The original csound code is under a educational+personal use only license. It is not GPL compliant, nor does it come close to matching to Open Source Definition(TM).
There are "clean room" implementations of most of the interesting csound opcodes that were made as part of the now defunct Quasimodo project. These were salvaged from Quasimodo and turned into LADSPA plugins. I don't remember who did this, but any decent LADSPA plug in collection will include them.
I think it might be more interesting to make csound and pd interface to JACK. No Windows version of JACK, however. Then there's also that control-stream vrs. audio-stream issue.
jfm3 hat gesagt: // jfm3 wrote:
Somehow my PD (0.35-test3) crashes when I try any q2ladspa plugin with plugin~ 0.2.0.
Other plugins work, though.
__ __
Frank Barknecht ____ ______ ____ __ trip\ \ / /wire ______
/ __// __ /__/ __// // __ \ / / __ \ ___\
/ / / ____/ / / / // ____// /\ \ ___\____ \
/_/ /_____/ /_/ /_//_____// / \ \_____\_____
/_/ _\
Hi,
Yes. Last Sunday Stefan Kersten and I managed to get csound running to and from pd using named pipes created on the fly. It works very well and we found out that it is possible to implement the very same functionality as the csound~ object for MAX/MSP (like being able to pause performance). We implemented most of it already but had some problems with option passing to csound and we would like to make it foolproof (with filecleanup etc.). But it should be ready for posting very soon (hopefully this weekend).
Like I said, it's basically an extention of the former approach. The advantages are obvious: 1. The object doesn't use any csound code and therefor nobody has to deal with copyright issues concerning the pd side and 2. since it just calls an external csound process the object doesn't have to get recompiled when a new csound version is released. Apart from that it runs as stable as csound itself (while I encountered many problems with the csound~ object for MAX/MSP like crashing the mac with a simple pluck instrument whenever it gets called). But of course it should be made clear that the object is a rather simple hack and merely a demonstration of the advantages of linux pipes than anything else. It simply can't compare to the project of Matt Ingalls.
Yours, Orm