[CM] CM and Csound API
Wed, 27 Sep 2006 09:48:13 -0700
I am interested in helping in any way I can. My skills are fairly
limited, but my motivation is high. I have played extensively with all
manner of commercial software in windows, and for me it is inevitable
that I create something a bit more personal for my own use, taking the
best ideas that I have seen. You seem to be moving towards what I'm
thinking, so perhaps I should wait for you to put your code in cvs, (you
mean Csound5 cvs, yes?). Meanwhile, I'll dig in and get myself setup.
I have no knowledge of the SndObj lib, does it obliterate the need for
Csound? I really like Reaktor, how one can create synths and fx
processors with presets and call them up. I am/was hoping to create
some kind of interface for Csound that can do that, thus my interest in
Juce. Same goes for CM. I really feel the need for a gui of some
kind. Are there other gui toolkits that would be better? I know Rick
did some work with GTK2.
Keep me in the loop,
todd ingalls wrote:
> Victor Lazzarini and I have just been talking about this and trying to
> fix up the CFFI interface so it is workable. What is in csound5 now is
> just generated from verazzano but I think swig would do a better job
> and also it is clearer to me how this should work.
> i would also like to point out that i have made cffi wrapper for
> Lazzarini SndObj library and I hope to add this to the cvs for that
> soon. I also have some additional code that uses portaudio and a few
> extra classes to define realtime "synths" in lisp and schedule them
> under rts. I hope to soon add this to cvs so people can start playing
> with it, but i keep finding extra functionality i want to add as i use
> it in my own work - i need to stop that and just put the code up
> as far as Juce goes, i have SWIG code that generates a C wrapper and
> CFFI bindings for juce. The main issue is that many of the UI classes
> rely on subclassing a "model" class and a "listener" class and
> overloading viurtual methods for them to work properly. this is fine
> but the only way i have been able to determine to make these useable
> from lisp is to make subclasses of these that have constructors that
> accept function pointers that are called by the overloaded methods. on
> top of this my C++ skills are not very good so it is taking soemtime.
> if anyone is interested in any of the above please contact me off list
> and maybe we can set up a group to speed up development.