On Thu, Sep 27, 2007 at 12:34:08PM +0200, Tim Blechmann wrote: > hi enrique, > > > i'm using pd since 0.37 and my focus is mainly on dynamic patching (gui > > & dsp). i often create instances of abstractions for example to add one > > or more effects to a channel. but that is also where i reach puredata's > > limit. so i had to get used to dropouts :-( > > > > so i wonder will dynamic patching / modifying the dsp-tree be possible > > in nova ? > > currently, there is no way to do dynamic patching in nova, mainly > because i haven't thought of a powerful way to do it. so on the side of > the language, patching features are still missing. > > i was thinking about using a messagebus like "thiscanvas" or > "thispatcher" or a |thispatcher| object in order to send messages to the > canvas, which could be used for dynamic patching, but i haven't found a > good solution, yet. the problems are mainly: > - how do you get the message to the canvas (message bus or special > object?) > - how do you reference objects in a patch (pd uses integer indexes, > which are very fragile, dyn~ uses symbolic names, which it maintains > itself, nova internally uses uuids, but i doubt you would want to use > them for dynamic patching) > > dynamically modifying the dsp graph is not a problem in nova, the > generation of the dsp chain is done in a background thread.
For the problem of referencing objects in a patch, I would consider user assigned names for each object. This would not only be useful for dynamic patching, but a keyboard driven interface as well. It also contributes to the patch readability since the name gives some clue towards the function of the thing. Naming an object could be optional; not doing so just means it can't be dynamically manipulated or directly accessed with the keyboard. _______________________________________________ nova-dev mailing list [email protected] http://klingt.org/cgi-bin/mailman/listinfo/nova-dev http://tim.klingt.org/nova
