On 08/02/2007, at 14.45, Roman Haefeli wrote:
On Thu, 2007-02-08 at 14:21 +0100, Steffen wrote:
On 08/02/2007, at 13.35, Frank Barknecht wrote:
Hallo, Roman Haefeli hat gesagt: // Roman Haefeli wrote:
[...] as far as i can see it. the only case i could think of, that would require litteral '$0's as abstraction arguments, would be, if you would use dynamic patching just as a quicker way of patching.
... which actually is a very sensible usecase!
In terms of state-saving?
normally state saving is used to set variable controllers of a 'hardwired' patch/synth/hardware to a specific state, without changing the patch/synth/hardware itself. that is why i'd say, that when a dynamically created patch (more accurate: a patch, that contains dynamically created parts) is saved, this shouldn't be considered as a state-saving mechanism.
what frank and i have been talking about, is rather that dynamic creation could also be used to speed up patching progress, so that dynamic creation is used as long the patch is not finished and when
the patch is finished, no dynamic creation is involved anymore. such a scenario would require the ability to create abstractions with
litteral '$0's as arguments dynamically.
The later part i understand. But i was thinking that some clever
algorithmic scoring could use dynamical patching in which case the
saving of the patch - with the dynamically crated parts - would be
state saving of the score. I might be out deep, as I'm not sure it
makes sense in Pd.