Subject: Re: [linux-audio-dev] Extending LADSPA to support Non-Causal Plugins
From: Conrad Parker (conradp_AT_cse.unsw.EDU.AU)
Date: Mon Oct 30 2000 - 13:50:47 EET
On Mon, Oct 30, 2000 at 11:52:59AM +0200, Jarno Seppanen wrote:
> "Richard W.E. Furse" <richard_AT_muse.demon.co.uk> writes:
>
> > Non-causal is a signal processing term, but I do agree it isn't terribly
> > intuitive or especially well-defined. I'm very open to suggestions!
>
> IMHO causality is a well-defined and very nonambiguous property -- if you do
> signal processing that is. From Oppenheim, Schafer, Discrete-Time Signal
> Processing, 1989, p. 20: "a system is causal if for every choice of $n_0$ the
> output sequence value at index $n = n_0$ depends only on the input sequence
> values for $n \leq n_0$." ... "That is, the system is nonanticipative."
>
> For example, according to the definition, the reverse operation is not
> causal, since there exist output values at $n = n_0$ that depend on input
> values at $n > n_0$.
>
> However, not being a native speaker I don't know of the connotations of the word.
sounds good to me: causalilty is all about things not being
influenced by stuff that hasn't happened yet. Let's not get
philosophical though ;)
as a general principle for this kind of thing in LADSPA, I think
we're best off defining things in terms that signal processing
people [and musicians] will understand, given that the vast
majority of users of the API *should be* people writing plugins,
not people writing hosts -- so it would be nice to make it Simple
for them. [says he who most recently wrote yet another LADSPA
host ;) ]
given the confusion about this term, if this stuff goes in then the
header should probably mention the definition from Oppenheim et al.,
which makes the meaning quite clear.
Conrad.
This archive was generated by hypermail 2b28 : Mon Oct 30 2000 - 14:25:39 EET