Subject: Re: [ecasound] Re: [linux-audio-dev] LADSPA output controller?
From: Kai Vehmanen (k_AT_eca.cx)
Date: Tue Nov 19 2002 - 06:09:03 EET
On Fri, 15 Nov 2002, Mario Lang wrote:
>> ecasound -b:128 -f:16,2,44100 \
>> -a:generator -i null -o null -el:analogueOsc,1,5,0.5,0.5 -eS:1 \
>> -a:target -i foo.wav -o alsa -el:pitchScale,1.0 -ksv:1,0.5,2,1,1 -c
> This connect an audio out to a controller in, right?
Yup.
> The backgroun is that I'd like to write a ladspa step-sequencer.
> That would output "cv" data.
Btw; one possible problem is the rate at which control parameters are
updated. For instance ecasound updates params every X sample frames, where
X is defined by -b:X. Having host determine the control rate is far from
ideal... especially as many people think that the whole concept of using a
static control rate (instead of sample-accurate parameter change events)
is a broken design (see linux-audio-dev archives for lots and lots of
discussion about this topic).
It would probably be easier to write the sequencer as a JACK client. It
doesn't really involve much extra work, and it provides you with much more
flexibility.
> I'm wondering what
> to use, audio or output controller. If i use audio,
> i guess the translation from 0.0-1.0 to hz from 30 to 15000
> will probably be not that accurate, right?
Well, the scaling is done by multiplying 32bit floats so it's actually
quite accurate.
> can i do the same with controller data too?
> i mean, connect a ladspa output controller to another
> controller port?
Unfortunately not supported at the moment, although not an impossible idea
at all. Basicly you'd just have to add a LADSPA controller source type
(-kladspa:x,y,z) that would use a LADSPA ctrl output to control some
target parameter (either native or another LADSPA parameter).
-- http://www.eca.cx Audio software for Linux!
This archive was generated by hypermail 2b28 : Tue Nov 19 2002 - 06:05:45 EET