Subject: Re: [ecasound] what is wrong with this?
From: Kai Vehmanen (email@example.com)
Date: Wed Jan 12 2000 - 12:57:15 EET
On Mon, 10 Jan 2000, Jeremy Hall wrote:
> -a:1,2 -i /tmp/in.raw -o /tmp/out.raw -a:1 -efl:120 -ev -a:2 -efh:120
> c-select 1
> c-select 1,2
> although you can now c-select both channels, adding the effect seems to
> only effect channel 1. I cannot apply two -ev effects because then I
Yes, ability to add effects to multiple chains (at the same time) was
removed in some early 1.6.x version. This has some API-level
advantages. But it wouldn't help much, as ecasound's design doesn't
allow chain operators to be connected to multiple chains.
> would lose on things--music and movies etc are dynamic. I ned a -ev for
> both channels, not two separate ones.
This is actually part of a bigger design issue. Hardware mixing desks
(which are emulated by many software audio packages) have two-stage
mixing - input-stripes and buses. Ecasound on the other hand only has
one stage of mixing. In most situations this is enough, and it makes
the user-interface simpler and is usually faster.
There are however a few things we can do without altering ecasound's
1) Input/output looping
ecasound -a:1 -i file1.wav -o cloop,3 -efh:120 \
-a:2 -i file2.wav -o cloop,3 -efl:400
-a:3 -i cloop,3 -o /dev/dsp -ev
2) Looping chain operators
ecasound -a:1 -i file1.wav -o null -efh:120 -elf:1 \
-a:2 -i file2.wav -o cloop,3 -efl:400 -elf:1 \
-a:3 -i cloop,3 -o /dev/dsp -elt:1 -ev
Both work in a similar fashion. Output is routed from chains 1 and 2
to chain 3. First one is easier to implement, but looping causes
latency (buffersize samples). Second one requires that chains are
processed in the right order. This is a major setback as chains
are currently fully independent (SMP systems, paraller processing).
-- Kai Vehmanen <firstname.lastname@example.org> -------- CS, University of Turku, Finland . http://www.wakkanet.fi/ecasound/ - linux audio processing . http://www.wakkanet.fi/sculpscape/ - ambient-idm-rock-... mp3/ra/wav
This archive was generated by hypermail 2a24 : Wed Jan 12 2000 - 15:32:51 EET