[linux-audio-user] ecasound/alsa question

Kai Vehmanen kai.vehmanen at wakkanet.fi
Tue Jul 13 20:40:52 EDT 2004


On Tue, 13 Jul 2004, Aaron Trumm wrote:

> i'm trying to record multiple tracks at once (different tracks into
> ecasound) and also play them back as seperate tracks sent to seperate
> outputs (as i am always trying to do)
[...]
> but i'd like to do it without jack now - possible?  I would think so -

Yes, it's with and without JACK, but much easier with JACK. I've had a
development item for making this more elegant/easier in Ecasound, but I've
now postponed the work as JACK provides a nice solution to the problem.

Anyways, here're some tips:

> ecasound -c -r -b:64 -f:16,1,48000 -a:1 -i jack_alsa,alsa_pcm:capture_1 -o track1.wav -a:2 -i jack_alsa,alsa_pcm:capture_2 -o track2.wav

ecasound -a:1,2,3,4 -f:16,4,48000 -i alsa,default \
      -a:1 -f:16,1,48000 -o mono-1.wav \
      -a:2 -f:16,1,48000 -o mono-2.wav -erc:2,1 \
      -a:3 -f:16,1,48000 -o mono-3.wav -erc:3,1 \
      -a:4 -f:16,1,48000 -o mono-4.wav -erc:4,1 

This is btw from:
  http://www.wakkanet.fi/~kaiv/ecasound/Documentation/examples.html

Alternatively you can first write all channels to one multichannel file
and later on split into separate tracks. This is probably easier if you
have _lots_ of channels.

> ecasound -c -r -a:1 -i track1.wav -o jack_alsa,alsa_pcm:playback_1 -a:2 -i track2.wav -o jack_alsa,alsa_pcm:playback_2

ecasound \
      -a:1 -f:16,1,48000 -i mono-1.wav -ea:400 \
      -a:2 -f:16,1,48000 -i mono-2.wav -erc:1,2 -eac:0,1 -ea:400 \
      -a:3 -f:16,1,48000 -i mono-3.wav -erc:1,3 -eac:0,1 -ea:400 \
      -a:4 -f:16,1,48000 -i mono-4.wav -erc:1,4 -eac:0,1 -ea:400 \
      -a:all -f:16,4,48000 -o 4ch-mix.wav

Now you might ask why the "-ea:400" (amplify by 400%) at the end of each
chain. Now I regret having this feature in Ecasound more and more each
day, but due to backwards compability issues, we just have to live with 
it. Basicly when Ecasound mixes together 'n' chains, it scales the 
amplitude to 1/n. This results in surprising results in the above case 
where each source chain has audible signal only one channel and the 
scaling doesn't make sense. Thus we scale it back up (no loss of 
information, just awkward.

To following link might help in understanding how the Ecasound chains 
work:
  http://www.wakkanet.fi/~kaiv/ecasound/Documentation/users_guide/html_uguide/index.html#htoc23

-- 
 http://www.eca.cx
 Audio software for Linux!




More information about the Linux-audio-user mailing list