Subject: Re: [ecasound] loop device trouble
From: Kai Vehmanen (email@example.com)
Date: Sun Dec 03 2000 - 19:15:56 EET
On Sat, 2 Dec 2000 firstname.lastname@example.org wrote:
> ecasound -r -c -b:128 -sr:44100 \
> -a:0 -i:/dev/dsp -eli:1060,0.75,0.50,0.50 \
> -eli:1123,0.00,0.90,0.20,0.30,0.40,1.00 -o:loop,0 \
> -a:1 -i:loop,0 -o:/dev/dsp \
> -a:2 -i:loop,0 -o:out.wav
> For some reason, there are two lines assigning chain 2's input to loop
> device 0, and nothing for chain 1. If I change one of the lines to
Ok, there's one error in there...:
-a:1,2 -i:loop,0 \
-a:1 -o:/dev/dsp \
... should work ok. I admit that I made the same error in a loop-related
example I posted to list a few days ago. This inspired me to finally
add a "Tips for writing chainsetups" section to ecasound's user's guide.
Here's the first version:
1. Every chain has exactly _one_ input and _one_ output.
2. All inputs and outputs must be connected to some chain.
3. For every input/output, there must be exactly _one_
input/output specification (example specification: '-i:file.ext').
4. All mixing is based on selecting multiple chains and then
specifying an input or output (example: '-a:1,2 -i:file.ext').
5. All audio copying and mixing is done channel-wise. If you attach
a 4-channel input and a two-channel output to a chain, chain
will have 4 channels of audio, but only the first two channels
will be written to the output file.
So your chainsetup file (as well as my earlier example) broke the third
rule. Although for some tasks, the current syntax feels a bit clumsy,
I feel that it is _very_ important to have a firm logic behind the
chainsetup syntax, without any exceptions. This way if someone
posts a chainsetup file that doesn't work, it's easy to pinpoint the
cause of the problem. We check the rules, and either fix ecasound or
the chainsetup file.
-- . http://www.eca.cx ... [ audio software for linux ] /\ . . http://www.eca.cx/aivastus ... [ aivastus net radio ] /\ .
-- To unsubscribe send message 'unsubscribe' in the body of the message to <email@example.com>.
This archive was generated by hypermail 2b28 : Sun Dec 03 2000 - 18:22:24 EET