Subject: Re: [ecasound] reverbs
From: Kai Vehmanen (email@example.com)
Date: Sat Sep 11 1999 - 15:58:50 EEST
On Fri, 10 Sep 1999, Jeremy Hall wrote:
> ecasound -i /tmp/in.raw -o /tmp/out.raw -efs:1000,500 -efr:1000,500 -c
> sounds similar to a 56K audio stream and is a bit quieter. In contrast,
Well, in this case it's just what it should sound like as...:
ecasound -i /tmp/in.raw -o /tmp/out.raw -efr:1000,500 -c
does nothing to the signal in 1.5.9/10.
> as noted before, -ev can produce some strange results, do we wrap
Hmm, strange in what way? Counters are reseted with every
call of "estatus". Otherwise they will wrap after about 27 hours
of sample data... this shouldn't be a problem. ;)
> The statistics table is hard to understand.
Well, this I admit. :) It's _very_ messy. I needed a quick&dirty
way to adjust the input signal level (in console mode!).
As ecasound wasn't publically released back then "-ev" did
the job nicely.
> enm: I think I am not setting it up right.
Check /usr/local/share/ecasound/singlechain_effect_presets ...
There should be at least one working noise-gate preset.
Btw; you can get some weird rhytmic effects if you use
low-frequency oscillators to control the noisegate parameters.
You can't do this with your average rack-multieffect! ;)
> (audiofx) Mono noise gate enabled; threshold level 75.0000%, threshold
> time 40.0000 ms, attack 80.0000 ms, hold 75.0000 ms and release 75.0000
This won't sound so good as it treats everything below 75%
(100% = max amplitude) as noise.
> -ep: works as advertised, but decreases volume
Yep, this should have been fixed a long time ago. In 1.5.11
"-epp" won't affect the volume level.
> [jhall@pongo jhall]$ ecasound -i /tmp/in.raw -eft:80
> - [ Chainsetup/Adding a new chain operator ]
These will be gone in 1.5.11r5. I'm rewriting some parts of
the effect-API and it will have a new way to handle
effect configurations (internally). When ecasound was just a
simple command line util, it wasn't a big deal if it dumped
core when invalid command line arguments were given, but now this
will have to change.
> -etf: doesn't make a signifficant diff on a stereo signal and that is whta
> I am testing with now
Same problem as with "-efr" (bandreject filter). Doesn't do
anything (1.5.0->1.5.10 versions).
> -ef1: produces a high squeal
This is under work (a bit experimental, -efs and -efb should
be used instead). I might drop this altogether.
> -efb: produces strange results no matter what its options are
Hmm, this works fine with me. I'll have to test it some more.
> -efh and -efl
> they seem to do something that is sortta right, but I am hearing pitches
> around 400 to 500hz when I set -efh of 1000 but they may be overtones I'll
> test with a sine wave later.
Yes, these should also work, but are not perfect.
> gotta go, will continue the evaluation later, but a friend is here.
Thanks! This has been __very__ helpful. Most of ecasound's effect
code dates back to 1997-1998. I've been aware of the problems,
but other issues (multitracking, qtecasound, etc) have taken
all my time.
-- 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 : Sat Sep 25 1999 - 19:41:11 EEST