Subject: Re: [linux-audio-dev] "pro" soundfile editors for linux
From: Kai Vehmanen (kaiv_AT_wakkanet.fi)
Date: to helmi 10 2000 - 17:38:05 EST
On Tue, 8 Feb 2000, Paul Barton-Davis wrote:
> Absolutely. But its the level required in a commercial studio, whether
> we like it or not.
True, although I still like to believe that I'm developing audio
software for my own purposes. ;) Motivation and interest are
extremely important, and even more so with free software.
> I think you need to try working with a 26 piece percussion ensemble,
> all of whom were out till 3am the night before, packed into a room too
> small to get them all in without acoustic bleed through, various
> personal tensions in place, and a lot of time pressure to get the
I see your point. ;)
>> a sequencer (MIDI, tracker, audio, ...) and then export the data
>> to the multitrack environment.
> Right. So the question is, can one do this sort of thing without going
> all the way to ProTools, or not ?
Well, it can be done, but isn't easy. Currently it's really hard
to route audio data between applications (in realtime that is).
ALSA and OSS both have loop devices, but routing from soundcard to
another just isn't enough. We need something more flexible...(= MuCoS).
>> Hmm, I'd record the punched material into a separate file (hey,
>> diskspace is cheap). Replacing previously recorded material makes
> OK, I'll buy that. But now, without an edit/play list, how do you
> propose to play it back correctly ? The material is no longer in a
> single contiguous file, and I thought that this was a requirement for
> ecasound ?
One of ecasound's strongest points is its controller subsystem.
You can control all effect parameters (even other controllers)
with various control sources (envelopes, oscillators, MIDI-knobs,
etc). To do a cross-fade between two tracks, you need two amplify
effects and linear envelopes controlling them. Of course, a good
GUI does all this automatically.
> As a point of reference, I asked some studio people how many punch
> in/out's they might typically do in the course of a 40 minute ADAT
> tape. The answer seems to vary a lot depending on the artists being
> recorded, but the order of magnitude was 10-100. The Alesis M30 allows
Here's something I try to avoid. Most audio apps imitate hardware
designs (we have virtual 303s, mixers, tape recorders, etc). This
approach has its advantages (familiar GUI, some concepts really
work "virtually"), but I'm more interested in trying out alternative
solutions. For instance, punch-in/outs aren't really necessary
when you can use as many tracks as you like. Similarly, without
hardware limitations it's easier to switch between different "modes"
(separate mixdown view, audio sequencing, separate mixer-views
for subgroups, etc etc). Earlier we discussed about using mixer
"stripes and buses" approach in audio apps.
-- Kai Vehmanen <kaiv_AT_wakkanet.fi> -------- CS, University of Turku, Finland . http://www.wakkanet.fi/ecasound/ - linux multitrack audio processing . http://www.wakkanet.fi/sculpscape/ - ambient-idm-rock-... mp3/ra/wav
This archive was generated by hypermail 2b28 : pe maalis 10 2000 - 07:23:27 EST