[LAD] automation on Linux (modular approach)

Ralf Mardorf ralf.mardorf at alice-dsl.net
Fri Mar 19 21:31:46 UTC 2010


Tim E. Real wrote:
> On March 19, 2010 07:53:23 am Ralf Mardorf wrote:
>   
>> I do not really understand what the problem is with using MIDI control
>> change for mixers. The resolution? For 2 data bytes there are 127 * 127
>> = 16129 steps. The number of channels? It's unlimited when using as much
>> IOs as needed. IMO there already is a standard for all apps, it's called
>> MIDI.
>> IMO automation is overrated, it's useful, but OTOH how often is it
>> needed to change settings during an opus? Most times a mix, selected
>> synth etc. are fixed from the start to the end of an opus. For example,
>> normally a musician plays an instrument dynamically by the touch or by
>> using a volume pedal. Dynamic for the loudness seldom is done by a fader
>> after the recording is done.
>>     
> I disagree.
> Automation, especially audio automation, is extremely important.
> Some examples:
> When recording real musicians playing real instruments, you don't just
>  record one take, you record several takes, then pick the best one and 
>  use the others to patch up the odd mistakes within the best take.
> Without automation, it is impossible to do this.
> It is easy to say "just cut and paste the corrected takes into the best take",
>  but this is not good because you cannot just place the tail end of one
>  wave up against the beginning of another wave - you will get a 'pop'.
> (Advanced apps might do the joining for you, with averaging or filtering).
> With audio automation, you do a quick, but not sudden, fade out of the
>  best take at the correction point, and simultaneously do a quick, 
>  but not sudden, fade in of the corrected take. Then you do the reverse
>  when the end of the corrected take arrives.
> This makes the transitions sound smoother.
>
> If the editing is done on the waves themselves, that's destructive
>  and when your session ends, well, you're stuck with those waves.
> (Non-daw claims to have unlimited 'go-back', but I've not tested this).
>
> You also want to be able to correct a track with a portion which is just
>  a bit too quiet or a bit too loud, without editing the waves themselves.
> I would say not altering the original waves is most important, to me.
>
> Apart from that, even simple song fade-outs are impossible without
>  automation, unless you like to do everything by hand, 'live' as the
>  song is playing.
>
> There are many cases where automation is required even in the simplest
>  of songs. If none of the linux mid/audio apps had automation, I would 
>  certainly have given up! I would not have been able to produce 
>  my songs without it.
>
> Tim.

You are right here. It's not the way I would make music, but you're 
arguments are all right. I'm not against automation, I like to have it 
myself, but we do have it already for Linux ;). I can life with what is 
possible for automation using Linux now, but I'm whining because of 
other issues ;). I had to laugh when I read that Rui plans to add 
automation to Qtractor with sample-accuracy, instead of sequencer ticks, 
it's good, but IMO overstated. For my needs there are other issues more 
important. Don't get me wrong, it's good that people wish to have 
meticulous automation, it's just less important for me, compared to some 
sync issues.

Ralf



More information about the Linux-audio-dev mailing list