[LAD] automation on Linux (modular approach)

Ralf Mardorf ralf.mardorf at alice-dsl.net
Mon Mar 22 17:53:35 UTC 2010


Nick Copeland wrote:
> > > Just 128 steps for filter cutoff for an oscillating emphasis does not
> > > work well
> > > if it needs to be tuned exactly to an VCO/DCO, it is basically 
> never in tune
> >
> > This is wrong.
> >
> > The number of bits in a parameter is unrelated to where and what some
> > filter is or is not tuned to.
> >
> Why not?
>
> Perhaps we are referring to different things. I was talking about the 
> number of
> steps in a filter cutoff parameter - if you only have 7 bits in the 
> filter cutoff setting
> then you can only select 127 discrete cutoff frequencies. If you 
> overdrive the filter
> emphasis then it should turn into something close to a sinewave 
> oscillator at the
> frequency of the cutoff - the steps in the frequencies become audible 
> which was a
> ongoing complaint against these systems. The issue is that if you want 
> to detune
> the filter oscillating frequencies by a 0.1 HZ to get some phasing 
> then you are out
> of luck since the cutoff steps are too big. The same is true if you 
> want to use this
> filter as an extra oscillator (something that a lot of people did on 
> the big synths to
> get hammond sounds) then it doesn't tune that well,especially at the 
> higher
> frequencies where the filter was used as if it were another drawbar.
>
> It doesn't mean the synth will not sound good, there are just some 
> things that it
> will not be able to do.
>
> Regards, nick.

Indeed a B3 emulated by an Oberheim Matrix-1000 might be no good choice, 
OTOH because of the automation, you won't have any synth with a good B3 
emulation to change 20 parameters during a song, while for any 
"synthetic" sound e.g. from an Oberheim Matrix-1000 you might wish to 
change a lot of parameters during a song and changing filter parameters 
for "synthetic" sounds 128 steps are enough.

Ralf



More information about the Linux-audio-dev mailing list