[linux-audio-dev] Additional LADSPA hints

janne halttunen jhalttun at pp.htv.fi
Thu Jan 16 11:49:01 UTC 2003


On Thu, 16 Jan 2003 10:58:40 -0500
Paul Davis <paul at linuxaudiosystems.com> wrote:

> >On Wed, 15 Jan 2003 18:13:35 +0000
> >Steve Harris <S.W.Harris at ecs.soton.ac.uk> wrote:
> >
> >> AUDIO_RATE_CONTROL. Hints than an audio control should/could be controlled
> >> by a high time res. slider or control data, but shouldn't be connected to
> >> the next audio signal by default. I can't think of any simple examples off
> >> hand, but combined with MOMENTARY it could be used for sample accurate
> >> tempo tapping.
> >> 
> >
> >Not to mention that AUDIO_RATE_CONTROL would completely useless in JACK?  
> >You'd have a buffer-full of samples and apply frequency shift to them, which w
> >ould change the length of that buffer, which you couldn't give to JACK unless 
> >you do some rebuffering.
> 
> this is true of *any* output paradigm. the audio interface only
> accepts N frames per second. if you had N frames to deliver in one
> second, then stretched them to N*2 frames, it will necessarily take
> you twice as long to deliver them - the interface isn't going to speed
> up for you.

Yes, but then, if you are also having input from realtime-type interface the buffer must be of fixed size.  

> >Correct me if I'm wrong.  We have similar issue with ecasound's Pitch Shifter 
> >operator in connection with JACK.
> 
> steve has written LADSPA pitch shifters and they work just fine with
> the fixed-sized buffer paradigm.

Yep, I was just intrigued with the tempo tapping, which cannot be achieved when in and out are realtime. (like jack-rack)  Or then I maybe misunderstood the term.


janne



More information about the Linux-audio-dev mailing list