Re: [linux-audio-dev] matrix model vs. memory

[Abramo Bagnara]

Paul Davis wrote:
> 
> >> >My proposal is to allow buffer sharing (like in my soundbox_port_link)
> >> >and to have two helper functions for not sharing case.
> >> >1) for each in_ch: out[dest[in_ch]] = in[in_ch]*gain[in_ch]
> >> >
> >> >2) for each out_ch: out[out_ch] = sum in[in_ch]*gain[in_ch,out_ch]
> >> >                              for each in_ch
> >>
> >> and please tell me where gain[] comes from?
> >
> >If you want it static: from an ordinary array.
> >
> >If you want it dynamic and presented to user: using an array exported
> >via my soundpanel API.
> >
> >What's the problem? ;-)
> 
> You've just suggested the very concept I've been arguing for. In your
> first comment about it, it was done as a plugin. you've now suggested
> that it be available for two helper functions that form the core of
> the system.
> 
> See my point? :)

But I'm not excluding the possibility to have it *also* in a plugin (a
matrix routing soundbox). Why to impose limits to fantasy? ;-)

-- 
Abramo Bagnara                       mailto:[EMAIL PROTECTED]

Opera Unica                          Phone: +39.546.656023
Via Emilia Interna, 140
48014 Castel Bolognese (RA) - Italy

ALSA project               http://www.alsa-project.org
It sounds good!