Subject: Re: [alsa-devel] Re: [linux-audio-dev] laaga, round 2
From: Karl MacMillan (karlmac_AT_peabody.jhu.edu)
Date: Fri May 11 2001 - 01:57:47 EEST
On Thu, 10 May 2001 eli+@gs211.sp.cs.cmu.edu wrote:
> Paul Davis wrote:
> > The user says "I want to connect the input of this internal object to
> > internal bus #5".
>
> As to the UI: I as a user don't want to have to carry around this
> magic number 5 in my head. I just want to say "connect the input to
> the left output of that panner". I want to build a signal-flow graph,
> with support for both fan-in to inputs and fan-out from outputs.
>
> But what I really want to say is that the UI model *doesn't matter* at
> this stage. I can build that signal-flow model on top of your bus
> system. Just because I don't like magic numbers for my UI doesn't
> mean you shouldn't go out and build your engine.
>
> One-to-one: A.out to B.in.
> Allocate a bus X. A.out writes to X; B.in reads from X.
>
> Fan-out: A.out to B.in and C.in.
> Allocate a bus X. A.out to X; B.in from X; C.in from X.
>
> Fan-in: (A.out + B.out) to C.in.
> Allocate a bus X. A.out to X; B.out to X; C.in from X.
>
> Many-to-many: my layer doesn't have explicit support for this. Use a
> "passthrough" ugen.
>
> I can hide the bus IDs, and the very existence of buses, from the user.
>
Eli,
This is all a very good point - thanks for the clear statement. The only
hesitation I have is that the current api prevents some possible server
implementations that are more flexible and efficient. But, as you say, it
has no bearing on what is eventually presented to the user.
Karl
> --
> Eli Brandt | eli+@cs.cmu.edu | http://www.cs.cmu.edu/~eli/
>
_____________________________________________________
| Karl W. MacMillan |
| Computer Music Department |
| Peabody Institute of the Johns Hopkins University |
| karlmac_AT_peabody.jhu.edu |
| mambo.peabody.jhu.edu/~karlmac |
-----------------------------------------------------
This archive was generated by hypermail 2b28 : Fri May 11 2001 - 02:40:40 EEST