Subject: Re: [linux-audio-dev] matrix model vs. memory
From: Paul Davis (pbd_AT_Op.Net)
Date: Sun May 20 2001 - 15:38:48 EEST
>Are we talking about a set of N output gains and a set of M input gains, or an
>NxM matrix of gains?
the latter.
>The latter option is probably easier to use (and now I think about it not
>much slower) but is going to be a IU nightmare.
>
>In pauls earlier memory example, something like:
>
>26 ins 26 outs
>32 ins 32 outs
>6 ins 6 outs
>
>I was just imagining this layed out on a screen. Thats 4096 knobs (no
>comments from anti gui knob people please!) which seems a bit of a
>eyeful. Maybe by colour hinting the matrix lines you could make it easy
>to use, something analogue synths have never done.
This is a good point, but I think you wouldn't present it that
way. Instead, you'd present a high-level view of all the plugins as
black boxes, then the user would select any two of them, and a matrix
for the interconnections would be displayed. It could still be rather
big, but by color-coding the matrix it could be fairly manageable, i
think.
And note that in many cases, the user can also control the gains from
much more conventional UI systems. You could still have a standard
slider/fader that just adjusted the relevant gain value in the matrix
(or some other gain value, if appropriate).
--p
This archive was generated by hypermail 2b28 : Sun May 20 2001 - 15:55:05 EEST