[specimen] save-master-volume bug

[Antonio "Willy" Malara]

2007/3/13, Atte André Jensen <atte.jensen at gmail.com>:
> Well I looked into the source (+ a .beef) and it's more like it's not
> implemented than it's a bug. I don't see any attempts anywhere in the
> code to either save or restore the master volume (for instance the
> slider is created directly using the #DEFINEd DEFAULT_VOLUME). Also the
> .beef doesn't seem to know about it, it only has the level of each sample.

yeah, the part of specimen that deals with parameters and CCs is broken down
in various parts of the codebase, for example in saving/loading the
beef file there is
a long list of istruction that programmatically sets all the
parameters one-by-one,
and in the midi-section too, there is a static map defined at
compile-time of mapping
between parameters and CC number, and scale of vaules...

now for adding a parameter (like the master volume) one should add it
both in the beef handling (twice: one for loading one for saving) and
one in the midi handling code (twice: one for alsa midi, one for jack
midi), due to the trend of osc in linux audio community, if one day
specimen would implement osc there will be code even in that part...

i was thinking maybe it's better to introduce an abstraction layer in
the code to
abstract the concept of "parameter" of specimen, so one can simplify
the code that
deals with the beef file, and we get even a much more dynamic binding
between midi or osc devices and specimen itself?

i never talked of this because i've no time to implement this idea,
but i think it can be useful to specimen :/

of course for little things one can just take the "patch-as-we-go"
approach, but i fear that this attitude in the long term can hurt the
codebase, what are your toughts, as developers?

-- 
Willy
------
Il jazz e' lasciare che la luce brilli, lasciarla essere.
                -- Keith Jarrett

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.