[LAD] starting out with tux audio
noisesmith at gmail.com
Wed Jun 18 04:53:58 UTC 2008
On Tue, Jun 17, 2008 at 7:39 PM, AlgoMantra <algomantra at gmail.com> wrote:
> [[[again, meant for the list. Dunno why Gmail is defaulting weirdly....]]]
> On 6/18/08, AlgoMantra <algomantra at gmail.com> wrote:
>>> Do you have any midi-keyboard around? Because stdio won't tell you that
>>> a key has been released, only that it has been hit.
>> Interesting!!! It'll be fun to find a way around this.
>>> This limits the "toy
>>> Casio experience" somewhat ... There is of course a workaround - given
>>> that you have root permission, you could read the keyboard
>>> from /dev/input/ instead. This is unfortunately a bit like having a
>>> "keyboard sniffer" installed, which may not be ideal in your
>>> environment. Next step up the ladder I think is fullblown X ...
>>> A virtual midi-keyboard would solve that, perhaps vkeybd?
>> Just for myself, I would love to play with a lot of different external
>> HIDs actually..I'm planning to code for playing C code music with
>> one of those old Nintendo joysticks modded for USB. It's quite
>> cheap here (around $8 USD).
>> But the class I teach coding barely knows how to TYPE!!!! Grrrr....
>>> ------- -.-
>>> 1/f ))) --.
>>> ------- ...
> ------- -.-
> 1/f ))) --.
> ------- ...
> Linux-audio-dev mailing list
> Linux-audio-dev at lists.linuxaudio.org
On the subject of input devices, I had a recent project that may be
just up your alley. First, I changed my xorg.conf to only read from
one specific mouse device, then I used the input event interface that
the kernel creates for each extra mouse, and wrote a c program that
mapped five mice (as many as I had laying around and could plug into
one machine) to seperate synthesis parameters via OSC, up to 7
parameters for each standard mouse if you ignore chords: 1) leftbutton
x 2) leftbutton y 3) rightbutton x 4) rightbutton y 5) scroll up/down
6) scrolldrag x 7) scrolldrag y (the number of possible parameters
with chording comes out to a much higher number, of course) (the
advantage of OSC here is that I write the code once and then it can be
an input to any synthesis setup I would actually care to use).
And on that subject: I understand the appeal of the simplicity of
starting from scratch, but if you ever get all the features you
expect, it is rarely simple any more. If you use jack for your audio,
or write your synth/filter as a ladspa, LV2 or dssi plugin, and use
OSC or MIDI for control input / output, you may just end up with
something someone else would use, because it will be simple to
integrate into someone's computer studio. Programs that use /dev/dsp
and read from the tty can't play together so easily, and are in the
long run difficult to compose and perform with (I no longer consider
any of them worth my time). Selfishly, I lament the time others spend
working on such projects that do so much less to benefit me than they
could. You of course don't owe me anything, so this is a request
rather than a demand.
More information about the Linux-audio-dev