OSS apps vs. ALSA apps

[raydar]

> On Sunday 24 February 2008, raydar wrote:
>   
>> I think I spoke too soon if I thought alsa-oss would be the silver
>> bullet for Gtick + Jack compatibility: I installed the alsa-oss package,
>> but even after a reboot, nothing has changed; Gtick still gives me the
>> "Couldn't start metronome. Please check if specified sound device and
>> sample file are accessible" error if I try to start it ticking after
>> Jack is running.
>>     
>
> I would expect this is the usual problem with hardware that won't accept 
> multiple connections (which is so common it's definitely the rule by a wide 
> margin, and not the exception.)  When JACK is running, it eats up the one 
> connection, and blocks the soundcard.  The usual old school way to deal with 
> this is to make sure every audio app you use can speak JACK, or run one or 
> the other.  (I think there used to be a little wrapper so you could run 
> jack_hack_wrapper_thingie my_old_dumb_app and get it to work that way, but if 
> I'm not dreaming that, I haven't been able to figure out how to do that in 
> years.)
>
> That looks like the culprit to me, after a look at the Gtick page.  I used to 
> prefer Sweep as my wave editor of choice, and it's an old 
> only_one_method_of_audio_production app that doesn't speak JACK either.  I 
> used to turn JACK off to use Sweep.  Now I use MHWaveEdit or reZound instead, 
> which speak JACK just fine.
>
> There is some new school way of dealing with this issue using something that I 
> think they call the "dmix plug layer" or something to that effect.  I must 
> confess that when I tried to read through the information on how to get this 
> to work with my ice1712, I got frustrated quickly, and gave up in total 
> despair.  I can't make heads or tails of all that blather about plug layers 
> and asoundrcs.  I've been using ALSA since 0.5.x without ever having to screw 
> with an asoundrc file, and I like it that way.  When I did try to do this, 
> the documentation I found was wretched, and assumed I already had a PhD in 
> audio engineering or something.  Just blather your boodles to your 
> scluppthiths and then scalate your eschillitons to the blintzfluffles, and 
> you should immediately grath your plarkitty splimmles.  Isn't that totally 
> obvious, stupid?
>
> Now the world sees what a piss poor expert I am.  I probably totally omitted 
> the Magic Sploofloodle everyone has known about and used since 2003.
>
> Oh well, this message is worth every penny paid for it.
>   
Oh, I dunno about that--I'm learning something! :)  

I discovered in the Ubuntu forums that running an app as an argument to 
"aoss" is a/the way to invoke alsa-oss, so I typed

    aoss gtick

in a terminal, got

    /dev/dsp: Device or resource busy

and also tried having Gtick use /dev/dsp1 after launching Gtick via 
aoss, but there was no difference from starting Gtick the normal way & 
using /dev/dsp1 as before (no error, but no output).

So, if the problem is Jack hogging /dev/dsp, and if /dev/dsp1 is not 
being hogged but it's output isn't "getting through," is there a way I 
can route /dev/dsp1's output to Jack's input so that it all goes out 
/dev/dsp?  (I'm thinking /dev/dsp1 might be the on-board sound that I'm 
not using, while /dev/dsp is the PCI sound card which I am using and 
have speakers & inputs plugged into. Maybe that onboard sound chip 
wasn't an involuntary waste of $ after all?)

Of course, developing a tic[k] may not be in my best interests anyway . 
. . .  :)