On Tue, Sep 2, 2014 at 3:57 PM, Philipp Überbacher <murks@tuxfamily.org> wrote:

On Tue, 2 Sep 2014 23:22:56 +0100

Harry van Haaren <harryhaaren@gmail.com> wrote:

> On Tue, Sep 2, 2014 at 11:06 PM, J. Liles <malnourite@gmail.com>
> wrote:
> > In my workflow, this presents no problem. If I implemented this JACK
> > configuration feature, then I would have CREATED a problem where
> > there were none before.
>
> > Hopefully you can see why I'm not interested in implementing this
> > feature.
> Yep. Good point, it seemed like something that would be nice to
> handle in NSM, but i now agree that NSM should *not* handle jack
> settings. The user has to do this manually: its not feasible to
> script / auto-save JACK settings reliably.
>
> So before starting the NSM UI / nsmd, it might be worth checking if
> JACK is running, and if its not, providing an option to ask
> A) Do you want to start JACK a la ~/.jackdrc
> B) Continue without JACK
> C) Quit
>
> Does that seem reasonable? I'm thinking about beginner users, who are
> not familiar with the eco-system, and making this as usable as
> possible.
>
> Cheers, -Harry

Well, I still think it is a good idea. It would not create problems for
you, simply because you're not forced to use this if it is realised as
a client in the spirit of jackconnect.

I did not know that nsm supports live switching of sessions without
restarting clients, but in case you need to change the jack sample rate
you most likely can't do that anyway. I guess that with that
functionality in use 'live switching' would not be possible or
only in special cases (identical jack settings).
That would be the only drawback I can see.

Theoretically there's no conflict between sample rate changes and switching--assuming of course that A) all of the clients involved can deal with runtime sample rate changes and B) that none of them will complain if the JACK sample rate and the project sample rate differ (however briefly).

And likewise, theoretically it's fine to change sample rate and buffer size without stopping JACK clients.

In fact, I do test all the Non stuff for this scenario. But I don't usually do that in practice though. Too many bad memories from the time when running jack_bufsize with clients attached would crash stuff.

Regards,
Philipp
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user