[Jack-Devel] Jack Problems

[Kjetil Matheussen]

On Tue, Mar 26, 2019 at 11:46 PM Chris Caudle <chris at chriscaudle.org> wrote:
>
> On Tue, March 26, 2019 4:42 pm, Kjetil Matheussen wrote:
> > Maybe not, but [per port gain control] makes sense from a
> > do-things-quickly-and-non-frustratedly point of view.
>
> That sounds like endless amounts of frustration to me.  It is already bad
> enough that some audio interfaces change gain without the change being
> obvious, and you have to verify that the input and output levels are what
> you expect.  Apparently that very behavior unfairly engendered much
> ill-will toward jack by the original poster of this thread.
>
You should not use it if you don't need it.

> > If jack had a port_set_gain function, it would take a day
> > to script up a general mixing gui that would automatically
> > create a mixer with all the volume controls you need.
>
> Or you could connect up non-mixer in 10 minutes.  Not sure what problem
> you are attempting to solve here.

Ah, good question. At least for me, the main problem is that it's
often inconvenient to set the output volume of the clients. It's not a
big problem, but it would be easily solved by a general mixer program.
This mixer program would probably be used in other ways as well, I
don't know. Thing is that it's a natural tool one expects to be there,
but for some reason is not. It makes jack look bad. And, yes, as I,
and many others afterwards, including you, have pointed out, you can
hack this by inserting clients in between connections and so forth,
but that very inconvenient and can lead to various problems such as
breaking automatic patching and left-over clients.



>
> And as Robin pointed out it would completely break the design of jackd.
>
That's probably not correct.


> > This would  be a convenient program that most
> > people would use now and then,.
>
> I think you are attributing what you think you want to "most" people, but
> I do not see any evidence given for that.  It sounds like a terrible idea
> to me, and I am sure that "most" of the people who understand how jackd
> works and how to take advantage of that would also think it is a terrible
> idea.
>
> And I am very sure that "most" of the people who understand how jackd
> works and have the ability to modify the design think it is a bad idea, or
> someone would have done it already.
> The source code is available, if it is such a good idea code up a
> prototype and see how it works out.
>

I don't think this is so important that I would spend time
implementing it, I'm just pointing out that it's something jack is
missing. It should be a fun project though for anyone who's got some
time to spare.