Re: [LAU] Questions about LV2

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: J. Liles <malnourite@...>
Cc: linux-audio-user <linux-audio-user@...>, David Robillard <d@...>
Date: Wednesday, May 15, 2013 - 9:16 am

--20cf307f382c2b63f004dcbe34a1
Content-Type: text/plain; charset=ISO-8859-1

On Tue, May 14, 2013 at 1:37 PM, J. Liles wrote:

>

this is false.

> but the universal human trait of laziness means that it will always and

this is arguably true.

>

forcing IPC on the GUI is (a) stupidly expensive (b) stupidly complex (c)
limits host options.

LV2 has followed the same model as AudioUnit, in which there is documented
emphasis on NOT using any backdoor communication techniques between the DSP
and GUI, including a mention that the plugin may not even run on the same
machine as the GUI, but there is no requirement that GUIs are in separate
processes.

the correct solution here is the same one that apple used: provide the
right, easy to use, powerful communication mechanisms between the GUI and
the DSP. the facilities made possible by the AU SDK are quite remarkable
for this.

--20cf307f382c2b63f004dcbe34a1
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

On Tue, May 14, 2013 at 1:37 PM, J. Liles <malnourite@gmail.com=
> wrote:
=
It is a flaw in the design of LV2 that custom plugin GUIs both run in the h=
ost process and have access to the DSP part of the plugin's internal st=
ate. This was deemed necessary, I believe, in order to cater to vendors of =
proprietary plugins,
this is false.=A0
but the universal human trait of laziness m=
eans that it will always and constantly be used as a way to bypass any diff=
icult aspect of GUI to DSP communication.=A0
this is arguably true.=A0

Without completely removing this mechanism and forcing custo=
m plugin GUIs to run in a separate process (and therefore use a formally de=
fined interface to the DSP component) LV2 will always be inadequate for you=
r purposes.
forcing IPC on the=
GUI is (a) stupidly expensive (b) stupidly complex (c) limits host options=
.LV2 has followed the same model as AudioUnit, in=
which there is documented emphasis on NOT using any backdoor communication=
techniques between the DSP and GUI, including a mention that the plugin ma=
y not even run on the same machine as the GUI, but there is no requirement =
that GUIs are in separate processes.
the correct solution here is the same one that apple us=
ed: provide the right, easy to use, powerful communication mechanisms betwe=
en the GUI and the DSP. the facilities made possible by the AU SDK are quit=
e remarkable for this.

--20cf307f382c2b63f004dcbe34a1--

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[LAU] Questions about LV2, Julien Claassen, (Tue May 14, 9:35 am)
Re: [LAU] Questions about LV2, Jeremy Jongepier, (Tue May 14, 9:42 am)
Re: [LAU] Questions about LV2, Julien Claassen, (Tue May 14, 10:05 am)
Re: [LAU] Questions about LV2, Florian Paul Schmidt, (Tue May 14, 10:15 am)
Re: [LAU] Questions about LV2, Julien Claassen, (Tue May 14, 10:27 am)
Re: [LAU] Questions about LV2, J. Liles, (Tue May 14, 5:37 pm)
Re: [LAU] Questions about LV2, Paul Davis, (Wed May 15, 9:16 am)
Re: [LAU] Questions about LV2, J. Liles, (Wed May 15, 5:27 pm)
Re: [LAU] Questions about LV2, Julien Claassen, (Tue May 14, 5:51 pm)
Re: [LAU] Questions about LV2, hermann meyer, (Tue May 14, 9:31 pm)
Re: [LAU] Questions about LV2, Harry van Haaren, (Tue May 14, 9:57 pm)
Re: [LAU] Questions about LV2, Julien Claassen, (Tue May 14, 9:43 pm)
Re: [LAU] Questions about LV2, hermann meyer, (Tue May 14, 10:01 pm)
Re: [LAU] Questions about LV2, Harry van Haaren, (Tue May 14, 9:59 pm)
Re: [LAU] Questions about LV2, Rui Nuno Capela, (Wed May 15, 11:20 am)
Re: [LAU] Questions about LV2, Florian Paul Schmidt, (Tue May 14, 10:42 am)