31 Jul
2012
31 Jul
'12
10:13 p.m.
I have adapted the GMPI requirements final draft document to a
comparison with the current state of LV2: http://lv2plug.in/gmpi.html
A couple of nonsense baroque ideas aside, most of the requirements are
met, though there are still important gaps. I mention it here in case
anyone has an interest, please feel free to address any of the points
made in this document.
It may also be useful to augment this document with additional
requirements, particularly since there's several knowledgeable folks who
may not grok the *how* of LV2 but know *what* they need in terms of
general requirements. I will add a section for this if anybody has any
input.
Perhaps this will serve as a good road map that is not too bogged down
with details.
Cheers,
-dr
31 Jul
31 Jul
10:43 p.m.
On Wed, Aug 1, 2012 at 12:13 AM, David Robillard <d(a)drobilla.net> wrote:
I have adapted the GMPI requirements final draft
document to a
comparison with the current state of LV2: http://lv2plug.in/gmpi.html
Excuse my curiosity, but whose government's permit to dig on the site
of GMPI did you have to obtain through bribing? :)
It may also be useful to augment this document with
additional
requirements, particularly since there's several knowledgeable folks who
may not grok the *how* of LV2 but know *what* they need in terms of
general requirements. I will add a section for this if anybody has any
input.
Perhaps this will serve as a good road map that is not too bogged down
with details.
I'm a bit concerned about your stance towards portability of LV2.
Audacity used to have that problem with regards to an existing
implementation of LV2, and they had to disable the code completely,
along with (quite useful, IMHO) automatic classification of plug-ins
in menu (there was another reason for the latter, though). So I'm
curious if an extra layer is really that much required.
Alexandre Prokoudine
http://libregraphicsworld.org
1 Aug
1 Aug
2:26 a.m.
On Wed, 2012-08-01 at 00:43 +0400, Alexandre Prokoudine wrote:
On Wed, Aug 1, 2012 at 12:13 AM, David Robillard
<d(a)drobilla.net> wrote:
It was a useful archaeological endeavour to perhaps dig up any useful
vestiges of a long gone civilization ;)
I have adapted the GMPI requirements final draft
document to a
comparison with the current state of LV2: http://lv2plug.in/gmpi.html
Excuse my curiosity, but whose government's permit to dig on the site
of GMPI did you have to obtain through bribing? :) I'm a bit concerned about your stance towards
portability of LV2.
LV2, the specification, is entirely portable.
Audacity used to have that problem with regards to an
existing
implementation of LV2, and they had to disable the code completely,
along with (quite useful, IMHO) automatic classification of plug-ins
in menu (there was another reason for the latter, though).
I have heard nothing about this situation. However my implementation,
lilv[1] (along with its few dependencies) is definitely portable. It is
tested and working on many flavours of POSIX including GNU/Linux,
FreeBSD, and OpenBSD, as well as Mac OS X, and Windows (natively, with
no imported POSIX layers, it even compiles with VC++, which is quite an
achievement in portability to horrifyingly terrible things).
Note this is purely host stuff. Plugins are simple C things and do not
require such libraries.
I have done a great deal of work making sure my entire LV2 stack is
portable, thoroughly well-tested, and working on any platform that
matters. I do not enjoy working with proprietary platforms, Windows
especially, but I did it anyway, so yes, I will proudly claim that all
of this stuff is entirely portable in practice, through deliberate
effort by myself. May you spread LV2 enlightenment to even the
shittiest of platform, in the hopes that the users will see the light ;)
It would be nice to establish support in hosts on foreign platforms to
resolve the chicken & egg problem. I encourage Audacity, and others, to
do so. Should there be any problems, please file a ticket[2], and they
will be resolved.
So I'm
curious if an extra layer is really that much required.
What do you mean by this? Which layer is or is not required?
Thanks,
-dr
[1] http://drobilla.net/software/lilv
[2] http://dev.drobilla.net/newticket
3 Aug
3 Aug
6:53 p.m.
On Tue, 2012-07-31 at 20:26 -0400, David Robillard wrote:
On Wed, 2012-08-01 at 00:43 +0400, Alexandre
Prokoudine wrote:
[...]
> I'm a bit concerned about your stance towards
portability of LV2.
[...]
I have heard nothing about this situation. However my
implementation,
lilv[1] (along with its few dependencies) is definitely portable.
... and I have the pictures to prove it :)
http://drobilla.net/2012/08/02/lv2-opengl-uis-on-windows/
-dr
1 Aug
1 Aug
1:21 p.m.
On 07/31/2012 10:13 PM, David Robillard wrote:
I have adapted the GMPI requirements final draft
document to a
comparison with the current state of LV2: http://lv2plug.in/gmpi.html
Great job!
Besides dynamic ports, most interesting to me seems to be the
Instruments section, http://lv2plug.in/gmpi.html#sec_3.17
In short: per voice control of any parameter.
--
Thorsten Wilms
thorwil's design for free software:
http://thorwil.wordpress.com/
8:48 p.m.
On Wed, 2012-08-01 at 09:45 +0200, Thorsten Wilms wrote:
On 07/31/2012 10:13 PM, David Robillard wrote:
I thought controlling parameters via events in general was the most
interesting, particularly since it's related to recent relevant
discussion on this list (the buffer size discussion) and an LV2 hot
topic lately for a few reasons.
Per voice parameters requires this, but is another can of worms. All
that's needed there is a convention for what kinds of events to send for
note ons and such. It would be relatively simple to build using LV2ey
atoms, though some might argue using OSC is more appropriate here since
sequencers are more likely to adopt such a standard. Of course, that's
another chicken & egg problem... since making a sequencer that would
actually do anything useful with this is no small task, this is more of
a fantasy feature IMO, plenty of things pragmatically actually needed
right now...
-dr
I have adapted the GMPI requirements final draft
document to a
comparison with the current state of LV2: http://lv2plug.in/gmpi.html
Great job!
Besides dynamic ports, most interesting to me seems to be the
Instruments section, http://lv2plug.in/gmpi.html#sec_3.17
In short: per voice control of any parameter.
4624
days inactive
4627
days old
linux-audio-dev@lists.linuxaudio.org
5 comments
3 participants
participants (3)
-
Alexandre Prokoudine -
David Robillard -
Thorsten Wilms