21 Jan
2008
21 Jan
'08
5:08 p.m.
Hi all,
the a bit provocative title is not here to start a flame war but to spark a
constructive discussion about the
viability and future of the LV2 plugin standard in the professional audio
application market.
Some background:
as you probably know Steinberg just released VST3 and developers do not seem
happy with it
as it is not backwards compatible, not many new features and it seems less
portable amongst platforms than VST2.4.
Users are unhappy and started a long discussion:
http://www.kvraudio.com/forum/viewtopic.php?t=204080&postdays=0&pos…
The discussion was picked up on the Reaper forum too.
For anyone not familiar with Reaper, it is a very good audio,midi sequencer.
http://www.reaper.fm
It is a windows app but runs very well under wine and the company writes
this on their page.
Users like Alex Stone are using it on wine in conjunction with LinuxSampler
The authors are Justin Frankel from Nullsoft and others. For those that
don't know the name, he is the one
that wrote winamp, gnutella, nullsoft installer etc.
On a forum he said he played with the idea of open sourcing reaper.
It is being ported to OS X and probably will get ported to Linux too, given
the very good performance it achieves on wine.
Now back to LV2:
The VST3 discussion on the reaper forum resulted in users proposing to
create a new plugin standard
in order to "break free" from proprietary standards so they are proposing to
add LV2 support to Reaper.
Justin from Reaper answered the following on the forum:
-------
I looked at LV2, there's a lot of stuff which I disliked.. for example,
"ports" being for parameters and audio buffers (and presumably MIDI events),
and all having the possibility of colliding, isnt well thought out.
Also if you want to add parameters to a new revision of a plug-in, then you
have to change the URI? ick.
Or what if you want to change the I/O of a plug-in on the fly..
-Justin
---
see here for the full thread:
http://www.cockos.com/forum/showthread.php?t=17198&page=2
I did not look at LV2 so I cannot judge, but I think LV2 developers should
discuss about these issues and concerns with
Justin in order sort out problems, and given the joung nature of LV2 , in
case important design flaws get uncovered, change the specs a bit.
Reaper is rapidly building up a large user base (users switching away from
cakewalk sonar and cubase to reaper) as the application provided excellent
performance, is easy to use
and new features are being incorporated at a fast pace.
Dave Philips seem to love the app too as he mentioned it in Linux Journal
and he posts frequently on the forum.
http://www.linuxjournal.com/node/1005911
I think LV2 and Reaper developers should join forces because together
perhaps it will be possible to impose a new open plugin standard
which will get adopted by other commercial applications too and supersede
VST2.4 over time.
not sure if the reaper devs are reading LAD, so perhaps LV2 developers
should answer on the reaper forum too in order to sort out the
issues raised by Justin.
Everyone is invited to add his own point of view and I hope that the outcome
will be a positive collaboration between LV2 and Reaper.
thanks everyone,
Benno
21 Jan
21 Jan
5:25 p.m.
New subject: LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
On Mon, 2008-01-21 at 22:08 +0100, Benno Senoner wrote:
On a forum he said he played with the idea of open
sourcing reaper.
If he'll actually make it free software, and it's interesting enough to
someone who knows how to code, LV2 support will probably be added to it
anyway, or to a fork.
--ll
5:47 p.m.
Lars,
Yes I agree with what you said,
but better to start the discussion with them now because
we will not know if and when if will become free software.
I think times are mature for a change and we should all join forces and see
if we can
impose a standard that please both open source developers and commercial
application developers too.
Since you are one of the LV2 developers I'd suggest you and all other LV2
devs to take part on that thread on the reaper forum
and enlinghten them about the advantages and flexibility of LV2.
I think both parties LV2 and Reaper devs have nothing to lose, the worst
that can happen is Reaper not integrating LV2, but
OTOH if they point out design problems you can use them to improve the LV2
specs as the project is still young.
It would be really a plus for LV2 if reaper started supporting it.
If reaper's development continue at the current fast pace it could be that
it will soon become one of the most
popular audio sequencers and having a large userbase always helps to get
lots of feedback.
After all if you created LV2 I guess the goal is to achieve widespread
adoption so having widely used apps
using the standard can only be positive.
thanks everyone,
Benno
2008/1/21, Lars Luthman <lars.luthman(a)gmail.com>om>:
On Mon, 2008-01-21 at 22:08 +0100, Benno Senoner wrote:
On a forum he said he played with the idea of
open sourcing reaper.
If he'll actually make it free software, and it's interesting enough to
someone who knows how to code, LV2 support will probably be added to it
anyway, or to a fork.
--ll
_______________________________________________
Linux-audio-dev mailing list
Linux-audio-dev(a)lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev
5:59 p.m.
New subject: LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
On Mon, 2008-01-21 at 22:47 +0100, Benno Senoner wrote:
Yes I agree with what you said,
but better to start the discussion with them now because
we will not know if and when if will become free software.
Until it actually becomes free software I'm not interested in it at all.
--ll
7:18 p.m.
New subject: [LAD] LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
Am Montag, 21. Januar 2008 21:58:24 schrieb Lars Luthman:
On Mon, 2008-01-21 at 22:47 +0100, Benno Senoner
wrote:
Good you let us know, but this is not about Reaper, it's about LV2 and whether
it could be improved.
BTW there were rumours Reaper might become open source one day. But as you
might imagine, such a process would presumably happen gently step by step, so
that kind of attitude could really destroy the whole process.
CU
Christian
Yes I agree with what you said,
but better to start the discussion with them now because
we will not know if and when if will become free software.
Until it actually becomes free software I'm not interested in it at all.
7:32 p.m.
New subject: LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
On Tue, 2008-01-22 at 00:18 +0000, Christian Schoenebeck wrote:
Am Montag, 21. Januar 2008 21:58:24 schrieb Lars
Luthman:
The mail i replied to was about Reaper.
But I really doubt that the authors of a piece of software that I
haven't even used would care about my attitude, or lack thereof, when
deciding what to do with their code.
--ll
On Mon, 2008-01-21 at 22:47 +0100, Benno Senoner
wrote:
Good you let us know, but this is not about Reaper, it's about LV2 and whether
it could be improved.
BTW there were rumours Reaper might become open source one day. But as you
might imagine, such a process would presumably happen gently step by step, so
that kind of attitude could really destroy the whole process. Yes I agree with what you said,
but better to start the discussion with them now because
we will not know if and when if will become free software.
Until it actually becomes free software I'm not interested in it at all.
7:48 p.m.
New subject: [LAD] LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
Am Montag, 21. Januar 2008 23:31:25 schrieb Lars Luthman:
But I really doubt that the authors of a piece of
software that I
haven't even used would care about my attitude, or lack thereof, when
deciding what to do with their code.
Of course not. But it might affect their decision when this is the general
attitude. And unfortunately this actually the case here as far as I can see
it at the moment.
CU
Christian
10:33 p.m.
New subject: LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
On Mon, 2008-01-21 at 22:47 +0100, Benno Senoner wrote:
Since you are one of the LV2 developers I'd
suggest you and all other
LV2 devs to take part on that thread on the reaper forum
and enlinghten them about the advantages and flexibility of LV2.
The Reaper forum isn't a useful place for development discussion.
Here, or http://lists.lv2plug.in/listinfo.cgi/devel-lv2plug.in
but better to start the discussion with them now
because
we will not know if and when if will become free software
Yeah, right..
Tell you what, if Reaper becomes fully and truly Free Software (by the
FSF definition), I hereby volunteer to implement LV2 support for
them /and/ write good LV2 extensions for any features present in
whatever version of VST they happen to favour. Ball's in their court.
You can't beg a mechanic to fix your buddy's car when he's welded the
hood shut - and plans to run over the mechanic's bike as soon as it
runs ;)
Until then, my bike is Ardour. Race ya.
-DR-
6:21 p.m.
New subject: LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
On Mon, 2008-01-21 at 22:08 +0100, Benno Senoner wrote:
Justin from Reaper answered the following on the
forum:
-------
I looked at LV2, there's a lot of stuff which I disliked.. for
example, "ports" being for parameters and audio buffers (and
presumably MIDI events), and all having the possibility of colliding,
isnt well thought out.
In short this doesn't make any sense. Colliding? What?
Also if you want to add parameters to a new revision
of a plug-in,
then you have to change the URI? ick.
This is for compatibility. To be more accurate, if you add ports to a
plugin which /must/ be used, then you need to change the URI, otherwise
upgrading a plugin would break patches (think about it, it's obvious).
There are trivial solutions to future "versions" of a plugin that are
incompatible as well, noone's defined it yet because noone cares yet
(this is the whole point of all that RDF stuff...).
In short: updating your plugin collection will NEVER break existing
patches, if things follow the rules. This sounds like a complaint about
LV2 being well thought out, not the other way around...
Or what if you want to change the I/O of a plug-in on
the fly..
The solution to this is "multi-ports", ie ports with several streams in
one. You'll almost certainly see this show up at some point (there is
an existing extension that sort of does this for "parameters", but it
needs to be generalised).
LV2 has been thought out an awful lot more than any other plugin spec.
Any concerns Justin or anyone else has can be resolved without having to
break the spec, because it's extensible (chances are somebody somewhere
already has a good idea of how to do it). A lot of people have put a
great deal of thought into making sure this is true.
Can you say that about VST? VST3? AU? Any of the others? No, you
can't. I can say with almost complete confidence that anything VST3 or
whatever can do, LV2 can do too (with the appropriate extensions). LV2
can do pretty much anything.
What isn't well thought out again?
I think LV2 and Reaper developers should join forces because together
perhaps it will be possible to impose a new open plugin standard
which will get adopted by other commercial applications too and
supersede VST2.4 over time.
[ Obligatory sidenote: I don't think anyone currently working on LV2
things gives a damn about proprietary software. I for one put a lot of
effort into LV2 to further free audio software, not the proprietary
competition to free audio software. Sure, Reaper 'joining forces' with
the l-a-d community would be great. Show us the code, then we'll talk;
until then, Reaper is the enemy as far as I'm concerned ;) ]
That said, if the Reaper developers look more closely at LV2 they will
see it can be whatever they want it to be... You want flexible IO?
figure out how, and just do it! You want an elegant scheme for
upgrading plugins that break compatibilty? figure out how, and just do
it!
See the pattern?
Cheers,
-DR-
P.S. there is now an LV2 development mailing list preferred to l-a-d for
LV2 specific stuff, everyone with an interest in LV2 development hop on
board: http://lists.lv2plug.in/listinfo.cgi/devel-lv2plug.in
22 Jan
22 Jan
6:02 a.m.
Benno Senoner wrote:
Everyone is invited to add his own point of view and I
hope that the
outcome will be a positive collaboration between LV2 and Reaper.
Well, Benno, what
do you think about the responses so far? Is the spirit
of open collaboration and the welcoming attitude alive & well here ?
If I've learned one thing here it's that open sources != open minds.
Best,
dp
6:21 a.m.
New subject: LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
On Tue, 2008-01-22 at 05:02 -0500, Dave Phillips wrote:
Benno Senoner wrote:
Dave just offered to do all the work for them - if that's not in the
spirit of collaboration I don't know what is.
If I've come across as a grumpy old bastard I apologise, because that
was not my intention. The thing is that I'm just not particularly
interested in software that can only be used in the way the original
author allows it to be used. If they actually make it free it's a
completely different issue.
--ll
Everyone is invited to add his own point of view
and I hope that the
outcome will be a positive collaboration between LV2 and Reaper.
Well, Benno, what
do you think about the responses so far? Is the spirit
of open collaboration and the welcoming attitude alive & well here ?
If I've learned one thing here it's that open sources != open minds.
7:45 a.m.
New subject: LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
Lars Luthman wrote:
The Reaper forum isn't a useful place for development discussion.
Here, or http://lists.lv2plug.in/listinfo.cgi/devel-lv2plug.in
Yeah, right..
Tell you what, if Reaper becomes fully and truly Free Software (by the
FSF definition), I hereby volunteer to implement LV2 support for
them /and/ write good LV2 extensions for any features present in
whatever version of VST they happen to favour. Ball's in their court.
You can't beg a mechanic to fix your buddy's car when he's welded the
hood shut - and plans to run over the mechanic's bike as soon as it
runs ;)
Until then, my bike is Ardour. Race ya.
-DR-
So responding with sarcasm, "We'll do it my way or not at all"
conditions, and a confrontive attitude qualifies as "the spirit of
collaboration" ? Geez, you guys are really winning me over.
On Tue, 2008-01-22 at 05:02 -0500, Dave Phillips
wrote:
I quote the entire response:
On Mon, 2008-01-21 at 22:47 +0100, Benno Senoner wrote:
Benno Senoner wrote:
Dave just offered to do all the work for them - if that's not in the
spirit of collaboration I don't know what is.
Everyone is invited to add his own point of view
and I hope that the
outcome will be a positive collaboration between LV2 and Reaper.
Well, Benno, what do you think about the responses so far? Is the spirit
of open collaboration and the welcoming attitude alive & well here ?
If I've learned one thing here it's that open sources != open minds.
Since you are
one of the LV2 developers I'd suggest you and all other
LV2 devs to take part on that thread on the reaper forum
and enlinghten them about the advantages and flexibility of LV2.
but better to
start the discussion with them now because
we will not know if and when if will become free software
If I've come across as a grumpy old bastard I
apologise, because that
was not my intention. The thing is that I'm just not particularly
interested in software that can only be used in the way the original
author allows it to be used. If they actually make it free it's a
completely different issue.
No fear, no-one's going to ask you or DR for help
in this matter.
Best,
dp
10:16 a.m.
New subject: LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
Hallo,
Lars Luthman hat gesagt: // Lars Luthman wrote:
If I've come across as a grumpy old bastard I
apologise, because that
was not my intention. The thing is that I'm just not particularly
interested in software that can only be used in the way the original
author allows it to be used. If they actually make it free it's a
completely different issue.
I think this is not about Reaper being free or not or about Reaper
being in need of a fix or not. The issue is: Someone on the Reaper
forum has criticized LV2. If that criticism is valid or not, I cannot
judge, but it is not made invalid *just* because it came from a
developer of non-free software.
Ciao
--
Frank Barknecht _ ______footils.org__
1:25 p.m.
I thought we were all on the same side?
I hate politics, and this email was not the first one I wanted to send
to this list since rejoining. The fact is, though, that I wanted to
leave the Windows world behind and go all-Linux for my music making, and
I am now regretting that decision. What have I come back to?
* ALSA refuses to document anything, so it may as well be closed-source.
A friend of mine, who coded the original Windows drivers for a certain
multi-channel interface, wanted to write ALSA drivers for it, and got
stopped by no documentation and uncooperative developers.
* Ardour looks, behaves, and has almost the exact same feature set as it
did two years ago. (Nobody in the commercial world is still using a
32-bit mixing engine, not even Pro-Tools.)
You can go ahead and tell me to leave if I don't like it, but the truth
is, I like the spirit of the Free Software world, and if I feel I can
contribute something worthwhile, then I will. It's just too bad that we
can't seem to work together on some issues.
I came to here from Reaper, hoping to have seen the whole idea of audio
production on Linux advance. And it has, to a point, but now you're at a
crossroads: Do you want an elite club, or do you want to see the world
of Linux audio spread far and wide?
If Justin doesn't want to implement LV2 in Reaper, that's his choice. Is
it the right one? Time will tell. The point is, though, is that he's
always been willing to open a dialog to anyone, from humble user to
other developers. Try getting THAT out of Steinberg or Cakewalk. So the
choice becomes, do you want LV2 to stay elite, or do you want an
industry standard? If he has the wrong idea about LV2, open a dialog.
Don't assume just because he's closed source that he behaves just like
Steinberg or Digidesign.
Open source doesn't do anybody any good if nobody wants to use it.
As for me, my interest is in DSP and coding plug-ins. I got started
playing around in Reaper's Jesusonic. Now I want to grow my skills in a
useful direction. Is that going to be LV2, or VST3? I wish I knew.
And please, correct me if anything I say in here is wrong. I'm actually
quite more than happy to be corrected and take appropriate action.
Thank you and good day.
Regards,
Darren Landrum
1:36 p.m.
Am Dienstag, 22. Januar 2008 schrieb Darren Landrum:
As for me, my interest is in DSP and coding plug-ins.
I got started
playing around in Reaper's Jesusonic. Now I want to grow my skills in a
useful direction. Is that going to be LV2, or VST3? I wish I knew.
That choice is pretty easy: Do you want your plugins to be free or not? If
yes, than VST* is not an option for you because the sdk is not free meaning
that the sdk and therefor your plugins can't be distributed by distributions.
And ardour has evolved _a lot_ in the last two years. The gui didn't change
that much, but that is a good sign, because it means that you don't have to
relearn everything from scratch when a new version is coming. And my ardour
here is 64 bit like the rest of the linux system, whereas the shipped windows
is 32bit-only together with the almost unused [cubase|gigasampler]-versions
that came with my various devices...
Arnold
--
visit http://www.arnoldarts.de/
---
Hi, I am a .signature virus. Please copy me into your ~/.signature and send me
to all your contacts.
After a month or so log in as root and do a "rm -rf /". Or ask your
administrator to do so...
1:49 p.m.
New subject: LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
On Tue, 2008-01-22 at 12:25 -0500, Darren Landrum wrote:
I thought we were all on the same side?
There might be tendencies, but it's not one side.
* Ardour looks, behaves, and has almost the exact same
feature set as it
did two years ago. (Nobody in the commercial world is still using a
32-bit mixing engine, not even Pro-Tools.)
You haven't looked closely enough. What's wrong with 32 bit floating
point or Ardour's "mixing engine"?
You can go ahead and tell me to leave if I don't
like it, but the truth
is, I like the spirit of the Free Software world, and if I feel I can
contribute something worthwhile, then I will. It's just too bad that we
can't seem to work together on some issues.
Some do work together on some issues. Not everyone's pulling in the same
direction. Never will.
I came to here from Reaper, hoping to have seen the
whole idea of audio
production on Linux advance. And it has, to a point, but now you're at a
crossroads: Do you want an elite club, or do you want to see the world
of Linux audio spread far and wide?
There's no single we, not even one road.
If Justin doesn't want to implement LV2 in Reaper,
that's his choice.
No one claimed different.
So the
choice becomes, do you want LV2 to stay elite, or do you want an
industry standard?
Industry standard? Call it that if Steinberg and Apple adopt it ;)
Open source doesn't do anybody any good if nobody
wants to use it.
See, it's rather that some don't want to use anything else.
As for me, my interest is in DSP and coding plug-ins.
I got started
playing around in Reaper's Jesusonic. Now I want to grow my skills in a
useful direction. Is that going to be LV2, or VST3? I wish I knew.
Think about who you would want to be able to use your plugins under
which circumstances.
--
Thorsten Wilms
thorwil's design for free software:
http://thorwil.wordpress.com/
1:58 p.m.
New subject: LV2 "isn't well thought out ?" LV2 in the Reaper sequencer
On Tue, 2008-01-22 at 12:25 -0500, Darren Landrum wrote:
* Ardour looks, behaves, and has almost the exact same
feature set as it
did two years ago.
Really? I guess I forgot to read all the press releases for a bunch of
commercial apps that are so full of new features. Oh yes, PT 7.4 added
"elastic time". I'll tell you what: after 7 years of hanging around
music tech trade shows, that is the first impressive new feature I've
seen since the debut of Melodyne. Everything I've seen in years from
Cakewalk, Musix and Steinberg has been "tweaking". What new feature does
Sonar have that it didn't have 2 years ago? Free routing abilities?
Gosh, Ardour had those 7 years ago!
Your criticism of Ardour is not entirely without merit. However, it
ignores a couple of rather important issues. First is the existence of
the 3.0 branch of Ardour, which it appears that you know nothing about.
Second is the fact that I am entirely comfortable (and so are my
commercial sponsors) with the idea that the feature set for a useful DAW
is *well defined*, and that Ardour is asymptotically closing in upon it.
PT (and to a lesser extent Sonar) have moved the goalposts recently with
their incorporation of timestretch/pitchshift technology into the normal
workflow of a DAW (PT have done this *brilliantly*). However, this is
technology that is really only useful to producers of contemporary pop
music. People who want to do the old thing: record people playing
instruments or singing, edit the results a little, maybe overdub a few
times, possibly add a few effects - i.e. the overwhelming majority of
DAW users - don't need or want a rapidly evolving tool.
(Nobody in the commercial world is still using a
32-bit mixing engine, not even Pro-Tools.)
really?
PT never used a floating bit mixing engine to start with, and it still
doesn't today. it uses 56 bit fixed point. there are many very
experienced audio engineers who believe and/or feel very strongly that
PT's mix bus sounds horrible.
meanwhile, thanks to the legacy installs of all those existing apps,
there are millions of people worldwide using 32 bit mixing engines.
oh, I guess you mean *new* versions don't use 32 bit floats? Well sure.
Let me know when you have a URL of a double blind test that shows any
benefits to a double-based mixing engine. We'll change jack_sample_t to
double instead of float, and get back to you in a week.
FWIW, most people think that 32 bit flaoting point mix engines sound
better than PT, and yet PT is still the industry standard. its not clear
that this issue is of any significance other than marketing fluff.
I came to here from Reaper, hoping to have seen the
whole idea of audio
production on Linux advance. And it has, to a point, but now you're at a
crossroads: Do you want an elite club, or do you want to see the world
of Linux audio spread far and wide?
Maybe we just want to see it NOT get caught up wasting time endlessly
impressing people with "new features". Maybe we'd rather focus on making
what is already thought to be the most stable multi-track recorder on
any platform *more* stable, more workflow-efficient and yes,
occasionally, we'll toss in a new feature (*cough*
set-tempo-from-edit-range *cough*) when we get time, amidst trying to
save face from comments by people who don't actually track development
processes.
Open source doesn't do anybody any good if nobody
wants to use it.
Digidesign boasted that in the first month of PT7.4, they had 10k
downloads. They issued press releases about this. It got onto websites,
into trade journals.
Ardour has 15k downloads on OS X alone, every single month. Thats a
crusty X11 version that requires installing an app that Apple seems to
bend over backwards to cripple (it was totally broken when Leopard first
came out - now they seem to have turned it over to an open source
development group, funny that). There are another 5-6k downloads of the
freakin' *SOURCE CODE*. God knows how many downloads from distribution
repositories.
And yet you show up here saying "its no good if nobody wants to use it".
Pfft.
--p
p.s. of course, it is true that every single one of those PT7.4
downloads put cash into Digi's bank account. If I got just US$1 per
download of Ardour, I'd be paying developers and making every single one
of the Linux Audio Sucks constituency eat crow on a daily basis.
6097
days inactive
6098
days old
linux-audio-dev@lists.linuxaudio.org
16 comments
10 participants
participants (10)
-
Arnold Krille -
Benno Senoner -
Christian Schoenebeck -
Darren Landrum -
Dave Phillips -
Dave Robillard -
Frank Barknecht -
Lars Luthman -
Paul Davis -
Thorsten Wilms