Hey Tres!
I've read your responses and to you and other participants of the list I'd
like to explain where I am coming from. Please bear with me, I tried to not
make this text unnecessarily long, but it is.
First, because I do program as an amateur from time to time and I'd worked
as a professional web developer, I do know that programming is tough. So I
do appreciate that programming in general is a very tough thing and putting
out a serious product is very, very difficult and takes a lot of time and
dedication.
Second, in my own creative work I find that honest criticism is best at
improving. And because I really care about the state of Linux Audio, I am
always an honest critic. However, when it comes to Linux Audio, I am not a
programmer, but a musician. And so I give musician's perspective.
Because musician's perspective is all about producing music, I have to
honestly disregard any "coding troubles" and sort of switch off the part of
my mind that understands and sympathizes with programming as an activity
with its challenges, losses and wins. I have to look at actual
functionality of the products in question.
What follows below is a musician's perspective of one person. I totally
understand it is just my opinion, so take this just for what it is - an
opinion. Why it might be useful? I believe I am a good enough musician and
a rather active one. And so perhaps some of it could be relevant.
I understand that LADSPA plugins "work" in that they operate. I also
understand that their coding can be non-trivial in the sense of algorithms
and implementation. However, this says nothing about their actual function
and I believe that music plugins are usually judged by their merits as
music tools, not by their non-triviality as coding exercises.
LADSPA plugins are very different in quality offering. However, most of
these 200 LADSPA are basically uninteresting. I would be surprised if many
musicians disagree. Countless limiters and countless compressors and very
basic eqs. And there are hardly any interesting effects that can be used
for actual music making. And the reason they work after 12 years would be
much more remarkable if they would do sophisticated things. What can break
in a basic limiter? I can open numerous VST plugins made more than 12 years
ago and they would still work, yes. So what?
The fact that these plugins are free of charge and that they are free as in
freedom is not an argument in light of their usefulness as music tools.
Now, I made a statement that there are many copies of same functions and
they are uninteresting. Let's talk actual numbers.
In my LADSPA collection I see this:
13 Limiters
11 Compressors
16 Eqs
7 Amplifiers
42 Filters
8 Reverbs
9 Choruses
7 Phasers
1 Vocoder
Let me stop there for now. There are some more categories, but this is the
most of it.
Now, it seems like a nice bunch, but what's important here is quality, not
quantity. A commercial-grade DAW can have only 10-20 built-in effects, but
they can be really quality stuff and it can be enough for serious work. In
fact, I still like and use many of the FL Studio built-in plugins, which
hardly change over the years. Their Reverb is fantastic.
With LADSPA, if you actually start exploring these offerings, it is clear
that the amount of plugins does not mean diversity and choice.
First of all, many of them are very basic. And I mean - basic. As in being
coding exercises. See those 42 filters? Well, more than half of them is
just 2-3 knobs, sometimes even one, like in Vowel filter. The sound is
nothing special - standard and usually of very low quality.
Many of then simply don't work. Just do nothing, no matter what knobs you
turn. Although I have went through the whole LADSPA collection years back,
when randomly exploring filters now, among the 15 I picked I came upon 3-4
non-working. Several (3-4) among them had very weak effect - basically,
just worthless stuff. Probably indeed someone's experiment in DSP. No
problem with that, but these are not "fantastic plugins" from the point of
view of music production.
And this story repeats with other categories. They are all either basic
and/or not working. Try it for yourself right now. Are you seriously going
to use this?
Take delays. Apart from Vintage Delay, all of them are almost unusable as
they have delay in ms. Musicians do not work in ms, they work in bpm. I
have to take a calculator each time I want to use a delay? Not only that,
many of these delays are, again, very basic, some are weird and I have no
idea what they are doing. Apart from Vintage delay, none of them have ping
pong mode and other things, which are today part of a delay plugin. Some of
them are literally "delay", meaning they delay the incoming sound a certain
amount of seconds. In other words, this is a different function altogether!
Take away CALF Vintage Delay - and Linux Audio has no normally
standard-grade delay plugin.
And, finally, the basic nature of most of these plugins seems like all of
them are just parts of a bigger thing that noone really put together. By
"effects" we can mean very sophisticated tools, like sweeping filters,
various filter sequencers. Things you see in Rakarrack, to take a Linux
example. They've managed to not only take from Zyn, but also write their
own cool effects. If rakarrack plugin set was available in LMMS, that would
be a very strong offering.
What is the conclusion?
The conclusion is that I propose to not try to focus on the quantity of
plugins and rather talk about quality. If LMMS would have 1 good Chorus, 1
good Flanger, 1 good Phaser, 1 good Filter, 1 good EQ, 1 good Delay and 1
good Reverb - this would be enough. We don't need 200 low quality stuff
that you have to sort for hours to find something that works.
200 LADSPA plugins is an illusion. It is 10-20 usable standard-grade
plugins at most and by usable I mean things that a musician might end up
actually using, not any plugin that is able to operate. Among those 10-20
there are almost none sophisticated, just basic stuff, but at least it
mostly does its job.
LADSPA collection has several bright spots. CALF set is an obvious choice.
Also Zyn stuff. You can find them in KXStudio repos. Previously they were
called HOLAP Plugins, but the site is now absent. You can contact Rakarrack
developers for that, it was they who made these effects into LADSPA.
And fellas - I do hope my letter would be takes as it was intended. Just an
honest analysis of what we really have on the platform. I mean no offense
to anyone and in fact I am ready to change my opinions if new data would be
provided.
Louigi Verona.
http://www.louigiverona.ru/
On Fri, Jun 27, 2014 at 7:47 PM, Tres Finocchiaro <
[email protected]> wrote:
> I just want us to be constructively honest here.
>
>
> Being constructive is getting these working. Reporting that they are
> broken is a good first step, but please understand the difference.
> Construct means to "build or erect" which is what we need more of.
>
> Many of them are unusable
>>
> It would probably be better to specify here as "unusable" can be literal
> or opinion. There are over 200 effects now so "many" could be "5" or it
> could be "100". If you are aware of broken LADSPA plugins (in the literal
> sense) please file a bug report so that they can be disabled until they are
> fixed. There are certainly duplicates and we'd entertain a better
> organizing method if someone wanted to provide some recommendations/mockup.
>
>
>> A whole class are just basic stuff cloned. They are not "fantastic".
>>
> I'm not sure if you've read the source code for these but it's not trivial
> <https://github.com/swh/ladspa>. The fact that some of these still work
> 12 years after they were written *is fantastic* and *reason for
> celebration :)*, but these plugins can do many things. They may be
> clunky and there may be bugs but they can reverb, delay, compress, distort,
> warm, filter, eq and much more. They may not be commercial grade, but they
> are completely free, built-in and provide out-of-the-box capabilities that
> tends of thousands of composers use and enjoy. They ARE fantastic and we're
> very lucky to have them. :)
>
>
>> Ladspa collection lacks a proper reverb and zynreverb is the only
>> non-convolution reverb on linux that can offer quality cathedral spaces.
>> Phaser and flanger from zyn are also extremely good. If it is possible to
>> have internal lmms effects, it would be good, but we had that discussion
>> already.
>>
> Thanks for the explanation. That'a good argument to include them by
> default once the bugs are addressed.
>
>
--
Louigi Verona
http://www.louigiverona.ru/
------------------------------------------------------------------------------
Open source business process management suite built on Java and Eclipse
Turn processes into business applications with Bonita BPM Community Edition
Quickly connect people, data, and systems into organized workflows
Winner of BOSSIE, CODIE, OW2 and Gartner awards
http://p.sf.net/sfu/Bonitasoft
_______________________________________________
LMMS-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/lmms-devel
