[LAD] [LAU] So what do you think sucks about Linux audio ?
louis.gorenfeld at gmail.com
Wed Feb 6 21:34:21 UTC 2013
I'm late to this thread, but I'd like to shine some light on why I
switched from Linux to Mac OS for my personal music work, and things
which I believe may continue to hold back Linux development. Keep in
mind that I've only picked at Linux audio for the last 5 years, so
maybe some of these complaints have since been addressed.
- Make it easier for authors to support Linux: I'm the author of a VST
plugin. I decided to go with VSTGUI for my Linux port. However,
despite being an open-source project, VSTGUI is somehow not supported
on Linux. Due to time constraints, this resulted in me just yanking
the GUI support from my Linux version. Also due to time constraints, I
was not able to produce an LV2 version which some people wanted. I get
that VST is technically not a free-free standard. But it's also the
most widely-supported plugin format, with a staggering number of free
plugins. Unless there's something I'm not seeing, this seems perfectly
support-able on Linux (it's halfway there already!).
- Fragmented and not well-supported plugin architectures: It seems
that plugins on Linux have gone through several phases. When I used
it, it was all DSSI. Now I guess it's LV2. I think the most promising
time was when Linux VST was gaining momentum, but now it seems that
maybe this isn't an accepted format on Linux. And don't say JACK--
that's for sharing audio between applications. If I go to KVRAudio.com
and search for Linux plugins, there are only 50.
- Make Linux friendlier for closed-source/commercial devs: Open source
is great; don't get me wrong! But music software and DSP are
specialized areas and DAW and associated software is incredibly
complex. I think for Linux to really succeed in this arena, it will
have to attract commercial development. Easier said than done, I know.
- In the same vein: One thing I notice is that many music packages on
Linux seem to be by-and-for programmers. It's all very flexible and
modular, but this is not necessarily convenient when your brain is in
"music making mode". Ask electronic musicians what their work flow is
like and design applications around that kind of feedback. On the
other hand, coders always want everything to be a modular synth. It's
literally the first kind of synth project everyone I know who's a
coder has done!
- Is there a good all-in-one sequencer for electronic music styles?
This was the thing that really made me switch. I was using Rosegarden
at the time. And I don't want to call out Rosegarden because I do
believe it's a wonderful piece of software for music notation. It just
wasn't suitable for my uses. It had issues handling a
reasonable-but-large amount of MIDI data on my 1ghz machine (I use
pitch wheel for vibrato), and I wasn't ever able to render my song
down to disk in a non-realtime mode. People pointed me to using JACK
freewheel mode, but it seemed like at the time this wasn't supported
by synth developers. Finally, there wasn't a way to use samples other
than SoundFonts (which, IMO, are both squarely within the consumer
realm and are not easily editable) -- again, this was a few years
back, so maybe this has all been addressed.
The irony is that, shortly after, Renoise came out for Linux! :) And
maybe Ardour is suitable for electronic music; I dunno. Like I said,
I've been partially out of the loop.
- Not knowing if your sound card will be supported. It's hard to shop
for sound devices with Linux. "Supported" may mean all kinds of
things. For example, I found issues with the AudioBox 1818VSL which
were down to ALSA incorrectly implementing the UAC2 standard. I do
realize that a lot of hardware support is down to the hardware
developers being more enthusiastic about Linux. But the bottom line
for would-be Linux musicians? You never know how sound hardware will
react with Linux. To be fair, this is still true of Windows to some
extent where latencies are concerned (but not with basic
- Clearer documentation for ALSA would help (pretty please :))
Hopefully this isn't too harsh or outdated. And I don't want to just
heap criticism on Linux, because I think it has amazing potential. The
realtime kernels offer amazing performance, and so does ALSA-- I got a
netbook I was using for Maker Faire down to an incredibly low latency
on Linux. It performed much better than ASIO did on the same hardware.
It can be lightweight and incredibly customizable.
Bring on the software!
On Tue, Feb 5, 2013 at 6:58 AM, Dave Phillips <dlphillips at woh.rr.com> wrote:
> I've been reading a lot of negative (read: vitriolic) commentary about the
> world of Linux audio development and applications. I won't bother to say
> where, just "the usual places" will have to suffice. Of greater interest to
> me is the commentary itself - it seems to boil down to the following plaints
> and lamentations (in no particular order) :
> Too many distros.
> Too many audio-optimized distros.
> Not enough native plugins, esp. instruments.
> Inconsistent support for VST/VSTi plugins.
> Too many unstable/unfinished applications.
> Too many "standards" (esp. wrt plugins).
> Poor external/internal session management.
> Poor support for certain modes of composition (think Ableton Live).
> Lack of support for contemporary hardware.
> Confusion re: desktops, and GUI toolkits.
> Too difficult to set up audio system.
> JACK is a pain.
> Too much conflict/fragmentation within the development community.
> I'm not so interested in comments on the commentary, I have my own, but say
> what you will about the list. I figure that most denizens of these lists
> already have ready replies and responses to these and other criticisms, many
> of which have been voiced here previously. What I'm more interested in is
> what *you* think is missing most or just plain wrong about the situation.
> Please, try to speak your piece without flames or dissing other developers
> and/or their work. Frankly speaking, I've had enough of that crap, and I'm
> most thankful these days for such forum amenities as "mute user" and
> autodiscard, along with the standard filters found in mail clients.
> I'm reminded of John Cage's comments regarding the behavior of the NY
> Philharmonic when they destroyed his equipment during the premire of Atlas
> Eclipticalis, something to the effect that his concerns had ceased to be
> musical and had become social, i.e. that he had to figure a way to allow
> people to be free yet behave themselves with respect towards the common goal
> (e.g. Cage's music and property). I'm going to guess that he was still
> working on that up to his death.
> So, in your honest and bold opinion as user and/or developer, what do we
> lack most and what can we do without that we already have ? Please feel free
> to expand your remarks as you like. I'm planning an article on the topic and
> will likely use selected comments, subject to approval of course.
> Linux-audio-user mailing list
> Linux-audio-user at lists.linuxaudio.org
More information about the Linux-audio-dev