Your vision of the future of Linux audio

Discuss how to promote using FLOSS to make music.

Moderators: MattKingUSA, khz

rghvdberg
Established Member
Posts: 777
Joined: Mon May 12, 2014 7:11 am

Re: Your vision of the future of Linux audio

Postby rghvdberg » Tue Jul 11, 2017 8:07 am

Hanging out in irc much.
Most questions / problems are about getting sound from the audio interface. Audio software itself doesn't pose big problems for users it seems.
Troubleshooting is a nightmare.
Is it the alsa kernel driver, jack, pulse, audio groups, limits.conf, kernel? It's insanity.

And I think people that actually get on irc are pretty techie already. That's a pretty wild assumption I know.

KX Studio has made a big impact for me. It's the closest thing to "Works out of the box".
Then again, I don't use pro audio soundcard devices.

User avatar
chaocrator
Established Member
Posts: 303
Joined: Fri Jun 26, 2015 8:11 pm
Location: Kyiv, Ukraine
Contact:

Re: Your vision of the future of Linux audio

Postby chaocrator » Tue Jul 11, 2017 3:37 pm

[*]"Too much latency".

this happen quite frequently on any OS (probably less frequently on Mac, because it supports less hardware), and often involves some kind of black magic to fix (like on one of my laptops, where NI Komplete Audio 6 works just fine with linux, but has latency issues with windoze 7).

User avatar
chaocrator
Established Member
Posts: 303
Joined: Fri Jun 26, 2015 8:11 pm
Location: Kyiv, Ukraine
Contact:

Re: Your vision of the future of Linux audio

Postby chaocrator » Wed Jul 12, 2017 9:43 am

another thing that might happen in the future in linux audio world — more people might realize how cool is the ability to run synths and other software in headless mode, and why.

but at the moment, doing fancy multimedia show on the screen considered obligatory for every single piece of software, so i even don't see any point to talk about headless boxes )

Jack Winter
Established Member
Posts: 323
Joined: Sun May 28, 2017 3:52 pm

Re: Your vision of the future of Linux audio

Postby Jack Winter » Wed Jul 12, 2017 10:22 am

After having watched and used linux audio for many years, albeit mostly reaper in wine, and other linux audio apps less. But also having spent many years trying to helps noobs, I can't help to think the following...

1. Distros, especially multimedia distros ought to have lowlatency and realtime kernels ready available.
2. There has to be an easy mechanism to add a realtime group, giving the user privs to use rtprio up to 98 and unlimited memlock.
3. Users need either a rt kernel or a lowlat with thredirqs.
4. Users need an easy way to change the priority of the soundcard or usb hub irq.
5. The confusion caused by jack1, jack2, jack-dbus, etc, needs to be attended to.
6. Problems running jack-dbus on a headless system.
7. Pulseaudio/jack integration issues (though this has gotten vastly better).

None of the above is unachievable with a little bit of elbow grease, but really ought to be automatic for new users.. For the rest linux audio is pretty easy, install some packages from a repo and start having fun..!
Reaper/KDE/Archlinux. i7-2600k/16GB + i7-4700HQ/16GB, RME Multiface/Babyface, Behringer X32, WA273-EQ, 2 x WA-412, ADL-600, Tegeler TRC, etc 8) For REAPER on Linux information: https://wiki.cockos.com/wiki/index.php/REAPER_for_Linux

User avatar
CrocoDuck
Established Member
Posts: 968
Joined: Sat May 05, 2012 6:12 pm
Contact:

Re: Your vision of the future of Linux audio

Postby CrocoDuck » Wed Jul 12, 2017 12:35 pm

ufug wrote:I am just hoping that we can keep moving forward. The community of developers is amazing, but it is small, and some/many projects are just one person. If there is a new job, or a baby, or health concerns, there is no guarantee that someone else will step in. Sometimes people say they will and then they disappear.


That's what I most concerned about. Linux audio is for sure much better now than what it was 10 years ago, but I cannot help feeling the whole thing is fragile. It evolves rapidly and software needs continuous attention.

As for the other points being raised about Linux being pretty hard for newbies, well I think I kinda agree in a way. However, I feel that the nature of Linux was overlooked in that analysis. Linux is first of all a very portable piece of software intended to being able to run on an ecosystem made of very diverse and disparate hardware platforms (servers, ARM embedded stuff, Inflight Entertainment Systems - yes, they all run Linux -, missile pointer systems for warships, ticket vending machines...) while retaining a monolithic kernel. If compared to Mac OS (which I use and with which I fight pretty often) it is a completely different beast: Mac OS is born to support -only- the restricted hardware with which Mac computers will be released (although it can be made to work on much more stuff, being essentially FreeBSD at its core). That comes with very different strengths and weaknesses. And I believe that one of the strengths is gonna be not having to understand what damn device is creating those nasty interrupts that makes xrun rain like a storm on that particular PC... Being so portable, Linux needs to retain deep configurability. That is why conf files are still the norm.

I get the point: bringing newcomers to Linux audio is hard as it requires to adapt to a new different world which require a deep comprehension. Many will not care, their goal being "music right away".

I don't believe this is only a disadvantage though: from adapting to Linux one gains more deep knowledge of audio machinery and a new mentality which I bet would increase productivity for many, if they had the strength (and/or time) to endure the learning curve. Also, I feel like open source software makes people to learn principles better, rather than commercial stuff.

For example, I learnt about FEM with COMSOL, and I was thinking I got the basics. Then I moved to ElmerFEM and I was suddenly in hell: not being able to set up a FEM problem that would converge. However, by studying the software I got much deeper insight of FEM itself, understood what good parameters are to set up the solver for my needs. After a month or so I brought my test projects to match COMSOL accuracy. And now I understood what it takes and why, from a FEM theoretical point of view.

In other words. COMSOL made me learn COMSOL, which makes a lot of automatic decisions on problems (damn you sneaky second order meshes...). ElmerFEM made me learn more about FEM itself, putting me in charge of designing good parameters for my solvers.

This is just an example, but I feel this is more or less what happens with open source software as a whole.
Check my Linux audio experiments on my SoundCloud.
Browse my AUR packages.
Fancying a swim in the pond?

User avatar
raboof
Established Member
Posts: 1558
Joined: Tue Apr 08, 2008 11:58 am
Location: Deventer, NL
Contact:

Re: Your vision of the future of Linux audio

Postby raboof » Wed Jul 12, 2017 7:40 pm

m4l3z wrote:what would you love to see software-wise, feature-wise, community-wise ..etc


What I hope to see is bringing developers and end-users together even more, and continuing to blur the lines between the two roles.

Sure, we have our 'entitled end-user' and 'ux-challenged developer' extremes, but there's a really colorful wide area in between.

Our broad community background gives us lots of opportunity for improving the "out of the box I just want to play music" experience as well as supporting the "crazy idea never been done before" experiment. I hope to see more of both.

User avatar
lucidbeaming
Established Member
Posts: 31
Joined: Tue Mar 21, 2017 8:44 am
Location: San Jose, CA
Contact:

Re: Your vision of the future of Linux audio

Postby lucidbeaming » Thu Jul 13, 2017 4:28 am

Embedded systems.

I've spent a lot of time working with Linux audio to create my own outboard gear running Linux on Raspberry. With the surge in modular synth popularity and new found interest in actual hardware synths, the ability to build your own is only going to gather more interest. Not everything needs a GUI.

m4l3z
Established Member
Posts: 3
Joined: Fri Jun 30, 2017 12:52 pm

Re: Your vision of the future of Linux audio

Postby m4l3z » Thu Jul 13, 2017 8:45 pm

Thanks for all the answers ! That was a nice surprise when I logged back :)
So
Nachei wrote:I think your question contains two possible interpretations that would be good to clarify:

yes, sorry if it wasn't clear enough, even though both questions are interesting I was specifically asking about how you would like the future of linux audio to be.

chaocrator wrote:another thing that might happen in the future in linux audio world — more people might realize how cool is the ability to run synths and other software in headless mode, and why.

What do you mean by headless ? as in no GUI ?

raboof wrote:What I hope to see is bringing developers and end-users together even more, and continuing to blur the lines between the two roles.

This.. I'm not sure how to make it happen. It was what I tried to say by "hub between users and dev". linuxaudio.org really inspired me because it acts a bit as a frontpage for linux audio as it references a lot of useful information, softwares, tutorials, toolkits and libraries for developing. But sadly a lot of links are dead and it's not up to date globally and not aesthetically very pleasant so I had the idea circling in my head to fork it and make a cute up to date front page. The point is that I think that a big picture of the state of linux audio helps knowing what is available, what do we miss, what is reusable/can we build upon. I also created an acount on linuxaudio.org, I should correct those dead links as soon as I have time.

If anyone is lurking on this thread, feel free to share your view/ideas !

folderol
Established Member
Posts: 764
Joined: Mon Sep 28, 2015 8:06 pm
Location: Here, of course!
Contact:

Re: Your vision of the future of Linux audio

Postby folderol » Fri Jul 14, 2017 7:40 am

I think we need more people doing the most uninteresting part of any project - documentation.
I've lost count of the number of programs that have great potential, but it's all 'hidden' with only some cryptic man page. Now I know there are people out that that actually enjoy this kind of work, but how do we find them?

User avatar
chaocrator
Established Member
Posts: 303
Joined: Fri Jun 26, 2015 8:11 pm
Location: Kyiv, Ukraine
Contact:

Re: Your vision of the future of Linux audio

Postby chaocrator » Fri Jul 14, 2017 9:06 am

m4l3z wrote:
chaocrator wrote:another thing that might happen in the future in linux audio world — more people might realize how cool is the ability to run synths and other software in headless mode, and why.

What do you mean by headless ? as in no GUI ?

as in no any UI )
1 (physical) machine is sufficient to control synths/plugins instances running on a few (physical) machines.

User avatar
CrocoDuck
Established Member
Posts: 968
Joined: Sat May 05, 2012 6:12 pm
Contact:

Re: Your vision of the future of Linux audio

Postby CrocoDuck » Fri Jul 14, 2017 10:20 am

m4l3z wrote: linuxaudio.org really inspired me because it acts a bit as a frontpage for linux audio as it references a lot of useful information, softwares, tutorials, toolkits and libraries for developing. But sadly a lot of links are dead and it's not up to date globally and not aesthetically very pleasant so I had the idea circling in my head to fork it and make a cute up to date front page. The point is that I think that a big picture of the state of linux audio helps knowing what is available, what do we miss, what is reusable/can we build upon. I also created an acount on linuxaudio.org, I should correct those dead links as soon as I have time.


THIS

plus this:

folderol wrote:I think we need more people doing the most uninteresting part of any project - documentation.
I've lost count of the number of programs that have great potential, but it's all 'hidden' with only some cryptic man page. Now I know there are people out that that actually enjoy this kind of work, but how do we find them?


I wanted to go through the documentation and start correcting links myself. I managed to do it just a couple of times before I run out of time. I can barely maintain the links I collected in my blog posts... But I believe I could find an hour or so every week in which I could give an hand.

By the way, about system configuration documentation, we found in this thread that it would be nice to have some experimental method to assess system audio performance, so that we can keep the audio configuration tips up to date or even discover new ones. Unfortunately, I cannot get this project started: too time consuming and I lack test computers. A clear way to test/discover tweaks, so that we keep the documentation effective, is something we need.
Check my Linux audio experiments on my SoundCloud.
Browse my AUR packages.
Fancying a swim in the pond?

User avatar
sadko4u
Established Member
Posts: 612
Joined: Mon Sep 28, 2015 9:03 pm

Re: Your vision of the future of Linux audio

Postby sadko4u » Fri Jul 14, 2017 10:57 am

folderol wrote:I think we need more people doing the most uninteresting part of any project - documentation.
I've lost count of the number of programs that have great potential, but it's all 'hidden' with only some cryptic man page. Now I know there are people out that that actually enjoy this kind of work, but how do we find them?

I think the best idea is to open LinuxAudio WIKI at LinuxMusicians to collect all information and documentation about up-to-date audio software. WIKI format is easy to use and edit by many people.
LSP (Linux Studio Plugins) Developer and Maintainer.

m4l3z
Established Member
Posts: 3
Joined: Fri Jun 30, 2017 12:52 pm

Re: Your vision of the future of Linux audio

Postby m4l3z » Fri Jul 14, 2017 11:15 am

CrocoDuck wrote: I wanted to go through the documentation and start correcting links myself. I managed to do it just a couple of times before I run out of time. I can barely maintain the links I collected in my blog posts... But I believe I could find an hour or so every week in which I could give an hand.


I started this morning and updated all versions and release dates for the category Sampler/synths(standalone) : https://wiki.linuxaudio.org/apps/synth_apps. I'll find some more time soon to work on another section.
I noticed that the website says that they are looking for front-end / web-devs so if you know someone spread the word, a more polished interface could really help.

Lenny
Established Member
Posts: 25
Joined: Tue Aug 22, 2017 5:40 am

Re: Your vision of the future of Linux audio

Postby Lenny » Wed Aug 23, 2017 2:15 am

I don't know exactly, but it for sure looks like open source is taking huge steps also the audio world.

When I started with Linux, there was only few buggy audio SW available, some sample processor and couple of humble sample trackers. That was a time when things like Reason were coming to Mac/Windows world, and the general consensus was that it's not gonna happen in OSS. Now I see Linux is in such a good shape that there's really no need to look for other options.

If people are looking for easy to use, polished products, Mac is still much better options. But I have this feeling that in the algorithmic music and music research Linux is a very good option. So if in the future we are seeing more and more music written by learning algorithms, maybe that's one direction where Linux audio could thrive.

User avatar
milk
Established Member
Posts: 94
Joined: Tue Jan 05, 2016 9:08 am
Location: Edinburgh
Contact:

Re: Your vision of the future of Linux audio

Postby milk » Tue Mar 20, 2018 5:44 pm

m4l3z wrote:Like what would you love to see software-wise, feature-wise, community-wise ..etc


* More Ableton/Bitwig style clip launch style loop sequencing - game-changing performance mechanics
* Beatmatching software for module music - so much classic hardcore techno, breakbeat hardcore, jungle, dnb, chiptune and more mods out there [1]
* Composers Desktop Project LV2
* An audio log/journal utility that does Opus. [2]
* Innovation in MIDI routing - more intuitive methods of automating send/insert patching. I can't really describe that until I get around to creating the like in Moony.lv2 or such..

[1] there is Chipdisco, with track muting, beatmatching and pitch alteration, but it's mono (left=cue, right=live) and buggy.
[2] there is fmedia, which does things like dynamic normalisation, but it doesn't have a GUI on Linux.

I'll be having a look at the LA wiki in the future, hopefully once OpenID login is fixed. I've spent many hundreds of hours working on my own wiki so there's a lot to update to a more 'central' information repo.
wiki.thingsandstuff.org/Audio etc. - a collection of mostly Linux and free links (OpenID auth for anti-spam, try DW)
Linux Digital Audio and Music Workstation Comparison Matrix - collaborative Google Sheet


Return to “Music & FOSS Advocacy”

Who is online

Users browsing this forum: No registered users and 1 guest