Greetings from Brisbane Australia

[Chrisblob]

Greetings to all, my name is Chris - to follow is a short description of me, and my musical gear/goals.

I'm a New Zealander living in Australia with my partner Shelley, and work in building services engineering.

I got started in music as a teen in the 80's, stumbling my way into bass guitar. I gigged with my father in a covers band in the early 90's, playing bass, but since then have treated music as a semi-serious hobby. My first setup in 1991 consisted of an Atari 1040STe, MasterTracks Pro MIDI software, and my bass. I upgraded to an Atari Falcon030 in 1994, dabbling with audio recording/engineering on that until 2002 when I got my first intel based PC running Mandrake Linux. I had a break from PC audio/MIDI after that, as audio tools in Linux were difficult to get working for me. In the last few years I wanted to get back into composition/recording so I invested in some gear, and did some work in Windows, as Linux was still making life difficult for me as as musician - mainly the software works, but not cooperatively - I just cannot stuff working together.

My gear consists of the following...

- Dell Inspiron 1720 w/ Core 2 Duo 2GHz, 2GB RAM, 400GB internal SATA disks. Runs Win Vista & openSUSE 11.2
- Tascam US122L stereo audio/MIDI interface
- M-Audio KeyStation 88es semi-weighted keyboard
- Alesis Trigger I|O drum trigger interface
- fretless Vester Stage Series bass, often downtuned to low D or even B.

I'm partway through building a DIY 8-piece edrum kit, and I'm also hoping to buy a couple of guitars (one elec, one acoustic), an alto sax made from bamboo, and an ocarina to complete my musical setup. The M-Audio keyboard, Alesis Trigger I|O, and Tascam interface are all seen by openSUSE 11.2, but the Tascam cannot be used - the keyboard & drum interface work fine.

I do all my recording in Windows, but would like to be able to do everything in Linux one day. The software is there, but I just need to somehow get it all working together, since due to the mess that is Linux audio it is too complicated to do so currently. So thats my goal - get all MIDI/recording/fx/synths working together under Linux with low latency, and of course free.

Regards,

Chris, Brisbane Australia.

[studio32]

Welcome!

Linux audio is not a mess once you launch JACK...

Urm not totally true... but it's far more then a mess, especially if you know what you're doing

[Chipfryer]

Welcome and I'm very interested in your building a kit too. This is something I also want to do.
Keep well.

[Chrisblob]

Welcome!

Linux audio is not a mess once you launch JACK...

Urm not totally true... but it's far more then a mess, especially if you know what you're doing

When I run jack I run it as root. I also need to run other audio softwares as root for most of them to be seen by jack, even then most sound bad, or don't work as expected. qsampler is the only software that runs 100% under jack. I've tried so many things to fix it I cannot remember half of them.

My biggest gripe is that the state of Linux audio is incoherent, fractured, and not likely to improve without a major overhaul. We have several subsystems, and then all the audio softwares pick & choose which subsystem they want to play best with, the end result being it works for some, not for others. For someone like me trying to create semi-pro quality audio it doesn't work. On their own there are some good softwares - zynaddsubfx, hydrogen, qsampler, rosegarden, ardour, audacity, and others to name those I use are good, but I cannot link them all together without something ruining the end result. Example - I have some nice grand piano giga files running in qsampler - I want to play along to some piano music (my piano is bad, I need to practice) but I cannot open an MP3 or CD because the MP3/CD players I have don't appear in the qjackctl connect dialog anywhere - all I get is silence. Another example - I'd like to record some piano from qsampler, but the only recording softwares that jack acknowledges are rosegarden & qarecord - jack kills both when I hit record.

I've come from Cubase SX in Windows to see if I can get a similar result in Linux - I'm not expecting powerful VST plugins or effects (I use minimal VST stuff in Cubase anyway), just the ability to record some clean live audio/MIDI, and mix/render down to good quality stereo audio to make into MP3's or CD's. It should be simple stuff, but unfortunately in Linux for me it isn't.

Thats partly why I'm here - the existing softwares I believe are up to the task, but the underlying subsystem(s) need a major shake-up. What I'm hoping to find here is advice on how to work around the various (and many) problems of Linux audio - you guys are possibly the best experts on that, as you go beyond just needing something to come out of the speakers.

Regards,

Chris, Brisbane

[Chrisblob]

Welcome and I'm very interested in your building a kit too. This is something I also want to do.
Keep well.

Thanks mate. I'll try put up some posts showing progress, but thus far I've only purchased the Alesis Trigger I|O, a couple of piezos for testing, and some misc parts. I'm close to the point of building the first drum head/shell, as a tester to validate my design - frankly they'll look ugly but so long as they don't break when I thwack them with a 7 weight, or have huge crosstalk, I'll be happy...

The Alesis Trigger I|O works fine in Linux - it will be dead easy to get it working with Hydrogen, and I've already created a monster hydrogen kit from some excellent DW drum samples. It will also be a big kit - 6 toms (incl 2 floor toms), 2 crashes, 1 ride, 1 splash, hi-hat, 1 snare, 2 kick pedals. There will not be any dual-zone stuff on this, just single-zone.

Regards,

Chris W, Brisbane Australia.

[studio32]

You clearly have a lot to learn on Linux, if you say you run all audio apps as root, bad practice! Is Jack behaving bad or is it just you...

Thats partly why I'm here - the existing softwares I believe are up to the task, but the underlying subsystem(s) need a major shake-up. What I'm hoping to find here is advice on how to work around the various (and many) problems of Linux audio - you guys are possibly the best experts on that, as you go beyond just needing something to come out of the speakers.

Please do me a favor and keep your advices for yourself for a while, you clearly don't understand anything from jack, so how should you be able to tell what is lacking and how it should be improved.

So welcome to this forum, if you take the time to learn things it might become something for you, otherwise you might be better off on Windows.


just my 2cents

[Chrisblob]

You clearly have a lot to learn on Linux, if you say you run all audio apps as root, bad practice! Is Jack behaving bad or is it just you...

Thats partly why I'm here - the existing softwares I believe are up to the task, but the underlying subsystem(s) need a major shake-up. What I'm hoping to find here is advice on how to work around the various (and many) problems of Linux audio - you guys are possibly the best experts on that, as you go beyond just needing something to come out of the speakers.

Please do me a favor and keep your advices for yourself for a while, you clearly don't understand anything from jack, so how should you be able to tell what is lacking and how it should be improved.

So welcome to this forum, if you take the time to learn things it might become something for you, otherwise you might be better off on Windows.

just my 2cents

Thanks for your 2 cents, you're entitled to give it, but leave out the insults please - if I've insulted you then I apologize, but looking at what I've written so far I fail to see how you could find *any* of that insulting. I'm telling it as it is, as I see it. No more no less.

Back on topic I need to run jack as root because jackd will not start with me as a normal user. I have added rtprio 100 and nice -10 to @audio in my limits.conf, but jackd still will not start unless I run it as root. I'm well aware that running as root is bad, but until I solve what is preventing normal users from running audio apps cleanly I'm stuck with that partial solution.

I'm all for learning, but wouldn't you rather focus on learning to use the end-user apps? I know I would. It frustrates me (and I'm sure many others) that in order to begin learning the end-user apps we first of all must learn how to fine-tune our audio sub-systems so those apps will even start and work properly. I'm all for removing that barrier from the creative process, so that as musicians we can just get on with what we do best - playing music.

Regards,

Chris W, Brisbane Australia.

[looplog]

Hi Chris,

I'm usually a lurker on this forum, but your post has brought me out to add to the noise.

I'm a relatively long-term linux audio user, and I have to say it's posts like yours that frustrate me. No offense. You were in fact the one that came in wielding insults, making accusations of all manner of things broken, without first asking if in fact your problems were localized to your specific system and configuration.

I'll be the first to admit that linux audio is downright infuriating at times, but those times are always when I'm forced to confront my own ignorance. In other words, I have to get out and learn how to use the system I chose to use. Linux requires you to learn, so if you feel that's a bad thing, as the previous poster said, maybe you should stick with windows. Then, pick up a sitar and see how well you go with no training.

The last I heard of running jack as root was a long time past, so it is safe to say the you *are* in fact doing something seriously wrong. So chill out on the accusations "that the state of Linux audio is incoherent, fractured, and not likely to improve without a major overhaul." Maybe that's just the state of your knowledge.

Here's some questions that you could answer before you make massive generalizations about the overall state of linux audio:
What distro are you running?
What JACK version are you running?
Are you running one of the custom audio distros or overlays or meta-distros (choose your term) designed to work with jack as regular user out of the box?
If not, why not?
If not, then you are in fact, choosing to run a self-configure linux audio system. If you choose to run a self-configure system, why do you complain about having to configure it yourself?
Where did you get the information regarding a rtprio setting of 100 (most guides seem at most to suggest 99, with many suggesting lower)?
More generally, what source of information are you using to learn how to configure and run your linux audio system?

Perhaps if you provide those details with a little less of an aggressive stance to linux audio in general, you'll get more helpful replies.

all the best
Michael

[Louigi Verona]

Dear Chris!

I am a several month linux user. If you study this forum and look for some of my old posts, you will see that I have had same opinions. I felt that linux audio is not professional, fractured and just a collection of homebrew halfbaked stuff. I felt no serious music can be produced with linux and that JACK and all the rest of the software, so praised by the linux audio community is not worth the fame.

Having worked very closely with linux audio for half a year now, I can say that I was wrong.

Linux audio is much more powerful than it seems. The reason why its power is not seen from the first glance is because computing with linux in general is done in a very different manner. In order to see what linux audio can do you have to:

1. Learn how GNU/Linux systems work in general. If you spend most of your free time fiddling with your system for the next few months, you will learn enough. If you have other interesting things besides a computer, it might take a longer time. Learning another operating system might be a tough but very interesting experience. It might take a while until you feel confident with a GNU/Linux system in general. But when you do, you will know it was well worth it.

2. Get to know all audio software that exists. And this is the part that will bring you surprise each time. In fact, I am still discovering new apps. Thing is - the amount of audio software for linux is enormous. Some of it is in beta and not complete, some if it is actually very complete, but most of it can be confidently used. Once you more or less get yourself familiar with all the software that exists, you will start to understand the possibilities. At the moment I do not think you can judge what linux can do simply because you don't have most of information.

3. Understand the method of modular audio. This is the killer concept of linux audio - a modular design. Taking apps and gluing them together into a system that works for you.

4. See that Windows is really almost as limited in what it offers. Thing is, a lot of the VST developed for Windows use only several filter, chorus, flanger and echo algorithms. There really aren't so many. A lot of VST designers give their VSTs a new look and work on nice presets, but the underlying mechanics and algorithms are not that different and for most of the time duplicate one another. Some might be written from scratch, but use the same mathematical principles, same physical models. In the Windows world there are several really good reverbs and algorithms for those that are open get used over and over in various plugins. Same for Linux - it has nice sets of plugins, like Calf pack, which offer confident results and which you can combine with each other thanks to the modular approach to create your own effect systems. On Windows there are more people and more development is aimed at Windows, but not a great deal more.

5. Note that Linux Audio is a long-term project which is always developing. Indeed. On a proprietary platform developers are usually paid substantial sums of money and that allows to speed up the development of a program. On Linux all projects are dispersed in time. But when the project IS complete, it is always there because it is free software, available for everyone and not susceptible to problems of "competing" market where developers push forward "new" versions which basically have little differences from previous ones. So a lot of Windows app progress is artificial and not real.
A lot of things are coming up and over the years you can see real substantial progress. That includes synthesizers, effects, DAWs, modular system itself. A session handler called LADISH which I've installed recently does wonders even in its present state.

Conclusion: Linux audio is much more than it seems. Without being a coder with a high level of expertise you simple have not a lot of real information to base your general conclusions on. As a user, you do not yet have enough experience to understand if Linux Audio suits your workflow. It might not, but chances are - it will. It'll just take time to see how to get there.

[Louigi Verona]

yeah, and note that I am not a tech geek, I am not a coder. I am a musician. But I now have no problem compiling stuff cause that's real easy once you know the process.

[Chrisblob]

Hi Chris,

I'm usually a lurker on this forum, but your post has brought me out to add to the noise.

I'm a relatively long-term linux audio user, and I have to say it's posts like yours that frustrate me. No offense. You were in fact the one that came in wielding insults, making accusations of all manner of things broken, without first asking if in fact your problems were localized to your specific system and configuration.

You call what I wrote insulting? The reason I call it broken is that out-of-box it does not work. That would seem to be a fairly simple clarification would it not? I've read plenty of forums where users experience similar troubles in getting the audio experience they want out of a Linux system, many with no solution, and many copping flak such as I am for seemingly daring to question the state of Linux audio, so I think for me to say its a system that requires work is both accurate and warranted. You're welcome to disagree.

I'll be the first to admit that linux audio is downright infuriating at times, but those times are always when I'm forced to confront my own ignorance. In other words, I have to get out and learn how to use the system I chose to use. Linux requires you to learn, so if you feel that's a bad thing, as the previous poster said, maybe you should stick with windows. Then, pick up a sitar and see how well you go with no training.

Confronting your own ignorance? It isn't you that is ignorant, far from it, nor am I ignorant. Our choice of Linux as an OS bears witness to that. I just do not adhere to a "this is what you get, lump it or leave it" attitude - thats too much like the world of proprietary software. Surely you can see the irony in your standpoint there? On one hand you agree Linux audio can be a PITA, then on the other hand you deride yourself for not accepting it at face value, then bow down to accepting its problems? I'm confused.

I've used Linux for 9 years now. In that time I've managed to coerce all manner of poorly supported hardware to work on my systems, such as webcams, TV-cards, GSM modems etc. There is a method I have learned to deal with each problem. In that whole time I've failed to get audio working in a state that would allow a good DAW system to work, but I would still much rather use Linux than Windows as my daily driver.

The last I heard of running jack as root was a long time past, so it is safe to say the you *are* in fact doing something seriously wrong. So chill out on the accusations "that the state of Linux audio is incoherent, fractured, and not likely to improve without a major overhaul." Maybe that's just the state of your knowledge.

I know running it as root is wrong. However as I said before its the only way I can get jack to run until I fix it. In terms of doing something "wrong" I just installed the audio softwares and tried to run them, individually to begin with, then when I tried to run more than one app at a time via jack thats when the fun began. Do I need to assume I cannot just install software and run it?

Here's some questions that you could answer before you make massive generalizations about the overall state of linux audio:
What distro are you running?
What JACK version are you running?

I'm running openSUSE 11.2 64-bit, with the default alsa 1.0.21, and jack 0.118.0 - all as packaged by the openSUSE distro guys. All my audio apps come prepackaged from packman or opensuse, most of which are quite recent versions.

Are you running one of the custom audio distros or overlays or meta-distros (choose your term) designed to work with jack as regular user out of the box?
If not, why not?

Negative. All the needed apps were already supplied with the standard openSUSE 11.2, so I could only assume they should work, why else are they there? I'm aware of dedicated distros for media, and meta-distros, but I would like to know what will happen to everything in openSUSE 11.2 that *does* work before I go over-writing anything.

If not, then you are in fact, choosing to run a self-configure linux audio system. If you choose to run a self-configure system, why do you complain about having to configure it yourself?

openSUSE 11.2 (compared to previous releases) has been great in getting some less common hardware working for me - most things are plug n play. Audio worked out of box, but only one app at a time. I'm obviously not running a self-configured system, in fact I've not needed to do much configuring at all - audio I haven't touched other than the rtprio and nice fixes for jack (which didn't work).

Where did you get the information regarding a rtprio setting of 100 (most guides seem at most to suggest 99, with many suggesting lower)?

Thats suggested in the error log that jack throws up when trying to start it as a non-root user. Jack says jump, I say how high, but it didn't work.

More generally, what source of information are you using to learn how to configure and run your linux audio system?

Google is my friend, installed alsa docs, dredging thru forums etc

Perhaps if you provide those details with a little less of an aggressive stance to linux audio in general, you'll get more helpful replies.

all the best
Michael

This is obviously a sensitive topic. At that I'm surprised at the stance you guys take, as you seem fully prepared to just accept the fact that Linux audio could be a lot more user friendly, with little or no perceived desire on your part to improve the user experience. I'm not here to annoy, I'm here to try and get advice on working around my audio troubles, and perhaps motivate people into helping better the experience for future users. If my desire to see a better user experience upsets people, I could think of worse things to get upset about.

Regards,

Chris W, Brisbane Australia.

[studio32]

Thanks for your 2 cents, you're entitled to give it, but leave out the insults please - if I've insulted you then I apologize, but looking at what I've written so far I fail to see how you could find *any* of that insulting. I'm telling it as it is, as I see it. No more no less.

AFAIK I didn't insult anything. I pointed out my opinion, clearly, but not insulting. It just has little sense that someone who uses primarily Windows, to tell how Linux audio should be, especially not if he shows that he has pretty little clue about Jack, which is no insult, we all had no clue about Jack at start.

As more people pointed out here, critique is welcome, that's not the problem, but if you're new you better try to understand why and how first, before you tell experienced people how it should be.

I suggest you to post your problems with Jack on this forum, in the topics where they belong (not in an introduction topic). When you didn't solve your problems or you still have a clear opinion about how Linux audio could be better, over about let say 3 months, you are free to post it here. Almost everyone of us is copyright holder of such an topic, so no problem at all.

Again, welcome and post your problems and question on the places where they belong, thanks mate

[Chrisblob]

yeah, and note that I am not a tech geek, I am not a coder. I am a musician. But I now have no problem compiling stuff cause that's real easy once you know the process.

Thanks for your input Louigi. I'm no coder, although I'm a bit of a tech geek. I'm also not a trained musician (mostly self-taught, read a step editor better than a score!), but I know every millimeter of the bass guitar fretboard, and do an OK job of drums & guitar - just don't ask about keyboard, I totally suck at that! Thats partly why I want to get my audio working better, to improve my keyboard skills.

I'm also no stranger to compiling, but I feel the problem could lie in the configuration of the software, not in the software itself - I could compile everything, but I'd still need to work out how to configure it. The audio subsystem is not known for its intelligent configuration tools. To me its like taking the cockpit out of a Boeing - you have this complex incredible machine, but with no cockpit you have no controls to adjust all the parameters for the aircraft systems. The pilot is not expected to climb into a service duct to connect a cable to actuate a flap - he does that all from the cockpit. I think Linux audio could do with better cockpits.

Regards,

Chris W, Brisbane Australia.

[spm_gl]

The reason I call it broken is that out-of-box it does not work.

Out of which box? Unless you use a distro where someone else has made sure the components work together, you'll have to do it yourself. And noone can be held responsible for the multitude of hardware configurations possible.

[Chrisblob]

Thanks for your 2 cents, you're entitled to give it, but leave out the insults please - if I've insulted you then I apologize, but looking at what I've written so far I fail to see how you could find *any* of that insulting. I'm telling it as it is, as I see it. No more no less.

AFAIK I didn't insult anything. I pointed out my opinion, clearly, but not insulting. It just has little sense that someone who uses primarily Windows, to tell how Linux audio should be, especially not if he shows that he has pretty little clue about Jack, which is no insult, we all had no clue about Jack at start.

Telling a 9 year Linux veteran he's better off in Windows could be viewed as insulting...

I use openSUSE 80% of the time, Windows running in a VM under Linux 10% of the time, and Windows running natively in the remaining 10% of time. So Windows is definitely not my primary, nor my preferred OS.

As more people pointed out here, critique is welcome, that's not the problem, but if you're new you better try to understand why and how first, before you tell experienced people how it should be.

Thats the thing, I'm not that new, and I'm an avid reader of forums, so these issues are not new either.

I suggest you to post your problems with Jack on this forum, in the topics where they belong (not in an introduction topic). When you didn't solve your problems or you still have a clear opinion about how Linux audio could be better, over about let say 3 months, you are free to post it here. Almost everyone of us is copyright holder of such an topic, so no problem at all.

Again, welcome and post your problems and question on the places where they belong, thanks mate

I will do. I apologize if you consider my posts to be an attack - they are not. It is more of a vent of my frustration at not figuring this out myself - in 9 years it is the only Linux issue ever I have not been able to figure out.

Regards,

Chris W, Brisbane Australia.