[GTALUG] Seeking help with a unique Linux project?

Tue Dec 22 15:51:19 UTC 2015

I don't think it'll make much of a difference for command line usage. The
differences are more in the graphical desktop environments, packaging and
distribution, etc.

The reason I like KXStudio is that you can install the distribution on the
machine, *or* you can just add the KXStudio repositories to a supported Debian
or Ubuntu installation.

Since I was already using Debian (Jessie), it was convenient for me to just add
the KXStudio repositories to my existing machine. (The only extra step was
getting the Liquorix kernel, since the KXStudio repositories don't include a low
latency kernel.) That way I could have all the stability and familiarity of
Debian, but with up-to-date audio packages, and KXStudio assistance in
optimizing configuration for real-time audio.

I haven't tried AVLinux myself, but I don't think it'll be all that different
from Debian or KXStudio as for as command line access goes.

Blaise

ps I was scanning the discussion on LAU about this. I think F. Silvain is right
that most of the differences between difference audio distributions aren't
relevant for command line usage (
http://lists.linuxaudio.org/pipermail/linux-audio-user/2015-December/103578.html
) , but he just overlooked a few things:

- package management, software updates: if you're familiar with Debian/Ubuntu,
then you might prefer aptitude/apt-get to Arch's repository management, or vice
versa. This is used for installing new packages, or getting updates to existing
packages

- The audio distributions still might help with some of the configuration for
low latency audio, and getting the latest pro audio packages. You could
certainly make those adjustments manually in a non-audio distro, but if you
haven't done it before, I still think it'd be worth going with something like
AVLinux or KXStudio to reduce that learning curve and get started faster

On Sat Dec 19 19:23:35 UTC 2015, Karen Lewellen wrote:
> Hi Blaise,
> Again thanks for that terrific article link.
> I am wondering which of the two options presented  is likely to allow for 
> the greatest command line access?
> I am feeling avlinux, but that is just a feeling.
> The command line aspect is critical since  a graphical structure will not 
> work  with speech for me if I come in from the outside.
> I will cripple check that to be certain though.
> Thanks again,
> Karen
> 
> 
> On Tue, 15 Dec 2015, Blaise Alleyne wrote:
> 
>> I'll keep this on-list since people seem to be interested!
>>
>> Libre Music Production has a good getting started guide on selecting a GNU/Linux
>> distribution for pro audio:
>> http://libremusicproduction.com/articles/advantages-choosing-audio-orientated-linux-distribution
>>
>> That article recommends an audio distribution, because they have certain tweaks
>> applied, like the low latency kernel selection, or changes to
>> /etc/security/limits.conf with respect to the audio group, etc. Using an audio
>> distribution, or the KXStudio repositories, handles much of that for you.
>>
>> (But the KXStudio repositories don't include the kernel.)
>>
>> In Ubuntu, there's a linux-lowlatency package. In Debian, I added the Liquorix
>> repositories to get that low latency kernel.
>>
>>
>> That Libre Music Production article doesn't go into a lot of technical detail,
>> but it explains:
>> """
>> Realtime audio performance is important to many when working with audio
>> software. What this means in real terms is having the recorded signal played
>> back through your playback system without any audible delay. This is important
>> when monitoring your performance. Latency is inherent in computers but there are
>> things that can be done to minimize it.
>>
>> There are many tweaks, big and small, that Linux distros make under the hood
>> that are beneficial to this and general audio performance. As an end user, you
>> don't need to know the technical ins and outs of what these do. You just need to
>> know that the end result is that audio processes are prioritized by the
>> operating system, thus, allowing you to more easily achieve stable and/or low
>> latency audio.
>> """
>>
>> There's more technical information on latency and audio production in the Ardour
>> manual, but far more in depth than you really need to understand!
>> http://manual.ardour.org/synchronization/latency-and-latency-compensation/
>>
>>
>> I've tried using the standard Debian Jessie kernel. With that on my machine, the
>> best I can do without audio glitches (xruns) is about ~21ms latency. With
>> liquorix, it's ~2ms.
>>
>> If you're playing something slow or flowy, you might not even notice the
>> difference. But if you're playing a series of fast notes, or something rhythmic
>> or percussive, it's really disorienting to have that slight delay between
>> pressing the keys on your keyboard and hearing the sound. It's often noticeable
>> in normal music performance.
>>
>> A low latency Linux kernel -- which is basically the standard Linux kernel, but
>> with some changes around scheduling and deadlines -- just allows for a more
>> natural feel for pro audio, where you can hear notice when you play them rather
>> than with a slight delay.
>>
>> Hope that helps...
>>
>> Blaise
>>
>>
>>
>> On 15/12/15 03:27 PM, Karen Lewellen wrote:
>>> Hi Blaise,
>>> one of two.
>>> Thanks for the off list invitation!  I shall certainly write you, but at least
>>> given another post, will keep some of this on list.
>>> i am posting the  software options  in that reply.  still can you explain, or
>>> can others give examples of low latency kernels?
>>> What a fascinating idea.
>>> More in a moment,
>>> Kare
>>>
>>>
>>> On Tue, 15 Dec 2015, Blaise Alleyne wrote:
>>>
>>>> Hi Karen,
>>>>
>>>> Thanks for the additional details!
>>>>
>>>> I'm Debian-based these days, with the KXStudio repositories for current audio
>>>> production software and a low latency liquorix kernel, though I'm also still
>>>> running Ubuntu Studio on my home desktop.
>>>>
>>>>
>>>> One thing that may be a challenge (for which others on the list might have more
>>>> knowledge) is your issue with speech in post-Squeeze Linux kernels. For pro
>>>> audio work, the standard Squeeze kernel isn't going to be great... you want a
>>>> low latency kernel if you want notes to play when you press them on a MIDI
>>>> keyboard, or to be able to play back audio from several instrument sources at
>>>> the same time reliably, nevermind to record any audio reliably.
>>>>
>>>> So I'm not sure how you solve that problem -- an older low latency kernel
>>>> perhaps, which supports your speech hardware requirements but also pro audio?
>>>> I'm not quite sure.
>>>>
>>>> (Also, are you using JACK yet? That may be another challenge with speech
>>>> software, though an ALSA or Pulseaudio bridge may do the trick.)
>>>>
>>>>
>>>> Beyond that, maybe best to take the discussion off-list? Feel free to email me
>>>> at blaise at apeironmusic.com.
>>>>
>>>>
>>>> Blaise
>>>>
>>>>
>>>> On 15/12/15 01:11 AM, Karen Lewellen wrote:
>>>>> Hello Blaise,
>>>>> Thanks so very much in advance for the wisdom I feel sure you can provide.
>>>>> I honestly admit  to being a touch of a singer snob who imagines one can only
>>>>> perform  live with well other live people smiles.
>>>>> So when Bob wrote of your creating, I confess I got rather curious as well.
>>>>> As for command line music making  tools, there are actually rather a few I  am
>>>>> learning.  I am compiling a list with links, since I do not know yet your
>>>>> distribution of preference, or exactly how you work.
>>>>> Articulating my desires musically may make more sense than the ssh telnet aspect
>>>>> truth be told.
>>>>> In all honesty however, I cannot say with hands on firmness that I can drive the
>>>>> car where I wish to go.
>>>>> I was more of a singer / arranger than a songwriter until the past few years.
>>>>> Now I have all this music I desire moving from the studio in my head and heart
>>>>> onto sheet music, into performing form, as I am sure you can understand.
>>>>>  Since I am writing full arrangements, all the parts with my lyrics, I want to
>>>>> play all of the parts, piano, strings, guitar etc.,  that are in my head and
>>>>> then
>>>>>  get the completed work into sheet certainly both audio and sheet possibly.
>>>>> How I imagine doing this, is with a quality   piano / keyboard that also has
>>>>> several additional instrument options to simulate sound wise for composition
>>>>> only for now at least, the other instrument's for which I am writing.
>>>>> <Still awake?>
>>>>> Now, this is where my visual memory, blended with my lack of first hand Linux
>>>>> experience starts to get a bit...well hopeful.
>>>>> I imagine connecting said piano / keyboard  to the m-audio audiophile 2496 sound
>>>>> card I have in mind,  using any one of the dozen or so tools for the command
>>>>> line I am finding  are included in Debian, writing each part one at a time,
>>>>> putting the piece together, and transferring the completed work to a composing
>>>>> package, you referenced Lillypoind,  which is terrific, but  there are a couple
>>>>> more.
>>>>> I am hunting the instrument along with the talent for the Linux side of my
>>>>> desire.
>>>>> The older the better in a way, I tend to be a buttons kind of girl, not a touch
>>>>> screen one.
>>>>> Granted I know some keyboards have  multi-track sequencers built in. Letting me
>>>>> <I think?  compose entirely on said keyboard and transfer the finished
>>>>> product...strong emphasis on the  I think.
>>>>> There may be another way that I have not considered, I mean besides the
>>>>> traditional way  of doing the recording, I am a radio producer too, so have pro
>>>>> tools, a mac  a control surface etc. etc.
>>>>> Still I would rather use Linux because I know the composition  programs exist,
>>>>> that i may be able to script in Linux some of what I need...not me, but tap into
>>>>> some sources for this.  There is no real program for composition  mac wise,
>>>>> certainly not going back to the edition I am using to run my pro tools setup.
>>>>> The ssh-telnet aspect simply lets me treat the Linux box like a server, running
>>>>> the commands  from  my main computer where I am already comfortable, both with
>>>>> my speech synthesizer,  and in a small way  with the command line based on my
>>>>> daily use of a comparative Linux shell.
>>>>> Oh that reminds me, I spoke of issues with speech in more  current Linux
>>>>> Kernels.  This is because the ability to use a hardware source for speech no
>>>>> longer exists in the new ones, like in Jessie.  Only software speech which for
>>>>> me is out of the question.
>>>>> One can compile the new distribution with the older Kennel, but that. is. way!!!
>>>>> beyond me laughs.
>>>>> Is all this making more sense?
>>>>> Will have those programs for you tomorrow.
>>>>> Thanks for joining the adventure!
>>>>> I likewise thank you Bob for inviting Blaise to the party.
>>>>> Kare
>>>>>
>>>>>
>>>>>
>>>>> On Mon, 14 Dec 2015, Blaise Alleyne wrote:
>>>>>
>>>>>> Hi Karen,
>>>>>>
>>>>>> I don't have very much experience with command line tools for music in
>>>>>> GNU/Linux, barring some pretty simple uses (encoding, monitoring MIDI notes,
>>>>>> making minor edits to sheet music typeset in Lilypond generated by Rosegarden).
>>>>>> But I have taken a head-first dive into audio production in GNU/Linux over the
>>>>>> past few years, and I use it for live performance and for composition and home
>>>>>> recording. I'd also be personally interested in at least some command line
>>>>>> tools
>>>>>> for doing some of the things I currently do in GUIs.
>>>>>>
>>>>>> I don't think I fully understand what you're trying to do (both musically, or
>>>>>> with the SSH component), but hopefully I can be of some assistance. Very
>>>>>> curious
>>>>>> to learn more about what you're hoping to do in case I can help!
>>>>>>
>>>>>> Regards,
>>>>>> Blaise
>>>>>>
>>>>>> ps thanks for pinging me, Bob!
>>>>>>
>>>>>>
>>>>>> On 11/12/15 06:38 PM, Karen Lewellen wrote:
>>>>>>> Wow that must have been amazing.  The Linux audio list at yahoo groups is full
>>>>>>> of people mostly in Europe who use command line Linux tools for their music
>>>>>>> work, some of them are authors of packages included.
>>>>>>> Still it would be fun to witness it firsthand.
>>>>>>> Kare
>>>>>>>
>>>>>>>
>>>>>>> On Fri, 11 Dec 2015, Bob Jonkman wrote:
>>>>>>>
>>>>>>> At Software Freedom Day we had a great presentation on using Free
>>>>>>> Software for Musical Performance by Blaise Alleyne. He was using
>>>>>>> graphical tools for his performance, but has the Music-On-Linux part
>>>>>>> covered. He may have expertise in command-line musical performance
>>>>>>> tools too. I'll ping him with this message.
>>>>>>>
>>>>>>> I wonder what it would take to convince Blaise to haul his studio
>>>>>>> equipment up to the room at Ryerson to do his presentation for
>>>>>>> GTALUG... :)
>>>>>>>
>>>>>>> --Bob.
>>>>>>>
>>>>>>> SFDToronto:
>>>>>>> http://wiki.softwarefreedomday.org/2015/Canada/Toronto/LibrePlanet
>>>>>>>
>>>>>>>
>>>>>>> Bob Jonkman <bjonkman at sobac.com>          Phone: +1-519-635-9413
>>>>>>> SOBAC Microcomputer Services             http://sobac.com/sobac/
>>>>>>> Software   ---   Office & Business Automation   ---   Consulting
>>>>>>> GnuPG Fngrprnt:04F7 742B 8F54 C40A E115 26C2 B912 89B0 D2CC E5EA
>>>>>>>
>>>>>>>
>>>>>>> On 10/12/15 07:17 PM, Karen Lewellen wrote:
>>>>>>>>>> greetings everyone, My name is Karen.  I have only recently found
>>>>>>>>>> this LUG, and joined the talk list. I am a professional singer/
>>>>>>>>>> songwriter and media producer with a unique Linux goal. I wish to
>>>>>>>>>> make use of the varied console or command line based tools in Linux
>>>>>>>>>> for my composition work, and if possible for accessing media
>>>>>>>>>> materials too. I use adaptive technology, which is part of why my
>>>>>>>>>> desire is to ssh telnet into my Linux box rather than work with it
>>>>>>>>>> directly. I tend to prefer more traditional tools.  for example one
>>>>>>>>>> sound card in this box will be an m-audio audiofile 2496
>>>>>>>>>> production card to which I will be attaching my music keyboard. In
>>>>>>>>>> any case while I have some of the tools I require, and firm
>>>>>>>>>> knowledge that my goals are possible. I have not, at least not
>>>>>>>>>> until now, found a Linux users group in Toronto where I can get the
>>>>>>>>>> extra wisdom required. I am very used to Linux and UNIX based shell
>>>>>>>>>> structures. I am using one at this moment  based in freedsb, to
>>>>>>>>>> draft my email, and use another based in Ubuntu for my nonprofit
>>>>>>>>>> media newsroom's hosting account. Such is also part of why I want
>>>>>>>>>> to start with a comparative way to use my Linux box.  I would
>>>>>>>>>> rather begin with a foundation where I can make use of and learn
>>>>>>>>>> more right away, than fail to use the Linux box well at all. Let me
>>>>>>>>>> be sure  my post is suitable before both asking and answering
>>>>>>>>>> questions. Thanks in advance, Karen
>>>>>>>>>>
>>>>>>>>>> --- Talk Mailing List talk at gtalug.org
>>>>>>>>>> http://gtalug.org/mailman/listinfo/talk
>>>>>>>> ---
>>>>>>>> Talk Mailing List
>>>>>>>> talk at gtalug.org
>>>>>>>> http://gtalug.org/mailman/listinfo/talk
>>>>>>>>
>>>>>>> ---
>>>>>>> Talk Mailing List
>>>>>>> talk at gtalug.org
>>>>>>> http://gtalug.org/mailman/listinfo/talk
>>>>>>
>>>>>> ---
>>>>>> Talk Mailing List
>>>>>> talk at gtalug.org
>>>>>> http://gtalug.org/mailman/listinfo/talk
>>>>>>
>>>>> ---
>>>>> Talk Mailing List
>>>>> talk at gtalug.org
>>>>> http://gtalug.org/mailman/listinfo/talk
>>>>
>>>> ---
>>>> Talk Mailing List
>>>> talk at gtalug.org
>>>> http://gtalug.org/mailman/listinfo/talk
>>>>
>>> ---
>>> Talk Mailing List
>>> talk at gtalug.org
>>> http://gtalug.org/mailman/listinfo/talk
>>
>> ---
>> Talk Mailing List
>> talk at gtalug.org
>> http://gtalug.org/mailman/listinfo/talk
>>