Linux setup for audio work

Discussion in 'all things UNIX' started by Gullible Jones, Mar 25, 2014.

Thread Status:
Not open for further replies.
  1. Gullible Jones

    Gullible Jones Registered Member

    Joined:
    May 16, 2013
    Posts:
    1,461
    So my dad does a lot of audio work, and is currently stuck on Windows 2000. This worked for a while, but it is becoming more and more of a pain for him to manage, and the security implications are a serious a problem given his usage patterns. I've been pestering him to upgrade for some time.

    I want to help him switch over to mostly using Linux; probably via dual boot with Win2k at first. (Upgrading to Windows 7 is not really doable ATM.) But it will be necessary to have a working and usable audio setup.

    What is needed:
    - Some semblance of a user-friendly desktop interface. (Not hard.)
    - Automounting of removable devices. (Easily covered in several ways.)
    - File manager that is fast, friendly, and can easily manage huge numbers of rather disorganized files. (There are plenty of orthodox file managers that can handle this, file browsers not so much. Maybe XFE?)
    - Availability of recent versions of Wine, for running some audio programs. (Easy enough, most of the programs are fairly old.)
    - Low latency audio mixing that can cooperate with Wine, preferable using Wine's more stable OSS backend. (This could be a real pain, but I really don't know...)
    - Support for playing MIDI files, and for a MIDI keyboard synthesizer that interfaces with the computer's serial port. With Wine as part of the picture, mind. (I don't know the first thing about this.)

    The current theoretical setup looks like this:
    - Debian Stable
    - IceWM (with Debian menu system)
    - XFE for file management
    - cron-apt for automatic updates
    - AutoFS for mounting the occasional removable volume
    - Google Chrome
    - Wine (probably the latest stable version from WineHQ)
    - VLC, Audacity (already being used on the Win2k install), various other applications
    ... And then the audio stuff comes into the picture, and I have no idea where to even get started.

    Does anyone here do audio work on Linux? After having migrated from XP perhaps? Any suggestions?
     
  2. Ocky

    Ocky Registered Member

    Joined:
    May 6, 2006
    Posts:
    2,677
    Location:
    George, S.Africa
    If you mean applications for converting, transcoding, editing, extracting etc. here are those I use in Scientific Linux. Screenshot taken from Sound&Video Menu. (Most are GTK). WinFF is useful. Sorry for being lazy in just posting a screenshot.

    Audio.png
     
    Last edited: Mar 26, 2014
  3. Gullible Jones

    Gullible Jones Registered Member

    Joined:
    May 16, 2013
    Posts:
    1,461
    I see you're using Pulseaudio, how good is that for low latency stuff? IIRC the big thing used to be JACK, but it was always an astonishing pain to set up...

    Also, is it possible/reasonable to get low latencies with just dmix? Are there any alternative libraries/daemons that can provide low latency mixing?

    Edit: two more questions:

    - Could I get better latency with dmix by using an alternative resampling algorithm, e.g. speex?

    - How important is software mixing anyway? Realtime mixing of streams is not necessary AFAIK, and only one program will ever be playing audio at a given time...
     
    Last edited: Mar 26, 2014
  4. mack_guy911

    mack_guy911 Registered Member

    Joined:
    Mar 21, 2007
    Posts:
    2,677
  5. Gullible Jones

    Gullible Jones Registered Member

    Joined:
    May 16, 2013
    Posts:
    1,461
    ... Okay, it looks like people are not quite understanding me. I know there are plenty of Linux audio applications (and most are not on the menu, because they would take too much time and effort to learn fully... thus Wine).

    What I need to know more about is the audio backend setup.

    (This is also the reason for the weird proposed setup with IceWM, instead of something "normal" like Xfce. Having a dozen daemons polling stuff in the background is very convenient and all, but I doubt it can be good for latency on a multitasking desktop OS.)
     
  6. Ocky

    Ocky Registered Member

    Joined:
    May 6, 2006
    Posts:
    2,677
    Location:
    George, S.Africa
  7. 0strodamus

    0strodamus Registered Member

    Joined:
    Aug 23, 2009
    Posts:
    1,047
    Location:
    United Surveillance States
    Something else to check is that his current audio card is supported by ALSA (or whatever you decide to use). When I switched to Linux, mine wasn't so I had to switch to another card.
     
  8. Gullible Jones

    Gullible Jones Registered Member

    Joined:
    May 16, 2013
    Posts:
    1,461
    It is, ALSA support for both the PCI Soundblaster and the onboard audio is very good.
     
  9. 0strodamus

    0strodamus Registered Member

    Joined:
    Aug 23, 2009
    Posts:
    1,047
    Location:
    United Surveillance States
    That's good. I can't help you with low latency stuff, but hope you find something that works for you.
     
  10. Gullible Jones

    Gullible Jones Registered Member

    Joined:
    May 16, 2013
    Posts:
    1,461
    So I did a little testing with zynaddsubfx, a small software synthesizer with a rather opaque interface.

    I could play around with this for quite a while (and figure out precious little). But basically it looks like latency will not be an issue for native applications, even when using just ALSA and dmix... And at that, even when using libsamplerate or speexrate plugins at the highest quality setting.

    My main concern now is Wine:
    a) the latency it introduces
    b) the 16 bit maximum for audio sampling

    The latter in particular seems problematic, because I can easily hear the difference between 16 bit and 24 bit in media players. Maybe I should see if I can find native apps to replace some of the old Windows ones.
     
  11. amarildojr

    amarildojr Registered Member

    Joined:
    Aug 8, 2013
    Posts:
    1,988
    Location:
    Brasil
    I'm a musician with +10 years on my back, and my current setup is:

    Arch + alsa-utils + pulseaudio + pavucontrol + pulseaudio-alsa + ardour + audacity + tuxguitar.

    Never been happier.
    Latency is not a problem. For input monitoring I just unmute my Mic channel on alsa-mixer and record listening to it + the channels already recorded.
     
  12. keithpeter

    keithpeter Registered Member

    Joined:
    Jan 4, 2014
    Posts:
    45
    Location:
    UK
    A computer that came with the client version Windows 2000 is going to be somewhat long in the tooth now I suspect.

    Do people here think that there would be any advantage in dedicating the machine to music production and disabling networking? Just using a recycled laptop or tablet for Internet? I'm thinking of saving interupts.

    Ignore that if it is some huge Xeon based workstation!

    The page at

    http://wiki.linuxaudio.org/wiki/system_configuration

    might help, and a post to Debian Forums / Desktop and Multimedia might get some concrete suggestions about sound packages for the set up that was proposed. The packages mentioned by amarildojr are all available in Debian repos. I think that Ardour may need the jack audio daemon installed - or at least it did some years ago when I had to help out with some audio stuff.
     
  13. amarildojr

    amarildojr Registered Member

    Joined:
    Aug 8, 2013
    Posts:
    1,988
    Location:
    Brasil
    Is there any specific reason you don't want internet? I mean, software on Linux updates fast, so it's nice to have a connection.
    Regarding your question, I don't think there are any disadvantages in making a dedicated sound PC if that's how you make your living. But you gotta keep it up to date with both software and hardware, and good quality hardware too. So I don't know if a laptop will do it.[/QUOTE]
     
Loading...
Thread Status:
Not open for further replies.