Audacity Forum

steve wrote:What was the question? Have we answered it?

At this time I forget what the question was - the resulting discussion has been so interesting and informative that I no longer really care! But to answer your question: "have you answered my question"? - I have to observe that this is Linux we are talking about here. Not only just Linux, but sound processing within Linux; how can there be 'a' question? Honestly, I've been trying to understand it since first installing Fedora 4 - and failing comprehensively. I've learned more in this thread than in the previous 5 or 6 years. I also learned that almost every answer to a Linux sound question uncovers further questions I need to pose.

This is a critical flaw in the 'community support' concept for Linux - it means those who have the attitude, personality and ability to provide advice and guidance find themselves inundated by demands from the ignorami, who have a desire to get problems solved (if not even to learn how to solve them themselves). Talking of which, I observe that there is no framework within the Linux world for well-structured, effective and logical problem source identification procedures - what used to be called MAPs 30 or 40 years ago. Everybody seems to use an ad-hoc process. To use a modern expression (whose real meaning is totally hidden from me ), this approach 'does me 'ead in'.

Now to return focus to your last post (no pun intended), having 'blacklisted' the VIA 82xx modules, "cat /proc/asound/modules" returns: where the usb device is the mic in a Logitech 4000 web-cam, which is not supported under Ubuntu, but is under WinXP. I don't understand why it carries 'index' 2, rather than 1. Maybe this is a setting in a config file somewhere, with 'index' 1 being used for the (now blacklisted) VIA 82xx hardware.

PulseAudio Volume Control now shows my Input Devices as a QuickCam Pro 4000 Analog mono (which I completely understand) and a single 'Internal Audio Digital Stereo (ICE958), whose description I fail to understand, based on what ALAS sees: i.e. iCE1712. But I assume that this 'Internal Audio' is my M_Audio card.

PAVC lists my output device as 'Internal Audio Analog Stereo'. It further lists a set of Profiles under the 'Configuration' tab for my 'Internal Audio'. None of these profiles match my expectation, which is Duplex Analog Stereo Output/Input. The closest is Analog Stereo Out + Digital Stereo (IEC958) Input. However, it seems to produce the effect I want in Audacity, so I guess I should stop moaning about it.

<System><Preferences><Sound> information is now directly comparable to that of Pulse Audio Volume Controller

So I think I'm now running with a stable set of Audacity preferences. Just to check, I've just opened Audacity and - voila - it shows both playback and recording devices set to 'default' - which, in my experience, means no recording and no playback. And this is confirmed in alsamixer - it also shows the sound card selection as 'default', without stating what that default is. It certainly isn't set to the M Audio card, which is where I last set it and I certainly haven't changed it.

As I look around at the squares on the floor, this one is marked with the number '1'..........

I'm not sure how to answer your problem... Your available options are rather awkward...
Digital input is probably not line-in nor mic-in (those should be analog inputs).

Try quiting pulseaudio and run audacity on alsa only, see if you get more options.

To run alsamixer on a different card type alsamixer -c <index> where <index> is the number you get on /proc/asound/modules
For example, to run alsamixer on the usb audio type: alsamixer -c 2

What options do you get on alsamixer for your card? Specially on the capture screen (I'm not on linux right now but I think it's F4 to switch to capture.. it should be stated in the alsamixer interface, you can also tell it from the command to run on capture mode, type alsamixer -h to see the available options). Anyway see if it shows anything else beside the digital input option.

bgravato wrote: To run alsamixer on a different card type alsamixer -c <index> where <index> is the number you get on /proc/asound/modules
For example, to run alsamixer on the usb audio type: alsamixer -c 2

What options do you get on alsamixer for your card? .... Anyway see if it shows anything else beside the digital input option.

After a shutdown/restart sequence, the out put from /proc/asound/modules now has the USB audio (logitech Webcam 4000) on index 1. Running alsamixer against this device show only 1 'option' for capture, labelled 'mic'. That's as I expect.

Running 'alsamixer -c 0' (M-Audio ICE1712 card) however gives a result which I don't fully understand. The capture 'options' shown are (reading left to right):

S/PDIF Multi
S/PDIF Multi 1
DAC
DAC1
H/W Multi
H/W Multi 1
Multi Track Volume

All 'devices' (if that is what they are) are set at about 80 on their 'thermometers'.

On this same card the Playback 'options' are labelled:

H/W 1 (also has the text 'PCM Out' in place of a thermometer)
Multi
Multi 1
Multi 2
Multi 3
Multi 4
Multi 5
Multi 6
Multi 7
Multi 8
Multi 9 (All 10 Multis are set with unmute, at volume value = 80)
Multi Track Internal Clock (4410) (value = 44100; cannot be changed)
Multi Track Internal Clock Default (4410) (value = 44100; cannot be changed)
Multi Track Rate Locking (value = 00)
Multi Track Rate Rest (value = 00)
Multi Track Volume Rate (value = 80; cannot be muted)


I presume that the 10 Multis come from the line in the associated /usr/share/alsa/cards/ICE1712.conf file, which mentions 10 slave.channels.

I have no idea what any of this stuff does. Interestingly, there is no mention of 'IEC958' in the alsamixer output.

bgravato wrote:Try quiting pulseaudio

I'm not sure how to do that on a default Ubuntu 10 installation as PulseAudio is set to "auto-spawn" - it will automatically restart as soon as you use sound.
However, Audacity should be able to use ALSA directly (without using PulseAudio), by selecting the Recording and Playback devices in Preferences > Devices ~(rather than selecting "default").

I suspect all of those "Multi" outputs are because the M-Audio card digital I/O supports surround-encoded AC-3 and DTS pass-through.

Rincewind, can we take a step back here - leaving Audacity out of the equation for a moment, are you able to use the M-Audio card with other applications?
If so, how do you do that?

steve wrote:Rincewind, can we take a step back here - leaving Audacity out of the equation for a moment, are you able to use the M-Audio card with other applications?
If so, how do you do that?

Yes, I can play sound through the M-Audio card from other applications. This confident claim has only 1 tangible item of supporting evidence: the speakers on my system are connected only to the analog outputs on the M Audio card. In this configuration I hear the expected sounds when playing an MP3 file (produced from a WAV captured by Audacity) through the following apps: (these are are running as I type):

RythmBox
Gnome MPlayer
VLC Media Player
Exaile Music Player
Audacious

Now that I have blacklisted the modules for VT82xx the choice of hardware for playback is limited to 1, so I don't have to do anything, having set the profiles for that device in Pulse Audio Device Chooser/Volume Control.

Prior to your guidance about using Pulse Audio I had no idea how to set the correct hardware in any of these apps. aside from blundering around in alsamixer. until finally I got some sound, somewhere - but without being able to form a convincing correlation between what I did and what worked.

I recall spending almost 6 months trying to get recording and playback both working in Audcaity when I used Fedora and a sound blaster audigy card. (It never worked), so now I know explicitly how to choose and set the sound hardware in that application. For all other applications, there does not, in general, seem to be a function in which I can say 'use this device for output (or input, as appropriate).

Aside from some vague concept, derived from thinking from first principles, that there must be a single point in the system where one defines the sound hardware to be used - and deciding that <System><Preferences><Sound> (in Ubuntu 10.10 at least) was that point - I really have no idea how apps. 'choose' a sound hardware device.

If you play something with RhythmBox, by default it uses PulseAudio, so if you then open PAVU you should see RhythmBox listed on the Playback tab.
Similarly, if Audacity is set to playback using "default" as the Playback device, then when you start playing something in Audacity you should see Audacity listed in the playback tab of PAVU. Note: it will only show up during playback, or paused playback and will disappear from the PAVU when you press stop.
While it is visible in the PAVU, you can set PulseAudio to send the audio output to any available output.

Rincewind wrote:I really have no idea how apps. 'choose' a sound hardware device.

Most apps (if not all) allow you to select the playing device.
How you do that will vary from app to app...
For example, on VLC, go to Preferences -> Audio -> Output

bgravato wrote:

Rincewind wrote:I really have no idea how apps. 'choose' a sound hardware device.

Most apps (if not all) allow you to select the playing device.
How you do that will vary from app to app...
For example, on VLC, go to Preferences -> Audio -> Output

Hmm, not exactly - not in the way I mean. For example in VLC I have the choice between 'PulseAudio' and ALSA (and other things). None of these are 'devices', like 'M Audio Delta 24/96' - which the average homo-(partially)-sapiens in the average street would recognise - are they?

It really does seem to me that there is some sort of hair-shirt macho philosophy in the Linux world to make things as difficult as possible, in the circumstances, to understand.

Anyway, I'm a bit off topic here - this is an Audacity forum and I should try to confine my bigotry to that application. Perhaps this thread has run its course now.

Rincewind wrote: in VLC I have the choice between 'PulseAudio' and ALSA (and other things). None of these are 'devices', like 'M Audio Delta 24/96' - which the average homo-(partially)-sapiens in the average street would recognise - are they?

And on Windows you might have the choice of WDM, Direct Sound, WASAPI or ASIO. They don't sound like 'M Audio Delta 24/96' either.
If everything works straight away with the default settings, then it's no problem, whatever platform you are on. If it doesn't, then it usually means a headache, whichever platform you are on.

The sound system on Linux is developing rapidly to make things work "straight out of the box", and when I did a clean install of Ubuntu 10.10, it did work straight out of the box. Unfortunately you were not so lucky (which is why we're having this conversation )

Rincewind wrote:Perhaps this thread has run its course now.

I was hoping that we'd end with you saying "Great it's all working now".
Are we at that stage?