William G. Unruh | Canadian Institute for| Tel: +1(604)822-3273
Physics&Astronomy | Advanced Research | Fax: +1(604)822-5324
UBC, Vancouver,BC | Program in Cosmology | unruh@xxxxxxxxxxxxxx
Canada V6T 1Z1 | and Gravity | www.theory.physics.ubc.ca/
On Fri, 14 Nov 2014, Alan McConnell wrote:
> On Fri, Nov 14, 2014 at 10:29:50AM -0800, Bill Unruh wrote:
>> JUst to clarify a bit. pulseaudio sits on top of alsa-- ie it needs alsa (or
>> oss) to be running properly in order for it to work. It takes audio streams
>> from the programs and mixes them together and sends them off to the alsa
>> drivers, etc (at least when it is working properly).
> An interesting description. What do the audio streams
> consist of? Digital data, for sure; but what kind? I
> like "mixes them together". Now I'm not being hostile
> here; I'm just pointing out that your discussion of what
> is going on lacks specificity.
If you want to know exactly what goes on you will need to read the source. As
far as I know pulseaudion has no document that details the exact way in which
it works, etc.
>
> What I _am_ inferring is that pulseaudio gets the data
> in some form, and gives them to ALSA, which is in
> charge of the cards, which are attached to the
> speakers(through the hole/pin arrangement in the back
> of my machine) ALSA formalizes this process through
> the "driver" mechanism(everything is a file!) that we
> all know, and some of us understand(not I, too
> well<g>)
On Unix, everything is a "file" or stream.
>
>> This means that if, for some reason, alsa is not working
>> properly, pulseaudion cannot work. But is pulseaudio is not
>> working, then also can still play the music as long as the
>> program is set up to send the music to alsa.
> Which program? The magic seems to be: whether I get
Whatever program you are using to play sounds.
> sound signals from a youtube video, or from a midi
> file that my Lilypond has created, or from an oggvorbis
> file, or whatever, still sound comes out of my
> speakers nicely. Obviously a complicated process.
>
>
>
>
>> On Fri, 14 Nov 2014, Alan McConnell wrote:
>>
>>> pulseaudio. From my "research"(aka searching helplessly
>>> around on the Internet) I discovered that ALSA needs
>>> pulseaudio to function. So I ran the command
>>> pulseaudio --start
>>> and got an error message. "well, that didn't work!"
>>
>> I am sure that was not the error message.
> <LOL> No, it wasn't! Although unfortunately very many
> of my thoughts are error messages to myself.
>
>> Had you told us what it was we might have been able to explain
>> to you want it was trying to say.
> Here it is:
> E: [pulseaudio] authkey.c: Failed to truncate cookie file: Invalid argument
> W: [pulseaudio] authkey.c: Failed to load authorization key
> +'/home/alan/.config/pulse/cookie': Invalid argument
> What do you make of it? If you had received such a daunting
> message, would you have anticipated that your command
> nevertheless went through? For your information: I had
> previously created the pulse directory in ~/.config and
> had put an empty file, named 'cookie', in it.<G>
And clearly an empty file is not a valid cookie file. You might want to try
removing it althogether. I suspect but do not know that when pulseaudio finds
a cookie file it assumes that authentication is required.
If you look up the documentation of pulseaudio you might find something
describing such files. Or maybe not. pulseaudio was developed by the same
redhat developers that developed systemd, which is also not know for extensive
comprehensible documentation. But I will not look for it.
>
>
>
>
>> Sound is still one of the areas of Linux which is pretty messed
>> up, in part because of the vast vast diversity of sound
>> hardware-- there are hundreds of companies out there making
>> sound cards, and each thinks they have to compete by making
>> their sound card's control more obscure than their
>> competitor. pulseaudion was supposed to unify the situation. It
>> is fine sometimes but if there is any problem lower down it is
>> terrible and hides rather than reveals those problems.
> Interesting. In view of the problems I indicated above:
> analog/digital, huge number of formats, etc etc, I think
All computer sound is digital.
> that Linux and ALSA have done an impressive job. Of
> course, very intelligent people have been working on
> computer sound problems for decades; Wikipedia tells me
> that the ALSA project was started in 1998.
And before that OSS. Yes, the task is daunting, esp because of the HUGE number
of sound cards available. On windows, MS does not bother to try to drive them.
That is the job of the manufacturer. Linux does not have that luxury, and the
work done by the developers and supporters of ALSA is truely impressive.
However, their main interest has been technical, and a good user interface has
not be high on their list of priorities.
>
>>> Alsomixer, also amixer, work fine now. I know I can turn the
>> Alsamixer?
> It is a nice ncurses program. Don't you have it on your
> system, Mr Unruh?
It was the spelling.
>
>
>>> work? It is of course no problem to run pulseaudio --start
>>> by hand, and I can probably put this command into my
>>> .bashrc, but still . . .
>>
>> You need to have an entry in /etc/rc?.d where the ? refers to the runlevel you
>> run at, which starts with S, has two digits (listed at the beginning of the
>> init.d/pulseaudio file) and then the name pulseaudio. It is a link to
>> /etc/init.d/pulseaudio. Eg, ln -s /etc/init.d/pulseaudio
>> /etc/rc3.d/S75pulseaudio
>> (I do not know that those are the right numbers for Debian-- I do not use
>> Debian)
> If this elaborate process is my alternative to putting the
> command into my .bashrc, you know which I'll choose!
That's elaborate?
>
>
>>> Just for curiosity's sake: what other Linux program
>>> could I substitute for pulseaudio? Someday, when I've
>>> picked myself up off the floor, I might feel like playing
>>> with it. I am a musician, play violin and viola, so
>>> the temptation is there.
>>
>> Jack? Or just the "bare" alsa.
> Jack is mentioned in the Wikipedia article on ALSA. My
> Debian Wheezy disks seem to have it. Thanks for the
> information!
It is far more elaborate than pulseaudio, but also allows much more control.
>
>
>> William G. Unruh | Canadian Institute for| Tel: +1(604)822-3273
>> Physics&Astronomy | Advanced Research | Fax: +1(604)822-5324
>> UBC, Vancouver,BC | Program in Cosmology | unruh@xxxxxxxxxxxxxx
>> Canada V6T 1Z1 | and Gravity | www.theory.physics.ubc.ca/
> I have visited a friend in your lovely city. Are you a
> believer in "dark energy"? I ask as an interested amateur;
Believer? The evidence appears to be that something with an effective energy
momentum tensor with positive energy but negative pressure seems to exist-- to
be driving the universe. I do not particularly like it ("Who orderd that") but
then again there is no reason why nature should listen to me.
> my professional training and my career were in mathematics.
>
> Best wishes to all,
>
> Alan
>
> --
> Alan McConnell : http://globaltap.com/~alan/
> No one minds what Jeffreys says . . it is not more than a week ago
> that I heard him speak disrespectfully of the Equator.(Sydney Smith)
>
------------------------------------------------------------------------------
Comprehensive Server Monitoring with Site24x7.
Monitor 10 servers for $9/Month.
Get alerted through email, SMS, voice calls or mobile push notifications.
Take corrective actions from your mobile device.
http://pubads.g.doubleclick.net/gampad/clk?id=154624111&iu=/4140/ostg.clktrk
_______________________________________________
Alsa-user mailing list
Alsa-user@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/alsa-user
