You are not logged in.

#1 2008-03-14 13:50:10

SpookyET
Member
Registered: 2008-01-27
Posts: 410

The Sorry State of Sound In Linux

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



READ: http://insanecoding.blogspot.com/2007/0 … linux.html

I've only discovered this because of an ALSA bug was making my life miserable.  That article was written about a month or two before OSS4 (Open Sound System v4) was released under GPL and CDDL. This past January, it was released under BSD as well. Unfortunately for the developer, he does not understand open source very well. He is now reporting that his revenue went significantly down..

OSS4 does not equal to the crap OSS in Linux 2.4 kernels. Even OSS3 != OSS3 in former linux kernels. He was neglecting the open source version in favour of the commercial one. Instead of improving it, someone forked it. And, the fork became popular. Unfortunately, the original author had no interest in working on the fork. So, he only focused on the commercial version.  Now, many years later, the commercial version is fully open source. There is no more commercial version.

I've tried it,  and I really like it. Music sounds much better than on ALSA. For example, screaming in alternative rock is more legible.  It's like enabling "enhance voice" in a phone. It supports higher PCM values without noise.

The mixer (vmix) is capable of 18 channels. It can also do per application volume control (like PulseAudio).

It totally makes sense that OSS gets back into the kernel because it works on almost every UNIX and UNIX-like system (except OS X). ALSA only works on Linux, and according to that article, developers still prefer the OSS API, even on ALSA. However, the ALSA OSS API is lacking according to an ALSA developer. I can confirm that the comment on the ALSA OSS Emulation API working better than the ALSA API. XMMS with ALSA enabled freezes my system. XMMS with OSS enabled on ALSA does not.

It worked without any configuration besides muting some channels to kill the noise.

THE BAD

ossmix and ossxmix are totally unusable because they do not name the channels properly. ossxmix uses 100% CPU if Compiz is enabled. Version 4.1 will fix this.
I had to figure out WTF ossxmix.codec1.connector.jack14.jack 54:54 means. Sane names need to be added to the jacks such as "mater, front, input, auxiliary, microphone, etc.."
It's best to use ossxmix (GTK+ mixer) and playing with all the jacks to figure out what each one does.
One interesting control is how OSS4 should behave (Fast, Medium High, Professional, etc.). It's probably a latency control.  ossxmix is a demo app on how you can control the mixer from GTK. It needs to be made usable.

If you use Media Player Daemon, VLC, and MPlayer, OSS4 works beautifully. They use OSS directly. Others have problems.

Totem does not play sound. Totem-Xine uses 100% CPU (Xine-UI works with low CPU usage).
In terminal I noticed some output from oss mixer control saying it received bad arguments. OSS4 is backwards compatible with OSS3, but xine-lib may be using the API badly.

The progress bar (seeker) in GStreamer based applications (Rhythmbox, Banshee) does not work if you use vmix, or it may work, but it will be unsure of the length of a song. You will see it constantly adjusting a few seconds up or down. By the middle of the song, the length reporting generally stabilises. You have to enable softoss (the old mixer). vmix and Gstreamer behave very badly. The Gstreamer developer responsible for OSS claims that it does on his system. Though, he is most likely using the trunk version.

You have to apply this patched gstreamer to make volume control in GNOME work.
While the progress bar problem is fixed with softoss, volume control does not work, even with the patched GStreamer. It only works with vmix.  System > Preferences > Sound lists nothing under "Default Mixer Tracks".

Sound does not work in KDE4 at all. While Phonon is based on Xine, which has OSS support, it queries HAL, which does not support OSS4 yet. Therefore, it thinks that there is no sound card.

As for KDE 3.5.9. Amarok works, Noatun works, and anything based on MPlayer should work. KMIX does not work. Same in KDE4.

DOWNLOAD IT (Popular Distributions Linux and Unix Distributions)
Gentoo ebuild
Arch Linux

The Arch Linux wiki has a good article. Read it.
A Ubuntu user has published an install guide as well.
Check Configuring Applications for OSSv4 after you install it.

Overall, I think it's a million times better than ALSA. It's cross platform, and it's stable. It was released a year ago on 15th of March 2007. Since he does not have a marketing department, no one has heard of it.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (GNU/Linux)

iEYEARECAAYFAkfuWboACgkQQxhfy6QbjiQbdwCgzHLsXO9hg7j4i9VJH0N9AWtW
UTEAoLkOsDfmKjYJ/u5Nv8i04PUBCN7M
=6qYu
-----END PGP SIGNATURE-----

Last edited by SpookyET (2008-03-29 15:01:19)

Offline

#2 2008-03-14 14:06:22

fwojciec
Member
Registered: 2007-05-20
Posts: 1,411

Re: The Sorry State of Sound In Linux

Your soundcard must have some peculiar issues with alsa...

I have alsa properly set up on two computers (laptop and desktop).  The desktop is a transport for a fairly high quality headphone sound system (with a separate dac and a headphone amplifier) and all I can say that alsa works perfectly well for me -- I've never experienced any problems you're attributing to it generally, and I've been, in fact, impressed with how easy it is to achieve things like bit perfect digital output and such with it -- stuff you need for high quality music reproduction (and things you need workarounds and hacks for under windows, for example).

You probably should qualify the statements like "Music sounds much better than on ALSA" a bit -- because they are most definitely referring to how alsa is set up on your system not to alsa in general.

Otherwise -- an interesting read.  I'll have to try this new OSS4 one of these days and see for myself.

Offline

#3 2008-03-14 14:34:52

.:B:.
Forum Fellow
Registered: 2006-11-26
Posts: 5,819
Website

Re: The Sorry State of Sound In Linux

I just installed OSS 4 on my laptop, and the positive 'reviews' have made me eager to test. I am wondering though if it would be possible to strip out the GTK+ 2 dependency (and interface)? Surely there has to be another way to interface with the volume stuff... I don't want to install GTK+ 2 and all the deps on my headless server just to have better sound.

Last edited by B (2008-03-14 14:36:32)


Got Leenucks? :: Arch: Power in simplicity :: Get Counted! Registered Linux User #392717 :: Blog thingy

Offline

#4 2008-03-14 14:43:53

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

B wrote:

I just installed OSS 4 on my laptop, and the positive 'reviews' have made me eager to test. I am wondering though if it would be possible to strip out the GTK+ 2 dependency (and interface)? Surely there has to be another way to interface with the volume stuff... I don't want to install GTK+ 2 and all the deps on my headless server just to have better sound.

ossmix can control it just as well as ossxmix. Just modify the scripts on the wiki page.

Offline

#5 2008-03-14 17:18:27

wyvern
Member
From: 'Old' Scottish/English border.
Registered: 2008-01-06
Posts: 70

Re: The Sorry State of Sound In Linux

Thanks for bringing this to my attention. I had actually heard of this, but back when it was proprietary - hadn't heard about it becoming open source.

I have just installed, configured and tested OSS4 and I think it's perfect (well, apart from one small issue). I think it does make a difference with regards to sound quality (and I can *finally* use my headphones now too with my HD Audio card <3 ).

I noticed in gstreamer apps (Quod Libet) that it no longer updates the progress bar, though pausing and unpausing will make it show the correct progress roll Other than that Quod Libet is as usable as normal. I run without any patched versions of Gstreamer though.

However, using Openbox/Xcompmgr there ae no noticable slowdowns or graphical issues for me. I also use MOC instead of Quod Libet, so the updating progress bar doesn't bother me smile I also haven't tried xine - I prefer mplayer. For me the gtk mixer uses no more than 13% cpu (dual core intel @1.66Ghz). I run with the (latency?) setting at 'medium'.

I edited my copies of the example scripts on the wiki to use my preferred mixer, since I have a united 'vmix0-vol' present. So much easier than the n:n values for me:D Mute/raise/lower volume all work fine using those. If anyone needs them I can post up my altered copies smile

I really recommend people at least give OSS4 a try, it could be what you've been looking for (with a little more tweaking than alsa perhaps needs).

Offline

#6 2008-03-14 18:38:33

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

wyvern wrote:

Thanks for bringing this to my attention. I had actually heard of this, but back when it was proprietary - hadn't heard about it becoming open source.

You are welcome

I noticed in gstreamer apps (Quod Libet) that it no longer updates the progress bar, though pausing and unpausing will make it show the correct progress roll Other than that Quod Libet is as usable as normal. I run without any patched versions of Gstreamer though.

I mentioned that. The gstreamer dev responsible for OSS claims that this not occurs for him. It may have been fixed in the latest gstreamer. Changing the version numbers in the ABS PKBUILDs could confirm that.

However, using Openbox/Xcompmgr there ae no noticable slowdowns or graphical issues for me. I also use MOC instead of Quod Libet, so the updating progress bar doesn't bother me smile I also haven't tried xine - I prefer mplayer. For me the gtk mixer uses no more than 13% cpu (dual core intel @1.66Ghz). I run with the (latency?) setting at 'medium'.

The progress bar problem affects players differently. Totem won't play any sound. Banshee takes ages to switch between songs.

I edited my copies of the example scripts on the wiki to use my preferred mixer, since I have a united 'vmix0-vol' present. So much easier than the n:n values for me:D Mute/raise/lower volume all work fine using those. If anyone needs them I can post up my altered copies smile

I want those scripts. Maybe if I provide you with my ossmix output, you can figure out all those jacks are for me. I'm having a hard time turning off microphone and inputs to kill remaining white noise.

Offline

#7 2008-03-14 18:50:35

wyvern
Member
From: 'Old' Scottish/English border.
Registered: 2008-01-06
Posts: 70

Re: The Sorry State of Sound In Linux

Posted up on file bin the raisevolume lowervolume and mute scripts for you.

I'll pop back later and have a look at your ossmix output - just grabbing some food at the minute smile

Oh, and I was just confirming the gstreamer issue - didn't mean to imply I had found something new smile Should have made myself clearer.

Offline

#8 2008-03-14 19:14:00

NoOneImportant
Member
From: Deep Southern California
Registered: 2007-02-13
Posts: 178

Re: The Sorry State of Sound In Linux

the cpu problem with ossxmix has already been fixed (and it has a tray icon) and will most likely be included in the next oss4 release (it's currently in the latest 4.1 testing source available by 4-front)

Offline

#9 2008-03-14 19:34:08

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

Some people have problems compiling it if ALSA 1.0.16 is installed. Check the sums after downloading.

oss-linux-free-4.0_1014-3-i686.pkg.tar.gz

MD5: 6e2c4a37daaf362d187ca2ff151b541c
SHA1: 3d1cde0a604abf4e9b2b8c0ef4eb350224fff777

Last edited by SpookyET (2008-03-14 20:10:51)

Offline

#10 2008-03-14 19:39:29

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

NoOneImportant wrote:

the cpu problem with ossxmix has already been fixed (and it has a tray icon) and will most likely be included in the next oss4 release (it's currently in the latest 4.1 testing source available by 4-front)

What good does it do for me if it's unreadable? They need to name the controls with Master, Front, Mic, etc.

Offline

#11 2008-03-14 19:41:06

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

wyvern wrote:

Posted up on file bin the raisevolume lowervolume and mute scripts for you.

I'll pop back later and have a look at your ossmix output - just grabbing some food at the minute smile

Oh, and I was just confirming the gstreamer issue - didn't mean to imply I had found something new smile Should have made myself clearer.

I switched to ALSA 1.0.16 to see if that fixes my problem. It did not. So, now I have to reinstall OSS4. However, I'm trying to help irc://irc.freenode.net/#alsa to diagnose the problem.

Offline

#12 2008-03-14 20:57:58

wyvern
Member
From: 'Old' Scottish/English border.
Registered: 2008-01-06
Posts: 70

Re: The Sorry State of Sound In Linux

SpookyET wrote:

I switched to ALSA 1.0.16 to see if that fixes my problem. It did not. So, now I have to reinstall OSS4. However, I'm trying to help irc://irc.freenode.net/#alsa to diagnose the problem.

Well, if either driver can fix your problems, then it doesn't matter which you use, does it? smile Good luck.

If you decide to stick with OSSv4 and I can help diagnose the ossmix output, I'll be around smile

And thanks again for pointing it out - for me OSSv4 is wonderful.

Offline

#13 2008-03-14 21:09:24

lucke
Member
From: Poland
Registered: 2004-11-30
Posts: 4,018

Re: The Sorry State of Sound In Linux

I've been using OSS for some time (it supports headphone output in M-Audio Revolution 5.1, alsa doesn't) - until one version introduced cracking. They don't seem to make other-than-the-latest versions available for download - and that blows.

Last edited by lucke (2008-03-14 21:12:17)

Offline

#14 2008-03-14 21:17:36

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

lucke wrote:

I've been using OSS for some time (it supports headphone output in M-Audio Revolution 5.1, alsa doesn't) - until one version introduced cracking. They don't seem to make other-than-the-latest versions available for download - and that blows.

Sounds like a perfect match with, "The Arch Way":D

Offline

#15 2008-03-14 23:46:49

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

Fix for the progress bar problem in gstreamer.

Disable vmix and enable softoss.

I didn't do that part of the wiki because I don't care for effects, but apparently it fixes it.

Sound in totem still does not work. Sound in totem-xine does not work as well. It was probably not built with oss support.

Sound in xine-ui works. No more 100% CPU after following the OSS4 Wiki.

Last edited by SpookyET (2008-03-15 00:53:10)

Offline

#16 2008-03-14 23:56:00

wyvern
Member
From: 'Old' Scottish/English border.
Registered: 2008-01-06
Posts: 70

Re: The Sorry State of Sound In Linux

Hmm, so it's a choice - have to enable effects and use an older mixer, or put up with a 'stuck' progress bar... Well, as I said, I use MOC, and I'm happy with my mixer setup so far, so I'll stick to what I've got.

I can't believe I finally have my headphones working, or external speakers, should I so wish - and what with the ATI divers *finally* providing tear-free video, it's like it's my birthday lol

Offline

#17 2008-03-15 01:13:17

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

wyvern wrote:

Hmm, so it's a choice - have to enable effects and use an older mixer, or put up with a 'stuck' progress bar... Well, as I said, I use MOC, and I'm happy with my mixer setup so far, so I'll stick to what I've got.

I can't believe I finally have my headphones working, or external speakers, should I so wish - and what with the ATI divers *finally* providing tear-free video, it's like it's my birthday lol

Since you mentioned ATI and went offtopic, can you pastebin your ATI xorg? It works fine for me, but If I enable compiz, i get flickering video and slow scrolling. So, I'm stuck with the open source driver, which uses a lot more CPU when I play video.

I can also confirm what the article said about the ALSA OSS emulation. It does work better than ALSA with the ALSA API. xmms with ALSA enabled freezes my sytem. xmms with OSS enabled running ALSA does not. I'm tired of this bollocks. We have to mount a campaign to get OSS back into the kernel.

Anyway, this is my OSS output.

OSSINFO

Version info: OSS 4.0 (b1014/200803130443) (0x00040003) GPL
Platform: Linux/i686 2.6.24-ARCH #1 SMP PREEMPT Wed Mar 5 12:07:52 UTC 2008 (mercurius)

Number of audio devices:    10
Number of audio engines:    13
Number of mixer devices:    2


Device objects
 0: osscore0 OSS core services
 1: hdaudio0 Intel HD Audio interrupts=1389433 (1526696)
    HD Audio controller Intel HD Audio
    Vendor ID    0x80862668
    Subvendor ID 0x10250070
     Codec  0: ALC880 (0x10ec0880/0x08800000)
     Codec  1: Unknown (0x11c13026)
 2: softoss0 OSS Virtual Mixer v3.0


Mixer devices
 0: High Definition Audio ALC880 (Mixer 0 of device object 1)
    Device file /dev/oss/hdaudio0/mix0, Legacy device /dev/mixer0
    Priority: 10
    Caps: 
    Device handle: PCI00701025-0000:00:1b.0-mx01
    Device priority: 10

 1: Virtual Mixer (Mixer 0 of device object 2)
    Device file /dev/oss/softoss0/mix0, Legacy device /dev/mixer1
    Priority: 1
    Caps: VIRTUAL 
    Device handle: softoss0-mx01
    Device priority: 1


Audio devices
HD Audio front                    /dev/oss/hdaudio0/pcm0  (device index 0)
    Legacy device /dev/dsp0
    Caps: TRIGGER MMAP 
    Modes: OUTPUT 
      Out engine  1: 0/HD Audio front
                     Available for use 
    Input formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Output formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Device handle: PCI00701025-0000:00:1b.0-au01
    Related mixer dev: 0
    Sample rate source: 0
    Preferred channel configuration: Not indicated
    Supported number of channels (min - max): 2 - 8
    Native sample rates (min - max): 44100 - 192000 (44100,48000,96000,192000)
    HW Type: Not indicated.
    Minimum latency: Not indicated

HD Audio rear                     /dev/oss/hdaudio0/pcm1  (device index 1)
    Legacy device /dev/dsp1
    Caps: TRIGGER MMAP 
    Modes: OUTPUT 
      Out engine  1: 1/HD Audio rear
                     Available for use 
    Input formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Output formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Device handle: PCI00701025-0000:00:1b.0-au02
    Related mixer dev: 0
    Sample rate source: 0
    Preferred channel configuration: Not indicated
    Supported number of channels (min - max): 2 - 2
    Native sample rates (min - max): 44100 - 192000 (44100,48000,96000,192000)
    HW Type: Not indicated.
    Minimum latency: Not indicated

HD Audio center/LFE               /dev/oss/hdaudio0/pcm2  (device index 2)
    Legacy device /dev/dsp2
    Caps: TRIGGER MMAP 
    Modes: OUTPUT 
      Out engine  1: 2/HD Audio center/LFE
                     Available for use 
    Input formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Output formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Device handle: PCI00701025-0000:00:1b.0-au03
    Related mixer dev: 0
    Sample rate source: 0
    Preferred channel configuration: Not indicated
    Supported number of channels (min - max): 2 - 2
    Native sample rates (min - max): 44100 - 96000 (44100,48000,96000)
    HW Type: Not indicated.
    Minimum latency: Not indicated

HD Audio side                     /dev/oss/hdaudio0/pcm3  (device index 3)
    Legacy device /dev/dsp3
    Caps: TRIGGER MMAP 
    Modes: OUTPUT 
      Out engine  1: 3/HD Audio side
                     Available for use 
    Input formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Output formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Device handle: PCI00701025-0000:00:1b.0-au04
    Related mixer dev: 0
    Sample rate source: 0
    Preferred channel configuration: Not indicated
    Supported number of channels (min - max): 2 - 2
    Native sample rates (min - max): 44100 - 96000 (44100,48000,96000)
    HW Type: Not indicated.
    Minimum latency: Not indicated

HD Audio spdif-out                /dev/oss/hdaudio0/spdout0  (device index 4)
    Legacy device /dev/dsp4
    Caps: TRIGGER MMAP 
    Modes: OUTPUT 
      Out engine  1: 4/HD Audio spdif-out
                     Available for use 
    Input formats (0x00001410):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_AC3        - AC3 (Dolby Digital) encoded audio
      AFMT_S32_LE    - 32 bit signed little endian
    Output formats (0x00001410):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_AC3        - AC3 (Dolby Digital) encoded audio
      AFMT_S32_LE    - 32 bit signed little endian
    Device handle: PCI00701025-0000:00:1b.0-au05
    Related mixer dev: 0
    Sample rate source: 0
    Preferred channel configuration: Not indicated
    Supported number of channels (min - max): 2 - 2
    Native sample rates (min - max): 44100 - 96000 (44100,48000,96000)
    HW Type: Not indicated.
    Minimum latency: Not indicated

High Definition Audio rec1        /dev/oss/hdaudio0/pcmin0  (device index 5)
    Legacy device /dev/dsp5
    Caps: TRIGGER MMAP 
    Modes: INPUT  
      In engine   1: 5/High Definition Audio rec1
                     Available for use 
    Input formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Output formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Device handle: PCI00701025-0000:00:1b.0-au06
    Related mixer dev: 0
    Sample rate source: 0
    Preferred channel configuration: Not indicated
    Supported number of channels (min - max): 2 - 2
    Native sample rates (min - max): 44100 - 96000 (44100,48000,96000)
    HW Type: Not indicated.
    Minimum latency: Not indicated

High Definition Audio rec2        /dev/oss/hdaudio0/pcmin1  (device index 6)
    Legacy device /dev/dsp6
    Caps: TRIGGER MMAP 
    Modes: INPUT  
      In engine   1: 6/High Definition Audio rec2
                     Available for use 
    Input formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Output formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Device handle: PCI00701025-0000:00:1b.0-au07
    Related mixer dev: 0
    Sample rate source: 0
    Preferred channel configuration: Not indicated
    Supported number of channels (min - max): 2 - 2
    Native sample rates (min - max): 44100 - 96000 (44100,48000,96000)
    HW Type: Not indicated.
    Minimum latency: Not indicated

High Definition Audio rec3        /dev/oss/hdaudio0/pcmin2  (device index 7)
    Legacy device /dev/dsp7
    Caps: TRIGGER MMAP 
    Modes: INPUT  
      In engine   1: 7/High Definition Audio rec3
                     Available for use 
    Input formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Output formats (0x00001010):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_S32_LE    - 32 bit signed little endian
    Device handle: PCI00701025-0000:00:1b.0-au08
    Related mixer dev: 0
    Sample rate source: 0
    Preferred channel configuration: Not indicated
    Supported number of channels (min - max): 2 - 2
    Native sample rates (min - max): 44100 - 96000 (44100,48000,96000)
    HW Type: Not indicated.
    Minimum latency: Not indicated

High Definition Audio spdif-in    /dev/oss/hdaudio0/spdin0  (device index 8)
    Legacy device /dev/dsp8
    Caps: TRIGGER MMAP 
    Modes: INPUT  
      In engine   1: 8/High Definition Audio spdif-in
                     Available for use 
    Input formats (0x00001410):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_AC3        - AC3 (Dolby Digital) encoded audio
      AFMT_S32_LE    - 32 bit signed little endian
    Output formats (0x00001410):
      AFMT_S16_LE    - 16 bit signed little endian
      AFMT_AC3        - AC3 (Dolby Digital) encoded audio
      AFMT_S32_LE    - 32 bit signed little endian
    Device handle: PCI00701025-0000:00:1b.0-au09
    Related mixer dev: 0
    Sample rate source: 0
    Preferred channel configuration: Not indicated
    Supported number of channels (min - max): 2 - 2
    Native sample rates (min - max): 44100 - 96000 (44100,48000,96000)
    HW Type: Not indicated.
    Minimum latency: Not indicated

OSS Virtual Mixer v3.0 Playback   /dev/oss/softoss0/pcm0  (device index 9)
    Legacy device /dev/dsp9
    Caps: TRIGGER MMAP VIRTUAL 
    Modes: OUTPUT 
      Out engine  1: 9/OSS Virtual Mixer v3.0 Playback
                     Available for use 
      Out engine  2: 10/OSS Virtual Mixer v3.0 Playback
                     Available for use 
      Out engine  3: 11/OSS Virtual Mixer v3.0 Playback
                     Available for use 
      Out engine  4: 12/OSS Virtual Mixer v3.0 Playback
                     Available for use 
    Input formats (0x00000010):
      AFMT_S16_LE    - 16 bit signed little endian
    Output formats (0x00000010):
      AFMT_S16_LE    - 16 bit signed little endian
    Device handle: softoss0-au01
    Related mixer dev: 1
    Sample rate source: 0
    Preferred channel configuration: Not indicated
    Supported number of channels (min - max): 1 - 2
    Native sample rates (min - max): 48000 - 48000
    HW Type: Not indicated.
    Minimum latency: Not indicated

OSSMIX -d0

Selected mixer 0/High Definition Audio ALC880
Known controls are:
codec1.connector.jack14.mode <jack|input> (currently jack)
codec1.connector.jack14.mute ON|OFF (currently OFF)
codec1.connector.jack14.front-m ON|OFF (currently OFF)
codec1.connector.jack14.inputmi ON|OFF (currently ON)
codec1.connector.jack14.jack <both/leftvol>[:<rightvol>] (currently 57.9:57.9 dB)
codec1.connector.jack15.mode <jack|input> (currently jack)
codec1.connector.jack15.mute ON|OFF (currently OFF)
codec1.connector.jack15.rear-mu ON|OFF (currently OFF)
codec1.connector.jack15.inputmi ON|OFF (currently OFF)
codec1.connector.jack15.jack <both/leftvol>[:<rightvol>] (currently 57.9:57.9 dB)
codec1.connector.jack16.mode <jack|input> (currently jack)
codec1.connector.jack16.mute ON|OFF (currently OFF)
codec1.connector.jack16.center/ ON|OFF (currently OFF)
codec1.connector.jack16.inputmi ON|OFF (currently OFF)
codec1.connector.jack16.jack <both/leftvol>[:<rightvol>] (currently 57.9:57.9 dB)
codec1.connector.jack17.mode <jack|input> (currently jack)
codec1.connector.jack17.mute ON|OFF (currently OFF)
codec1.connector.jack17.side-mu ON|OFF (currently OFF)
codec1.connector.jack17.inputmi ON|OFF (currently OFF)
codec1.connector.jack17.jack <both/leftvol>[:<rightvol>] (currently 57.9:57.9 dB)
codec1.connector.jack18.mode <jack|input> (currently jack)
codec1.connector.jack18.mute ON|OFF (currently OFF)
codec1.connector.jack18.jack <jack|jack|jack|jack> (currently jack)
codec1.connector.jack19.mode <jack|input> (currently jack)
codec1.connector.jack19.mute ON|OFF (currently OFF)
codec1.connector.jack19.jack <jack|jack|jack|jack> (currently jack)
codec1.connector.jack1a.mode <jack|input> (currently jack)
codec1.connector.jack1a.mute ON|OFF (currently OFF)
codec1.connector.jack1a.jack <jack|jack|jack|jack> (currently jack)
codec1.connector.jack1b.mode <jack|input> (currently jack)
codec1.connector.jack1b.mute ON|OFF (currently OFF)
codec1.connector.jack1b.jack <jack|jack|jack|jack> (currently jack)
codec1.record.rec1 <both/leftvol>[:<rightvol>] (currently 31.9:31.9 dB)
codec1.record.rec1.rec1 <jack|jack|jack|jack|jack|jack|jack> (currently jack)
codec1.record.rec2 <both/leftvol>[:<rightvol>] (currently 31.9:31.9 dB)
codec1.record.rec2.rec2 <jack|jack|jack|jack|jack|jack|jack> (currently jack)
codec1.record.rec3 <both/leftvol>[:<rightvol>] (currently 31.9:31.9 dB)
codec1.record.rec3.rec3 <jack|jack|jack|jack|jack|inputmix|jack|jack|jack|jack> (currently jack)
codec1.misc.jack1 <both/leftvol>[:<rightvol>] (currently 58.9:58.9 dB)
codec1.misc.jack2 <both/leftvol>[:<rightvol>] (currently 58.9:58.9 dB)
codec1.misc.jack3 <both/leftvol>[:<rightvol>] (currently 58.9:58.9 dB)
codec1.misc.jack4 <both/leftvol>[:<rightvol>] (currently 58.9:58.9 dB)
codec1.misc.jack5 <both/leftvol>[:<rightvol>] (currently 58.9:58.9 dB)
codec1.misc.jack6 <both/leftvol>[:<rightvol>] (currently 58.9:58.9 dB)
codec1.misc.jack7 <both/leftvol>[:<rightvol>] (currently 58.9:58.9 dB)
codec1.misc.jack8 <both/leftvol>[:<rightvol>] (currently 58.9:58.9 dB)
codec1.misc.inputmix <jack|jack|jack|jack|jack|jack|jack|jack> (currently jack)

OSSMIX -d1

Selected mixer 1/Virtual Mixer
Known controls are:
synth <both/leftvol>[:<rightvol>] (currently 100:100)
pcm <both/leftvol>[:<rightvol>] (currently 100:100)
2 <leftVU>:<rightVU>] (currently 137:136)
autoreset ON|OFF (currently OFF)
effects.eq.prescale <monovol> (currently 255)
effects.eq.lo <monovol> (currently 128)
effects.eq.mid <monovol> (currently 128)
effects.eq.hi <monovol> (currently 128)
effects.eq.xhi <monovol> (currently 128)
effects.eq.bypass ON|OFF (currently OFF)
voices.pcm9 <both/leftvol>[:<rightvol>] (currently 100:100)
voices1 <leftVU>:<rightVU>] (currently 135:135)
voices.pcm10 <both/leftvol>[:<rightvol>] (currently 100:100)
voices2 <leftVU>:<rightVU>] (currently 118:119)
voices.pcm11 <both/leftvol>[:<rightvol>] (currently 100:100)
voices3 <leftVU>:<rightVU>] (currently 0:0)
voices.pcm12 <both/leftvol>[:<rightvol>] (currently 100:100)
voices4 <leftVU>:<rightVU>] (currently 0:0)

Last edited by SpookyET (2008-03-15 01:17:07)

Offline

#18 2008-03-15 03:04:39

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

OSS4 fails to shutdown properly on shutdown because it is installed as /etc/rc.d/oss-linux-free instead of /etc/rc.d/oss4

Change lines 21 and 43. in oss-linux-free.

While the progress bar is fixed with softoss. Volume control does not work, even with the patched gstreamer. It only work with vmix.  System > Preferences > Sound lists nothing under "Default Mixer Tracks".

Last edited by SpookyET (2008-03-15 14:42:48)

Offline

#19 2008-03-15 14:46:30

wyvern
Member
From: 'Old' Scottish/English border.
Registered: 2008-01-06
Posts: 70

Re: The Sorry State of Sound In Linux

SpookyET wrote:

Since you mentioned ATI and went offtopic, can you pastebin your ATI xorg? It works fine for me, but If I enable compiz, i get flickering video and slow scrolling. So, I'm stuck with the open source driver, which uses a lot more CPU when I play video.

Well, since I don't use Compiz, I don't have a lot of the useful 'compiz-related' lines in my xorg, but I've posted it up anyways here

And as for your controls, have you tried for each 'codec1.connector.jack14-18', and 1a, 1b:

ossmix -d0 codec1.connector.jack14.mute OFF

Also, to mute those 'codec.record.rec1-3' , try:

ossmix -d0 codec1.record.rec1 0:0

I think that should leave you with the main channels ('misc.jack1-8') which you can experiment with muting one-by-one, if needs be. Your setup is very different to my own, so this is very much guess work, sorry.

Offline

#20 2008-03-15 17:29:32

Cotton
Member
From: Cornwall, UK
Registered: 2004-09-17
Posts: 568

Re: The Sorry State of Sound In Linux

lucke wrote:

I've been using OSS for some time (it supports headphone output in M-Audio Revolution 5.1, alsa doesn't) - until one version introduced cracking. They don't seem to make other-than-the-latest versions available for download - and that blows.

This what you're after?:  http://www.4front-tech.com/developer/so … table/gpl/

Offline

#21 2008-03-16 01:03:17

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

I request KDE 3.5.9/KDE4 users to report if their players are working.
KDE mostly uses xine. While xine-ui works, Totem-xine failed miserably.
Spare me some work please.

Offline

#22 2008-03-16 08:49:43

schivmeister
Developer/TU
From: Singapore
Registered: 2007-05-17
Posts: 971
Website

Re: The Sorry State of Sound In Linux

ALSA is fine for me (KDE stripped of anything aRts). OSS is just louder, so is JACK. But until JACK (applications) can work well with OSSv4, I'm not even trying it. Do not downplay ALSA just because you (or your hardware) failed to function as expected.

Xine is a framework and engine, OSS is a module providing drivers and a mixer. Xine has got to have the relevant output driver, so see if you can compile in the support for the app in concern.

Last edited by schivmeister (2008-03-16 08:51:26)


I need real, proper pen and paper for this.

Offline

#23 2008-03-16 13:59:38

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

How do you know that Jack applications do not work properly if you have not tried it?

Offline

#24 2008-03-16 14:58:18

SpookyET
Member
Registered: 2008-01-27
Posts: 410

Re: The Sorry State of Sound In Linux

OSS4 fails to resume after suspend. I'm using pm-utils-opensuse (by me) from AUR.

Offline

#25 2008-03-16 15:15:39

schivmeister
Developer/TU
From: Singapore
Registered: 2007-05-17
Posts: 971
Website

Re: The Sorry State of Sound In Linux

s/not even trying it/not replacing ALSA with it

Last edited by schivmeister (2008-03-16 15:16:08)


I need real, proper pen and paper for this.

Offline

Board footer

Powered by FluxBB