Skip to contentSkip to conversation

ALSA vs. OSS

DISCUSSION Markus Hästbacka 18 participants 31 messages Jan 19, 2004 – Feb 25, 2004 2 messages not in the archive

lore
Share & tools
lore
#0 lore raw

Hello list,
I wonder what's the difference with ALSA and OSS. I have tried both, someone may say that ALSA is much better than OSS, but with my experience with ALSA I wouldn't say that, I would probably say it should be removed from the kernel totally.

So, what are the reasons for ALSA to become "default" in 2.6? I know it gives somekind of nice features, but ALSA didn't let me to open two sound sources (like XMMS and Quake3) at the same time, so I guess it is not really done yet, or is it?

Ignore this if you don't care.
Thanks,
Markus.

Signature · 2 lines
"Software is like sex, it's better when it's free."
Markus Hästbacka <midian at ihme dot org>
#1 lore raw
Quoted 3 lines · Markus Hästbacka

but ALSA didn't let me to open two sound sources (like XMMS and Quake3) at the same time, so I guess it is not really done yet, or is it?

Works for me. Right now I've got 3 instances of mpg123 playing 3 different MP3s and XEmacs playing a big .wav file and an audio CD playing. It's a horrible jumbled mess of noise coming out of my speakers, but it is working.

Signature · 4 lines
|---<Steve Youngs>---------------<GnuPG KeyID: A94B3003>---|
|              Ashes to ashes, dust to dust.               |
|      The proof of the pudding, is under the crust.       |
|------------------------------<[email protected]>---|
#26 lore raw
In reply to Markus Hästbacka
Quoted 6 lines · Markus Hästbacka

Hello list,
I wonder what's the difference with ALSA and OSS. I have tried both, someone may say that ALSA is much better than OSS, but with my experience with ALSA I wouldn't say that, I would probably say it should be removed from the kernel totally.

Quoted 8 lines

So, what are the reasons for ALSA to become "default" in 2.6? I know it gives somekind of nice features, but ALSA didn't let me to open two sound sources (like XMMS and Quake3) at the same time, so I guess it is not really done yet, or is it?

Ignore this if you don't care.
Thanks,
Markus.

http://www.alsa-project.org/alsa-doc/doc-php/asoundrc.php3?company=Generic&card=Generic&chip=Generic&module=Generic#softmix
This will tell you how to enable software mixing with ALSA.
I'm not able to give a perfect answer for the reasons why ALSA has been chosen to replace OSS so i will let someone else answer :)

Raphaël.

#2 lore raw
In reply to Steve Youngs

So far I sort of tend to agree with you on OSS being better.

I have a soundblaster Live Value card. I can no longer control the output level through my digital out. With OSS my PCM volume used to affect both the headphone jack and the digital out. With ALSA it affects only the headphone jack.

I have loaded up alsamixer and played with every level in there and it doesn't seem possible to adjust the level anymore unless I adjust the wave volume. As a result I've been unable to get xmms or gkrellm to adjust the volume coming out of my stereo.

Now I like the idea of seperate volume controls, but this doesn't do that.

Regards,
Travis M

Quoted 10 lines · Steve Youngs
Quoted 3 lines · Markus Hästbacka

but ALSA didn't let me to open two sound sources (like XMMS and Quake3) at the same time, so I guess it is not really done yet, or is it?

Works for me. Right now I've got 3 instances of mpg123 playing 3 different MP3s and XEmacs playing a big .wav file and an audio CD playing. It's a horrible jumbled mess of noise coming out of my speakers, but it is working.

#27 lore raw
In reply to Markus Hästbacka
Quoted 5 lines · Markus Hästbacka

Hello list,
I wonder what's the difference with ALSA and OSS. I have tried both, someone may say that ALSA is much better than OSS, but with my experience with ALSA I wouldn't say that, I would probably say it should be removed from the kernel totally.

It seems that you don't understand our goals. Please, look to our web pages - http://www.alsa-project.org. If you use audio devices only for consumer use, you probably don't notice anything special.

Quoted 4 lines

So, what are the reasons for ALSA to become "default" in 2.6? I know it gives somekind of nice features, but ALSA didn't let me to open two sound sources (like XMMS and Quake3) at the same time, so I guess it is not really done yet, or is it?

We don't do this in kernel. We implemented the direct stream mixing in our library (userspace). If your applications already uses ALSA APIs or if you redirect the OSS ioctls to ALSA library (our aoss library), you can enjoy multiple sounds.

Of course, using hardware which can do the hardware mixing is still better. It's the same difference like between sw 3D rendering and hw 3D rendering.

Jaroslav

-----
Jaroslav Kysela <[email protected]>
Linux Kernel Sound Maintainer
ALSA Project, SuSE Labs

#3 lore raw
In reply to Travis Morgan
Quoted 1 line · Travis Morgan

So far I sort of tend to agree with you on OSS being better.

Please. This is a minor point, the important part is hidden from an average user, that ALSA is modularized, supports SMP, multiple sound cards etc etc, so it is much better structured than OSS/Free. Of course some DRIVER is in better/worse in ALSA than in OSS/Free. But it's a minor point, the main advantage is the whole structure of the sound layer Linux has, which is much more better with ALSA than with OSS/Free. The base structure is the hard work, porting drivers from eg OSS/Free or enhance it in ALSA can be minor work. Also, user base of OSS/Free is MUCH larger than ALSA's just because ALSA _was_ a separated project till now, so maybe features provided towards users are not so clean than in the case of OSS/Free which was the part of kernel since ages. But I think this is exactly the reason OSS/Free and ALSA are available in paralell for a while, so developers have got time to do something.

It's like when 'new operating system' is described after its GUI in the m$ world, while it's not a major point when speaking about an OS ;-)

- Gábor (larta'H)

#4 lore raw
In reply to Travis Morgan
Quoted 1 line · Travis Morgan

I have a soundblaster Live Value card. I can no longer control the

I also have a SB Live!, and it doesn't work with ALSA at all - the AC97 codec doesn't load. I haven't taken the time to track it down as it does work just fine with OSS (under SMP at that).

Quoted 3 lines

output level through my digital out. With OSS my PCM volume used to affect both the headphone jack and the digital out. With ALSA it affects only the headphone jack.

That's a purely firmware thing with this card; you should just have to load the right patches. I don't know whether there is a loader utility for alsa, though. Perhaps the old utils will work?

Anyway, even if it's not working for me at the moment, it's still the superior architecture; just wait until the bugs affecting your specific situation are ironed out and userland utilities are available...

MfG, Ulrich

Signature · 2 lines
Heinz Ulrich Stille / Tel.: +49-541-9400463 / Fax: +49-541-9400450
design_d gmbh / Lortzingstr. 2 / 49074 Osnabrück / www.design-d.de
#5 lore raw

Me too. I cannot get ALSA working on my SB Live.

If I may be so bold as to make a suggestion: Maybe the developer in charge of ALSA's e-mu driver could work with us poor unfortunates. There may be some commonality between our systems that causes this (besides the sound blaster live).

My system:
Sound: SBLive Value
Redhat 7.3 (w/ piecemeal recompiles & upgrades)
Kernel: 2.6.1
CPU: Athlon XP 2100+
Mobo: ASUS (I think it's A7V333. I can confirm this later.)

-- Mark "runnin' dirty and deprecated" Borgerding

Quoted 31 lines · Heinz Ulrich Stille
Quoted 3 lines · Travis Morgan

I have a soundblaster Live Value card. I can no longer control the

I also have a SB Live!, and it doesn't work with ALSA at all - the AC97 codec doesn't load. I haven't taken the time to track it down as it does work just fine with OSS (under SMP at that).

Quoted 5 lines

output level through my digital out. With OSS my PCM volume used to affect both the headphone jack and the digital out. With ALSA it affects only the headphone jack.

That's a purely firmware thing with this card; you should just have to load the right patches. I don't know whether there is a loader utility for alsa, though. Perhaps the old utils will work?

Anyway, even if it's not working for me at the moment, it's still the superior architecture; just wait until the bugs affecting your specific situation are ironed out and userland utilities are available...

MfG, Ulrich

#6 lore raw
Quoted 7 lines · Mark Borgerding

Me too. I cannot get ALSA working on my SB Live.

If I may be so bold as to make a suggestion: Maybe the developer in charge of ALSA's e-mu driver could work with us poor unfortunates. There may be some commonality between our systems that causes this (besides the sound blaster live).
....

Ok, here's mine:
Sound: Creative Labs SB Live! EMU10k1
Debian Sid
Kernel: 2.6.1-mm4 atm
CPU: Athlon XP 2600+
Mobo: epox EP-8RDA3+ w/ NForce2 chipset

Signature · 2 lines
"Software is like sex, it's better when it's free."
Markus Hästbacka <midian at ihme dot org>
#7 lore raw

At Tue, 20 Jan 2004 16:03:09 +0200,

Quoted 10 lines · Markus Hästbacka

[1 <text/plain; iso-8859-15 (quoted-printable)>]

Quoted 7 lines · Mark Borgerding

Me too. I cannot get ALSA working on my SB Live.

If I may be so bold as to make a suggestion: Maybe the developer in charge of ALSA's e-mu driver could work with us poor unfortunates. There may be some commonality between our systems that causes this (besides the sound blaster live).
....

Quoted 6 lines

Ok, here's mine:
Sound: Creative Labs SB Live! EMU10k1
Debian Sid
Kernel: 2.6.1-mm4 atm
CPU: Athlon XP 2600+
Mobo: epox EP-8RDA3+ w/ NForce2 chipset

what kernel messages did you get?

mm4 must include ALSA 1.0.1, so i don't know of this problem until now.

--
Takashi Iwai <[email protected]> ALSA Developer - www.alsa-project.org

#15 lore raw
In reply to Mark Borgerding
Quoted 2 lines · Mark Borgerding

If I may be so bold as to make a suggestion: Maybe the developer in charge of ALSA's e-mu driver could work with us poor unfortunates.

The first thing would be to get the latest alsa version (1.0.1) to work with a 2.6 kernel (or wait for it to be integrated); so far I had no luck either with the packaged 1.0.1 or the CVS version.

Loading the kernel version of the modules always fails with:
ALSA sound/pci/ac97/ac97_codec.c:1671: AC'97 0:0 does not respond - RESET
EMU10K1_Audigy: probe of 0000:02:04.0 failed with error -6

Quoted 1 line

Sound: SBLive Value

SB Live! Platinum; lspci says "Creative Labs SB Live! EMU10k1 (rev 07)"

Quoted 1 line

Redhat 7.3 (w/ piecemeal recompiles & upgrades)

Based somewhere around RedHat 8 or 9, but this should not matter, apart perhaps from the compiler, which I compiled myself: gcc 3.3.1 (also tried gcc 2.95.3) and binutils 2.14.

Quoted 1 line

Kernel: 2.6.1

2.6.0

Quoted 2 lines

CPU: Athlon XP 2100+
Mobo: ASUS (I think it's A7V333. I can confirm this later.)

Dual Athlon XP 2000+ on Tyan Tiger MPX

Signature · 2 lines
Heinz Ulrich Stille / Tel.: +49-541-9400463 / Fax: +49-541-9400450
design_d gmbh / Lortzingstr. 2 / 49074 Osnabrück / www.design-d.de
#8 lore raw
In reply to Takashi Iwai
Quoted 3 lines · Takashi Iwai

<..snip..>
what kernel messages did you get?

No messages anywhere, it just doesn't let me open two sound sources at the same time. (the second app freezes without any messages)

Quoted 3 lines

mm4 must include ALSA 1.0.1, so i don't know of this problem until now.

Haven't tried yet with -mm4, I'll do it when I get time to do it. and the problem have been there for a long time, tested first time on test2, didn't work then either.

Kind regards,
Markus.

Signature · 2 lines
"Software is like sex, it's better when it's free."
Markus Hästbacka <midian at ihme dot org>
#28 lore raw
In reply to Jaroslav Kysela
Quoted 8 lines · Jaroslav Kysela
Quoted 4 lines

So, what are the reasons for ALSA to become "default" in 2.6? I know it gives somekind of nice features, but ALSA didn't let me to open two sound sources (like XMMS and Quake3) at the same time, so I guess it is not really done yet, or is it?

We don't do this in kernel. We implemented the direct stream mixing in our library (userspace). If your applications already uses ALSA APIs or if you redirect the OSS ioctls to ALSA library (our aoss library), you can enjoy

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
How can this be done? Just by creating symlinks?

Quoted 3 lines

multiple sounds.

Of course, using hardware which can do the hardware mixing is still

             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Oh, so ALSA does not use the hardware mixing capabilities of the emu10k-chips?

Will this be possible sometime?

Quoted 4 lines

better. It's the same difference like between sw 3D rendering and hw 3D rendering.

Jaroslav

Signature · 1 line
Olaf Dabrunz (od / odabrunz), SUSE Linux AG, Nürnberg
#9 lore raw
In reply to Markus Hästbacka

At Tue, 20 Jan 2004 16:17:13 +0200,

Quoted 8 lines · Markus Hästbacka

[1 <text/plain; iso-8859-15 (quoted-printable)>]

Quoted 3 lines · Takashi Iwai

<..snip..>
what kernel messages did you get?

No messages anywhere, it just doesn't let me open two sound sources at the same time. (the second app freezes without any messages)

then it's a different problem.

sb live has a single capture (record) device although it can play multiple streams at the same time. when both apps try to open fullduplex, the later one is blocked, because the capture device is already occupied.
if it's the case, it is not a bug but the correct POSIX behavior.

as a workaround, try to add the following to /etc/modprobe.conf:

alias snd-pcm-oss nonblock_open=1

this will set the OSS PCM devices as non-blocking.

ciao,

Takashi

#10 lore raw
Quoted 6 lines · Takashi Iwai

At Tue, 20 Jan 2004 16:17:13 +0200,
Markus Hästbacka wrote:
...

then it's a different problem.

Aha.

Quoted 6 lines

sb live has a single capture (record) device although it can play multiple streams at the same time. when both apps try to open fullduplex, the later one is blocked, because the capture device is already occupied.
if it's the case, it is not a bug but the correct POSIX behavior.

Quoted 6 lines

as a workaround, try to add the following to /etc/modprobe.conf:

alias snd-pcm-oss nonblock_open=1

this will set the OSS PCM devices as non-blocking.

I'll try that
Thanks,
Markus

#11 lore raw

Hi,

there was a typo:

Quoted 3 lines
Quoted 3 lines

as a workaround, try to add the following to /etc/modprobe.conf:

alias snd-pcm-oss nonblock_open=1

        ^^^^^
        options

Takashi

#29 lore raw
In reply to Olaf Dabrunz

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

Olaf Dabrunz wrote:
| On 19-Jan-04, Jaroslav Kysela wrote:
|>We don't do this in kernel. We implemented the direct stream mixing in our
|>library (userspace). If your applications already uses ALSA APIs or if you

|>redirect the OSS ioctls to ALSA library (our aoss library), you can enjoy
|
|   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
| How can this be done? Just by creating symlinks?

Reread what was written. the aoss library does that.

-Thomas
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFADT6y+mwggWqA1pQRAnqYAJ4nj30y/yvITXvcgyq4AEMuJ87/ZwCeP4hv lP7TW5saZHHZh5F8NJ2k/aI=
=5HeP
-----END PGP SIGNATURE-----

#31 lore raw
In reply to a message not in the archive
Quoted 9 lines · Jaroslav Kysela
Quoted 5 lines · Olaf Dabrunz
Quoted 3 lines

We don't do this in kernel. We implemented the direct stream mixing in our library (userspace). If your applications already uses ALSA APIs or if you redirect the OSS ioctls to ALSA library (our aoss library), you can enjoy

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
How can this be done? Just by creating symlinks?

No. Use aoss script in our alsa-oss package.

Ah, I see.

# objdump -t /usr/lib/libaoss.so.0.0.0

/usr/lib/libaoss.so.0.0.0: file format elf32-i386

SYMBOL TABLE:
[...]
000062a0 g F .text 00000088 ioctl
[...]
00006490 g F .text 0000008f munmap
[...]
00006bc0 g F .text 00000857 select
[...]
000063c0 g F .text 000000cf mmap
[...]
00006180 g F .text 00000088 write
[...]
00006210 g F .text 00000088 read
[...]
00006890 g F .text 00000325 poll
[...]
00005fd0 g F .text 000000da open
[...]
00006330 g F .text 00000088 fcntl
[...]
000060b0 g F .text 000000c9 close
[...]

So libaoss.so is a wrapper for all file-related system-calls, I suppose to catch calls involving /dev/dsp and /dev/audio.

Signature · 1 line
Olaf Dabrunz (od / odabrunz), SUSE Linux AG, Nürnberg
#32 lore raw
In reply to Olaf Dabrunz

At Tue, 20 Jan 2004 15:24:22 +0100,

Quoted 5 lines · Olaf Dabrunz
Quoted 2 lines

Of course, using hardware which can do the hardware mixing is still

             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Oh, so ALSA does not use the hardware mixing capabilities of the emu10k-chips?

it does.

not only emu10k1 but trident, ymfpci, es1968, maestro3, ali5451 and cs46xx can do. maybe i missed something...

ciao,

Takashi

#16 lore raw
In reply to Mark Borgerding
Quoted 7 lines · Mark Borgerding

Me too. I cannot get ALSA working on my SB Live.

If I may be so bold as to make a suggestion: Maybe the developer in charge of ALSA's e-mu driver could work with us poor unfortunates. There may be some commonality between our systems that causes this (besides the sound blaster live).

Quoted 7 lines

My system:
Sound: SBLive Value
Redhat 7.3 (w/ piecemeal recompiles & upgrades)
Kernel: 2.6.1
CPU: Athlon XP 2100+
Mobo: ASUS (I think it's A7V333. I can confirm this later.)

ALSA works fine with my EMU10K1; you might find you've selected the "Virtual MIDI" device in the kernel config -- for some strange reason if you build ALSA into the kernel, this always gets device #0, and it breaks stuff that's looking for /dev/dsp (not /dev/dsp1).

However, "fine" from the above paragraph is fairly subjective. The ALSA driver is noticably inferior to the OSS driver in that the ALSA developers, despite multiple bug reports and complaints, still persist to use the most horrible software tone controls. Enabling them is an utter waste of time, as putting them above 60 causes clipping and artifacts.

The OSS userspace utilities, however, program the EMU10k1 dsp with a very nice tone control patch that produces a very high quality control with no clipping.

If ALSA does or could support working with the programmable dsp, I'd be happy to switch to it. Right now my "deprecated" SBLive! OSS drivers output higher quality audio.

This is all getting a little OT, because it seems the problems most people have with ALSA are userspace concerns, not a problem with the kernel architecture. Certainly in my case, this is true.

Signature · 8 lines
Cheers,
Alistair.

personal:   alistair()devzero!co!uk
university: s0348365()sms!ed!ac!uk
student:    CS/AI Undergraduate
contact:    7/10 Darroch Court,
            University of Edinburgh.
#17 lore raw
Quoted 16 lines · Alistair John Strachan
Quoted 7 lines · Mark Borgerding

Me too. I cannot get ALSA working on my SB Live.

If I may be so bold as to make a suggestion: Maybe the developer in charge of ALSA's e-mu driver could work with us poor unfortunates. There may be some commonality between our systems that causes this (besides the sound blaster live).

Quoted 7 lines

My system:
Sound: SBLive Value
Redhat 7.3 (w/ piecemeal recompiles & upgrades)
Kernel: 2.6.1
CPU: Athlon XP 2100+
Mobo: ASUS (I think it's A7V333. I can confirm this later.)

Quoted 4 lines

ALSA works fine with my EMU10K1; you might find you've selected the "Virtual MIDI" device in the kernel config -- for some strange reason if you build ALSA into the kernel, this always gets device #0, and it breaks stuff that's looking for /dev/dsp (not /dev/dsp1).

You may try to append 'snd-virmidi=1,1' (first 1 means enable and second index starting with 0, so virmidi will be the second card in your system).

Quoted 5 lines

However, "fine" from the above paragraph is fairly subjective. The ALSA driver is noticably inferior to the OSS driver in that the ALSA developers, despite multiple bug reports and complaints, still persist to use the most horrible software tone controls. Enabling them is an utter waste of time, as putting them above 60 causes clipping and artifacts.

It's not a priority for us. I ported this code from OSS driver at some time, but I probably did some mistake.

Quoted 7 lines

The OSS userspace utilities, however, program the EMU10k1 dsp with a very nice tone control patch that produces a very high quality control with no clipping.

If ALSA does or could support working with the programmable dsp, I'd be happy to switch to it. Right now my "deprecated" SBLive! OSS drivers output higher quality audio.

We don't have user space tools to update DSP code although our emu10k1 driver is capable to do it. Sure, we are doing things differently than OSS driver so you cannot simply use the OSS utilities.

Perhaps, time to help us?

Jaroslav

-----
Jaroslav Kysela <[email protected]>
Linux Kernel Sound Maintainer
ALSA Project, SuSE Labs

#14 lore raw
In reply to a message not in the archive

At Tue, 20 Jan 2004 12:06:58 -0500,

Quoted 6 lines · Charles Shannon Hendrix
Quoted 2 lines · Takashi Iwai

sb live has a single capture (record) device although it can play multiple streams at the same time. when both apps try to open

Quoted 8 lines

I don't believe that is true of all Live! cards.

There are some minor differences between them.

I record from multiple sources all the time on mine.

It came with a utility that let you mix all the inputs, perform effects on them, etc.

yes, the sources are multiple but they are mixed up. the point is that the access to the device is exclusive. i.e. you cannot record the mic input with two applications concurrently (ALSA can do via dsnoop plugin, BTW).

Takashi

#20 lore raw
In reply to Heinz Ulrich Stille

Quoting Heinz Ulrich Stille <[email protected]>:

Quoted 6 lines · Travis Morgan
Quoted 1 line

I have a soundblaster Live Value card. I can no longer control the

I also have a SB Live!, and it doesn't work with ALSA at all - the AC97 codec doesn't load. I haven't taken the time to track it down as it does work just fine with OSS (under SMP at that).

My SBLive! has been working with ALSA since I first used it (0.9.6 or something) on both 2.4 and 2.6 kernels.

Asus nForce2 board.
01:0a.0 Multimedia audio controller: Creative Labs SB Live! EMU10k1 (rev 07) 01:0a.1 Input device controller: Creative Labs SB Live! MIDI/Game Port (rev 07)

-sandalle

Signature · 17 lines
PGP Key Fingerprint:  FCFF 26A1 BE21 08F4 BB91  FAED 1D7B 7D74 A8EF DD61
http://search.keyserver.net:11371/pks/lookup?op=get&search=0xA8EFDD61

-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GCS/E/IT$ d-- s++:+>: a-- C++(+++) BL++++VIS>$ P+(++) L+++ E-(---) W++ N+@ o?
K? w++++>-- O M-@ V-- PS+(+++) PE(-) Y++(+) PGP++(+) t+() 5++ X(+) R+(++)
tv(--)b++(+++) DI+@ D++(+++) G>+++ e>+++ h---(++) r++ y+
------END GEEK CODE BLOCK------

Eric Sandall                     |  Source Mage GNU/Linux Developer
[email protected]                  |  http://www.sourcemage.org/
http://eric.sandall.us/          |  SysAdmin @ Inst. Shock Physics @ WSU
http://counter.li.org/  #196285  |  http://www.shock.wsu.edu/

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.
#24 lore raw
In reply to Steve Youngs
Quoted 10 lines · Steve Youngs
Quoted 3 lines · Markus H?stbacka

but ALSA didn't let me to open two sound sources (like XMMS and Quake3) at the same time, so I guess it is not really done yet, or is it?

Works for me. Right now I've got 3 instances of mpg123 playing 3 different MP3s and XEmacs playing a big .wav file and an audio CD playing. It's a horrible jumbled mess of noise coming out of my speakers, but it is working.

You probably have a Soundblaster Live or similar, which has multiple hardware wave outputs.

OSS has software mixing. ALSA seems designed for people relying on esd, aRts or similar multiplexing daemons.

It's possible to run a program via 'esddsp' or 'artsdsp' to reroute /dev/dsp to the daemon, but the overhead isn't so nice, and the output quality is often wanting.

#25 lore raw
Quoted 12 lines · Brian McGroarty
Quoted 10 lines · Steve Youngs
Quoted 3 lines · Markus H?stbacka

but ALSA didn't let me to open two sound sources (like XMMS and Quake3) at the same time, so I guess it is not really done yet, or is it?

Works for me. Right now I've got 3 instances of mpg123 playing 3 different MP3s and XEmacs playing a big .wav file and an audio CD playing. It's a horrible jumbled mess of noise coming out of my speakers, but it is working.

Quoted 5 lines

You probably have a Soundblaster Live or similar, which has multiple hardware wave outputs.

OSS has software mixing. ALSA seems designed for people relying on esd, aRts or similar multiplexing daemons.

Don't you mean the other way around?

Quoted 3 lines

It's possible to run a program via 'esddsp' or 'artsdsp' to reroute /dev/dsp to the daemon, but the overhead isn't so nice, and the output quality is often wanting.

True.

Signature · 2 lines
Måns Rullgård
[email protected]
#21 lore raw
In reply to Eric Sandall
Quoted 10 lines · Eric Sandall

Quoting Heinz Ulrich Stille <[email protected]>:

Quoted 6 lines · Travis Morgan
Quoted 1 line

I have a soundblaster Live Value card. I can no longer control the

I also have a SB Live!, and it doesn't work with ALSA at all - the AC97 codec doesn't load. I haven't taken the time to track it down as it does work just fine with OSS (under SMP at that).

My SBLive! has been working with ALSA since I first used it (0.9.6 or something) on both 2.4 and 2.6 kernels.

My SBLive! 5.1 card has always worked great with ALSA (kernels 2.4 and now
2.6), and I started using ALSA at about v9.0 (I think). It's on an A-Bit KR7A-RAID board now.

	8-Dale
Signature · 3 lines
The Dynaplex Network - http://www.thedynaplex.org
We support Open Source 100% with Linux and FreeBSD!
Currently running Gentoo Linux 1.4 and FreeBSD 4.9
#22 lore raw
Quoted 9 lines · Dale Weber
Quoted 8 lines · Eric Sandall

Quoting Heinz Ulrich Stille <[email protected]>:

Quoted 6 lines · Travis Morgan
Quoted 1 line

I have a soundblaster Live Value card. I can no longer control the

I also have a SB Live!, and it doesn't work with ALSA at all - the AC97 codec doesn't load. I haven't taken the time to track it down as it does work just fine with OSS (under SMP at that).

Quoted 7 lines
Quoted 2 lines

My SBLive! has been working with ALSA since I first used it (0.9.6 or something) on both 2.4 and 2.6 kernels.

My SBLive! 5.1 card has always worked great with ALSA (kernels 2.4 and now
2.6), and I started using ALSA at about v9.0 (I think). It's on an A-Bit KR7A-RAID board now.

I have an Intel 8x0 that doesn't work with the alsa driver, but the OSS driver works perfectly. I can load the alsa driver, and get interrupts from the card (built-in on the MB), but no sound comes out, and I tried changing the volume in alsamixer with no results except for hearing static when I unmuted the headphone output.

I'm using 2.6.1, and I've never used alsa in 2.4.

00:00.0 Host bridge: Intel Corp. 82845 845 (Brookdale) Chipset Host Bridge (rev 04) 00:01.0 PCI bridge: Intel Corp. 82845 845 (Brookdale) Chipset AGP Bridge (rev 04) 00:1e.0 PCI bridge: Intel Corp. 82801BA/CA/DB/EB PCI Bridge (rev 05)
00:1f.0 ISA bridge: Intel Corp. 82801BA ISA Bridge (LPC) (rev 05)
00:1f.1 IDE interface: Intel Corp. 82801BA IDE U100 (rev 05)
00:1f.2 USB Controller: Intel Corp. 82801BA/BAM USB (Hub #1) (rev 05)
00:1f.3 SMBus: Intel Corp. 82801BA/BAM SMBus (rev 05)
00:1f.4 USB Controller: Intel Corp. 82801BA/BAM USB (Hub #2) (rev 05)
00:1f.5 Multimedia audio controller: Intel Corp. 82801BA/BAM AC'97 Audio (rev 05) 01:00.0 VGA compatible controller: ATI Technologies Inc Rage 128 Pro Ultra TF 02:09.0 Ethernet controller: Digital Equipment Corporation DECchip 21140 [FasterNet] (rev 22) 02:0a.0 SCSI storage controller: Adaptec AHA-2940U/UW/D / AIC-7881U (rev 01)

#18 lore raw
In reply to Jaroslav Kysela
Quoted 10 lines · Jaroslav Kysela
Quoted 3 lines · Alistair John Strachan

If ALSA does or could support working with the programmable dsp, I'd be happy to switch to it. Right now my "deprecated" SBLive! OSS drivers output higher quality audio.

We don't have user space tools to update DSP code although our emu10k1 driver is capable to do it. Sure, we are doing things differently than OSS driver so you cannot simply use the OSS utilities.

Perhaps, time to help us?

Is there any documentation on the interface for uploading new DSP code to the emu10k1?

Such would be /very/ useful for the task of writing tools to do the job.

Signature · 6 lines
	1024D/E65A7801 Zephaniah E. Hull <[email protected]>
	   92ED 94E4 B1E6 3624 226D  5727 4453 008B E65A 7801
	    CCs of replies from mailing lists are requested.

   "Never attribute to malice that which can be adequately explained by
stupidity" - Hanlon's Razor
#19 lore raw
Quoted 10 lines · Zephaniah E. Hull
Quoted 9 lines · Jaroslav Kysela
Quoted 3 lines · Alistair John Strachan

If ALSA does or could support working with the programmable dsp, I'd be happy to switch to it. Right now my "deprecated" SBLive! OSS drivers output higher quality audio.

We don't have user space tools to update DSP code although our emu10k1 driver is capable to do it. Sure, we are doing things differently than OSS driver so you cannot simply use the OSS utilities.

Quoted 6 lines
Quoted 1 line

Perhaps, time to help us?

Is there any documentation on the interface for uploading new DSP code to the emu10k1?

Such would be /very/ useful for the task of writing tools to do the job.

There is a preliminary emu10k* loader:

http://ld10k1.sourceforge.net/

The author is also reachable at <[email protected]>.

The whole API is in linux/include/sound/emu10k1.h (look at bottom).

Jaroslav

-----
Jaroslav Kysela <[email protected]>
Linux Kernel Sound Maintainer
ALSA Project, SuSE Labs

#23 lore raw
In reply to Travis Morgan

For what it's worth, I figured out the below problem. I was able to set XMMS' ALSA output to Wave instead of PCM and can now adjust the volume as I could with OSS. Also, for gkrellm, the digital out volume is now PCM2 rather than just PCM.

Now I can adjust my XMMS volume with my Playstation controller again! :) The USB converter and a USB extension were two of the best purchases I've made for my system.

Bus 002 Device 002: ID 6666:0667 Prototype product Vendor ID Smart Joy PSX, PS-PC Smart JoyPad

Linux castle 2.6.2-rc1-mm2 #1 Fri Jan 23 23:37:56 MST 2004 i686 AMD Athlon(tm) XP 2600+ AuthenticAMD GNU/Linux

Regards,
Travis M

Quoted 17 lines · Travis Morgan

So far I sort of tend to agree with you on OSS being better.

I have a soundblaster Live Value card. I can no longer control the output level through my digital out. With OSS my PCM volume used to affect both the headphone jack and the digital out. With ALSA it affects only the headphone jack.

I have loaded up alsamixer and played with every level in there and it doesn't seem possible to adjust the level anymore unless I adjust the wave volume. As a result I've been unable to get xmms or gkrellm to adjust the volume coming out of my stereo.

Now I like the idea of seperate volume controls, but this doesn't do that.

Regards,
Travis M

#12 lore raw
In reply to Takashi Iwai
Quoted 9 lines · Takashi Iwai

Hi,

there was a typo:

Quoted 3 lines
Quoted 3 lines

as a workaround, try to add the following to /etc/modprobe.conf:

alias snd-pcm-oss nonblock_open=1

        ^^^^^
        options

Hi,

I tried this, but the problem still occurs.
Ok, here's a explanation of what I did:
1. I started xmms and started playing something (NOTE!!!: This doesn't seem to matter at all, because I just tried without it and 3. still hangs.)
2. I started TeamSpeak2
3. I started Quake3, Quake3 hangs.

So where could the problem occur?

There's no messages anywhere, no error messages anywhere.

I wonder if it matters if ALSA with it's oss things are compiled in? (Not as module?)

Markus