[lucid regression] HDA: pulseaudio volume control sets alsa master to 0 when still at 15% (incorrect db data?)

[Toby Smithe]

As described in the summary. Not sure if this is an alsa bug or a pulseaudio one. I tried to figure out how to use dbverify, but it defeated me.

As far as I can remember, this seems to be a regression.

ProblemType: Bug
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.21.
Architecture: amd64
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: STAC92xx Analog [STAC92xx Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: toby 1829 F.... pulseaudio
                      toby 3827 F.... spotify.exe
 /dev/snd/pcmC0D0p: toby 1829 F...m pulseaudio
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xfc500000 irq 22'
   Mixer name : 'IDT 92HD73C1X5'
   Components : 'HDA:111d7675,1028029f,00100103'
   Controls : 17
   Simple ctrls : 11
Card1.Amixer.info:
 Card hw:1 'HDMI'/'HDA ATI HDMI at 0xfc010000 irq 17'
   Mixer name : 'ATI R6xx HDMI'
   Components : 'HDA:1002aa01,00aa0100,00100000'
   Controls : 4
   Simple ctrls : 1
Card1.Amixer.values:
 Simple mixer control 'IEC958',0
   Capabilities: pswitch pswitch-joined penum
   Playback channels: Mono
   Mono: Playback [off]
CheckboxSubmission: cb8560608953d46833a1bf646e78598a
CheckboxSystem: d00f84de8a555815fa1c4660280da308
Date: Sun Mar 7 15:40:28 2010
DistroRelease: Ubuntu 10.04
EcryptfsInUse: Yes
NonfreeKernelModules: wl
Package: alsa-base 1.0.22.1+dfsg-0ubuntu3
PackageArchitecture: all
ProcEnviron:
 LANGUAGE=fr_FR:fr:en_GB:en
 PATH=(custom, user)
 LANG=fr_FR.utf8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.32-15.22-generic
SourcePackage: alsa-driver
Uname: Linux 2.6.32-15-generic x86_64

[Daniel T Chen (crimsun)]

http://pulseaudio.org/wiki/PulseAudioStoleMyVolumes

[Toby Smithe (tsmithe)]

I read the document, but it didn't really explain the bug; I understand how the logic works, but I don't understand how pulseaudio makes the error. Unless, of course, the db data is suddenly wrong, or if there is a bug in pa's implementation of the logic on this hardware. It's curious, though, that this did work in the past.

It also seems that the regression has caused the "fade" and "balance" controls (accessed from gnome-volume-control) to be entirely dysfunctional. This was not the case before.

Finally, why is the importance only wishlist? (And why won't-fix in lucid?) The bug fulfils the criteria for medium: moderate and confusing impact on core program; problem with non-essential hardware.

[Daniel T Chen (crimsun)]

I intend to freeze as much of the userspace audio stack as possible before Beta 1. You should provide a verbose PA log and carefully check for dB output from snd_pcm_dump(). Furthermore, unless you intend to chase this one yourself, this symptom is not the highest priority audio bug.

[Dudude (dudude1990)]

This issue also affects me.

The audio card that I use, as given by 'lspci -nn' is:

00:1b.0 Audio device [0403]: Intel Corporation 82801H (ICH8 Family) HD Audio Controller [8086:284b] (rev 02)

Attached is a sceenshot showing the audio levels being different in AlsaMixer and in the PulseAudio Sound Preferences window.

[PaulR (paulr-bex)]

I've not been able to use Lucid because of this. And it looks like Karmic won't work either. I'm not an expert, so would require detailed instructions on how to get info to you; but would certainly be willing to do that.

My computer is a Lenovo 3000N100 with the Intel ALC861VD sound processor (if I read this right). In Jaunty I run it in OSS because ALSA doesn't work there either.

So far as I know, this is a fairly common computer. If sound won't work in it, it seems to me, we have a problem.

[PaulR (paulr-bex)]

My computer is a Lenovo 3000N100; specifically a 0768-04U. It has the ALC 861-VD sound card.

My following another thread I found the following:

Edit (use sudo) /etc/modprobe.d/alsa-base.conf
by adding the line

options snd-hda-intel model=3stack

to the end of the file. Restart. No promises, but it worked for me.

[Maarten Bosmans (mkbosmans)]

I have an HDA Intel (ICH 7) soundcard on a Samsung NC10 netbook.
With the pulseaudio dbverify utility I verified that the dB values for my soundcard are wrong. The Master mixer control is OK, but the PCM mixer (with fine-grained steps 0-255 of 0.2 dB each) seems off.

If this can be corrected, I'll gladly submit the right dB scale using the dbmeasure utility. As HDA-Intel is a very generic soundchip/driver, are the dB values expected to be the same for every laptop/netbook it is used in?

[Daniel T Chen (crimsun)]

The endpoint value should match, but the rest is codec- and
controller-dependent.

On Jul 13, 2010 11:41 AM, "Maarten Bosmans" <email address hidden> wrote:

I have an HDA Intel (ICH 7) soundcard on a Samsung NC10 netbook.
With the pulseaudio dbverify utility I verified that the dB values for my
soundcard are wrong. The Master mixer control is OK, but the PCM mixer (with
fine-grained steps 0-255 of 0.2 dB each) seems off.

If this can be corrected, I'll gladly submit the right dB scale using
the dbmeasure utility. As HDA-Intel is a very generic soundchip/driver,
are the dB values expected to be the same for every laptop/netbook it is
used in?

--
[lucid regression] HDA: pulseaudio volume control sets alsa master to 0 when
still at 15% (inco...

Audio Team, which is subscribed to pulseaudio in ubuntu.

Status in “pulseaudio” package in Ubuntu: ...

Bug description:
As described in the summary. Not sure if this is an alsa bug or a pulseaudio
one. I...

[Maarten Bosmans (mkbosmans)]

So is there a way to differentiate between the codecs and controllers such that the dB measurements can be used for my specific combination?

[Daniel T Chen (crimsun)]

Sure; see the individual codec dumps in /proc/asound/card*/codec*

On Jul 14, 2010 4:31 AM, "Maarten Bosmans" <email address hidden> wrote:

So is there a way to differentiate between the codecs and controllers
such that the dB measurements can be used for my specific combination?

--
[lucid regression] HDA: pulseaudio volume control sets alsa master to 0 when
still at 15% (inco...

Audio Team, which is subscribed to pulseaudio in ubuntu.

Status in “pulseaudio” package in Ubuntu: Confirmed
Status in “pulseaudio” source package in Lucid: ...

Bug description:
As described in the summary. Not sure if this is an alsa bug or a pulseaudio
one. I...

[machrider (machrider)]

I believe Daniel Chen set this to WontFix prematurely. This report isn't "wahh, I don't like how pulseaudio works." It's an actual bug.

When I set the volume with gnome's mixer applet or with pavucontrol, and pulseaudio's level reaches 15%, then Alsa's "Master" channel reaches zero. Therefore, at all levels below 15%, there is no sound at all. Alsa's PCM volume stays at 100% until pulse reaches 2% or so, and then PCM finally drops to 0%. Of course this has no effect, because Alsa's Master has been silenced the entire time. I'm guessing the intent is that Master should stay at a small, nonzero volume, until pulse is finally lowered to zero.

This is happening on a Dell M6400 with Intel HDA:
00:1b.0 Audio device: Intel Corporation 82801I (ICH9 Family) HD Audio Controller (rev 02)

I've attached a screenshot of puvacontrol and alsamixer side-by-side at 15% volume.

[machrider (machrider)]

Sorry, forgot to note that I'm seeing this behavior on a system running Maverick, packages fully updated as of a few hours ago.

[Daniel T Chen (crimsun)]

@Mike Are you using the same hardware as the original reporter, Toby?
If not, you should file a separate bug report against alsa-driver
(`ubuntu-bug alsa-base`).

[dino99 (9d9)]

That version is no more supported