strange behaviour from internal mic on Thinkpad

Have a new Lenovo Thinkpad SL510 that I have installed Opensuse 11.3 on to.
Everything seems to be working pretty well except for a problem with the inbuilt microphone.

Got some good information from this site about some settings with alsamixer

I should first explain that alsamixer recognizes the microphone as “mic” and not “inbuilt mic”

The problem I have is that after booting, if I record something all I get is white noise in the recording (using arecord)
If I load alsamixer and change to “inbuilt mic” and then straight back to “mic” and record something there is no problem.

Although I have entered the command “alsactl store” the problem returns on reboot.

Here is some info that might be useful


rpm -qa '*alsa*'
alsa-utils-1.0.23-1.8.i586
alsa-plugins-1.0.23-1.9.i586
alsa-1.0.23-2.12.i586
alsa-oss-1.0.17-29.2.i586


rpm -qa '*pulse*'
libpulse0-0.9.21-10.1.1.i586
libpulse-mainloop-glib0-0.9.21-10.1.1.i586
libxine1-pulse-1.1.19-2.pm.48.13.i586


rpm -q libasound2
libasound2-1.0.23-2.12.i586


uname -a
Linux linux-orxz.site 2.6.34.7-0.7-default #1 SMP 2010-12-13 11:13:53 +0100 i686 i686 i386 GNU/Linux


cat /etc/modprobe.d/50-sound.conf

options snd slots=snd-hda-intel
# u1Nb.kXwNWe4doZB:82801I (ICH9 Family) HD Audio Controller
alias snd-card-0 snd-hda-intel

Please let me know if any other info is required.

really appreciate any help with this.

On my SL510 with 11.3: alsamixer and KMix shows the microphone selections (under channel “Input Source”) as “Mic” and “Front Mic”. It’s the same on my main and test 11.3 systems, although test has 2.6.36.2-3 kernel-desktop and KDE 4.5.4.

Then I notice you are running 32bit 11.3 on a 64bit notebook. Any special reason? I have 64bit 11.3 installed. Also, you are running default kernel, whereas I have kernel-desktop. I didn’t pick that up in the other thread we are involved in. Other than those differences, my info is similar to yours above. Also:

>arecord -l
**** List of CAPTURE Hardware Devices ****
card 0: Intel [HDA Intel], device 0: ALC269 Analog [ALC269 Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0

Haven’t managed to record anything yet, but still working on that…

hmmmmm . . .

I hadn’t actually considered that it was a 64 bit computer as it came preinstalled with a 32bit version of windows.

Are there any advantages to 64bit suse over the 32bit version.

I don’t really want to go through the install process again but might look at 64bit Suse 11.4 when released in a couple of months.

As per the mic issue - over a two day period of changing nothing it appears to have resolved itself. Not sure why. Will keep an eye on it and see if sticks.

Incidentally - what works for you? “Mic” or “Front Mic”

For me - alsamixer shows the choice of “mic” & “inbuilt mic” while Kmix shows “mic” & “front mic” - this difference also seems a bit strange. :slight_smile:

Since your mic works, I recommend you open your mixer and do a screen print of the settings. Save that. That will then be your baseline record of functional mic settings.

If sometime in the future it does not work, open up the saved screen print and compare settings. It might help solve a future problem. That works for my wife.

good idea - thanks. :slight_smile:

So did mine. There are quite a few long threads discussing pros and cons of 64bit v. 32bit, worth a search when you have the time. On this size of m/c I think 64bit is probably a good idea particularly if you do a lot of video work, and I think there are increasingly more openSUSE 64bit users to discover and help fix issues. However I understand your reluctance to change now, considering you have a couple of things working I don’t (e.g. suspend to RAM, and the builtin mic). :wink:

You may have missed this question in the other thread. What BIOS release level do you have? I think you can check it via the initial ThinkPad screen via a key press (sorry can’t recall without looking).

Incidentally - what works for you? “Mic” or “Front Mic”

When I tried to record with Audacity, Front Mic was the only one to register on the level meters, but I don’t know what noise it was picking up.

For me - alsamixer shows the choice of “mic” & “inbuilt mic” while Kmix shows “mic” & “front mic” - this difference also seems a bit strange.

Yes, maybe difference is in 32bit v. 64bit distros, or does it pick names up from BIOS data? “inbuilt” seems a better description to me.

sorry - yes, missed that one. I’ll take a look when I get home tonight.

When I tried to record with Audacity, Front Mic was the only one to register on the level meters, but I don’t know what noise it was picking up.

before I got the mic to work I was getting white noise generated from somewhere. Has gone now that mic is working.

Yes, same here. Just got it working with “Front Mic” using arecord. It gives a nice clean recording, picking up a radio a couple of metres away in the background. :wink:

hmmm . . .intersting. This raises another possible prob.
ON boot - to enter bios you are supposed to hit the blue thinkvantagae button. Pressing that does nothing for me and grub comes up.
but that would be another thread again.

Anyway . . .booted into Win7 and got bios version via thinkvantage

Bios version = 6JET83WW (1.41)

In Linux one can often get the bios version by running the program (with root permissions):

dmidecode


For example on my Core i7 920 with an Asus P6T Deluxe V2 motherboard, I get after typing dmidecoe:


# dmidecode 2.10
SMBIOS 2.5 present.
81 structures occupying 3016 bytes.
Table at 0x000F0710.

Handle 0x0000, DMI type 0, 24 bytes
**BIOS Information
        Vendor: American Megatrends Inc.
        Version: 0504   
        Release Date: 05/19/2009**

thanks - wasn’t aware of that.

I get . . .


# dmidecode
# dmidecode 2.10
SMBIOS 2.5 present.
44 structures occupying 1523 bytes.
Table at 0x000E0010.

Handle 0x0000, DMI type 0, 24 bytes
BIOS Information
        Vendor: LENOVO
        Version: 6JET83WW (1.41 )
        Release Date: 09/21/2010
        Address: 0xE45F0
        Runtime Size: 113168 bytes
        ROM Size: 2048 kB
        Characteristics:
                PCI is supported
                PNP is supported
                BIOS is upgradeable
                BIOS shadowing is allowed
                ESCD support is available
                Boot from CD is supported
                ACPI is supported
                USB legacy is supported
                BIOS boot specification is supported
                Targeted content distribution is supported
        BIOS Revision: 1.65
        Firmware Revision: 1.37

Excellent. I didn’t know that either, thanks. :slight_smile:

Also @farcusnz, thanks for yours, so here is mine and it is older, as expected:


# dmidecode
# dmidecode 2.10
SMBIOS 2.5 present.
44 structures occupying 1523 bytes.
Table at 0x000E0010.

Handle 0x0000, DMI type 0, 24 bytes
BIOS Information
	Vendor: LENOVO
	Version: 6JET58WW (1.16 )
	Release Date: 09/17/2009
	Address: 0xE4670
	Runtime Size: 113040 bytes
	ROM Size: 2048 kB
	Characteristics:
		PCI is supported
		PNP is supported
		BIOS is upgradeable
		BIOS shadowing is allowed
		ESCD support is available
		Boot from CD is supported
		ACPI is supported
		USB legacy is supported
		BIOS boot specification is supported
		Targeted content distribution is supported
	BIOS Revision: 1.22
	Firmware Revision: 1.8

Don’t normally like to update the BIOS unless a fix is indicated. Now I am uncertain wrt Suspend to Ram, since I learnt our 11.3’s are different wrt architecture and kernel type. :\

BTW, internal mic worked OOTB on my Gnome partition. alsamixer targets pulseaudio as device, and you only get an input and an output channel, with input choice of Microphone 1/2 (1 is the internal). :wink:

I’ve now completed testing my SL510’s internal Mic on three systems running 11.3 with 64bit kernel-desktop. Note the following changes to Input Source, particularly with the newer kernel:

Standard system: 2.6.34.7-0.7 with KDE 4.4.4: select Input Source “Mic” (not “Front Mic”)

Standard system: 2.6.34.7-0.7 with Gnome: select Input Source “Microphone 1” (not “Microphone 2”).

Testing system: 2.6.36.2-3 with KDE 4.5.4: select Input Source Front “Mic” (not “Mic”). Here, the chip/mixer/codec is now detected as “Intel Cantiga HDMI”, whereas previously it’s detected as “Intel G45 DEVCTG”.

For clarification, Gnome’s “Microphone 1/2” is used in Sound Preferences. The alsa-info script reports those Input Source devices as “Mic”/“FrontMic”, as for KDE. The input Mic device selection for KDE/Gnome is equivalent, given the same kernel.

I’m back to my mic not working ‘sometimes’ on boot. Have added input source to kmix options now so can just change to front mic and back to mic and works again.

not sure what would be causing this behaviour but seems very random as sometimes no problem on boot at all.

I’ll check for this on my standard 11.3 KDE system. So far not getting your issue on my test system with kernel-desktop 2.6.36.2-3 and “Front Mic”.

When I booted my standard 11.3, checked mixer settings (ok with no changes) and used arecord, the resulting test.wav was silent (checked with both mplayer and audacity). Switching from “Mic” to “Front Mic” and back, appeared to fix the problem and a good recording resulted, using audacity. I haven’t seen this problem with the newer kernel (“Intel Cantiga HDMI” support - see above).

With kernel-desktop 2.6.34.7-0.7, /var/log/boot.msg contains the line:


<7>   16.696074] ALSA hda_codec.c:4405:    inputs: mic=0x12, fmic=0x18, line=0x0, fline=0x0, cd=0x0, aux=0x0

With kernel-desktop 2.6.36.2.3, /var/log/boot.msg contains the line:


<7>   17.157486] ALSA hda_codec.c:4637:    inputs: mic=0x18, fmic=0x12, line=0x0, fline=0x0, cd=0x0, aux=0x0

That shows the reversed Mic alternatives (in blue text). There are some differences in other ALSA messages, but I don’t know how relevant these are. Between the two kernels, there are significant differences with thinkpad_acpi messages. However I will post those in that other thread we were using, as it is more relevant.

If I can repeat your problem again, I will confirm that here.

I suspect being random makes it more difficult to get fixed.

This reads like there could be a bug in the alsa driver designed for this hardware. You could consider writig a bug report to have this looked at by the SuSE-GmbH sound packager (who is also an alsa developer). Bug report would need be written on openSUSE-11.3 component “sound”.

Indeed, I am considering the bug report with some reservation. Perhaps you could advise after considering the info below. :slight_smile:

Firstly to confirm the random nature of @farcusnz’s problem, and my seeing the same problem exactly as he describes, and the workaround. After booting my standard 11.3 KDE system today, the internal Mic (as “Mic”) worked perfectly. However, the problem immediately showed after booting my standard 11.3 Gnome system, “Microphone 1” was dead. Switching to “Microphone 2” and back cleared the problem. Again booting my 11.3 KDE system with newer kernel today, worked properly.

For me that points the finger clearly at the standard 11.3 kernel/modules e.g. the alsa driver. However, please consider the following short clips from running alsa-info which I combined for comparison (differences in blue):

!!DMI Information
!!---------------

Manufacturer:      LENOVO                         LENOVO
Product Name:      28473PG                        28473PG


!!Kernel Information
!!------------------

Kernel release:    2.6.34.7-0.7-desktop           2.6.36.2-3-desktop
Operating System:  GNU/Linux                      GNU/Linux
Architecture:      x86_64                         x86_64
Processor:         x86_64                         x86_64
SMP Enabled:       Yes                            Yes


!!ALSA Version
!!------------

Driver version:     1.0.22.1                      1.0.23
Library version:    
Utilities version:  1.0.23                        1.0.23


!!Loaded ALSA modules
!!-------------------

snd_hda_intel                                     snd_hda_intel
                                                  thinkpad_acpi

Note that thinkpad_acpi fails to load with standard kernel-desktop, /var/log/boot.msg reveals:

<3>   16.498777] thinkpad_acpi: Not yet supported ThinkPad detected!

My concern is that alsa developer(s) can do little without a thinkpad_acpi module and correct kernel support (i.e. “Intel Cantiga HDMI”, not as current “Intel G45 DEVCTG”). Would you expect the the newer thinkpad_acpi to be backported to the earlier kernel?

What do you think re a bug report submission?

I do not know enough to provide a technical troubleshooting suggestion.

What I do know is the alsa developer’s worked with the user who created the alsa-info.sh script (who used to frequent IRC chat #alsa) to modify the alsa-info.sh script to provide them all sorts of useful information on a PCs hardware and software sound configuration.

So I recommend write a bug report on openSUSE component “sound”, describe the problem, and attach to the bug report the appropriate /tmp/alsa-info.sh output(s) of running the diagnostic script:

 /sbin/alsa-info.sh --no-upload