Re: testers for SB Extigy wanted

[Takashi Iwai <tiwai@suse.de>]

At Fri, 11 Oct 2002 10:34:11 -0500,
mark@undertone.org wrote:
> 
> On Fri, Oct 11, 2002 at 05:07:25PM +0200, Takashi Iwai wrote:
> > > Here is everything syslog spits out at me when I start up alsa (the
> > > cant locate module stuff is obviously nothing, along with one or two
> > > other things, but I'll leave them in for completeness sake):
> > > 
> > > Oct 11 09:30:39 atrophy kernel: PCI: Found IRQ 10 for device 00:08.0
> > > Oct 11 09:30:39 atrophy kernel: usb.c: registered new driver snd-usb-audio
> > > Oct 11 09:30:40 atrophy kernel: ALSA ../alsa-kernel/usb/usbaudio.c:1662: 4:1:2 : unsupported format type 0
> > 
> > this should be harmless.  just ignored a bogus descriptor.
> > to be sure, could you send the output of "lsusb -v" for this device?
> 
> This is fairly long, so I have it included as an attachment.

thanks.

> 
> > so far, all looks good.
> > 
> > just try to start "alsamixer -c1" and unmute the switch and change the
> > volume as you like.
> > you can play via hw:1,0 pcm (you might need to use plughw:1,0
> > instead), for example,
> > 
> > 	% aplay -Dplughw:1,0 foo.wav
> > 
> > and for capture
> > 
> > 	% arecord -Dplughw:1,0 -fcd bar.wav
> 
> I'd love to do this, but:
> 
> atrophy:/etc/modutils# amixer -c1
> amixer: Mixer load error: hw:1
> atrophy:/etc/modutils# alsamixer -c1
> 
> alsamixer: function snd_mixer_load failed: Invalid argument
> atrophy:/etc/modutils# 

hmm, could you run "alsactl store" and show the generated
/etc/asound.state (only for the second card is enough)?
at least we can see whether the controls are parsed properly.


> > and /dev/dsp1 is the OSS device for Extigy (available only if you load
> > snd-pcm-oss module).
> 
> Hmm, when I load snd-mixer-oss and snd-pcm-oss, only /dev/dsp appears as a compatibility link to sound/dsp. Sound (which im guessing is the oss stuff in the dev tree) has:
> 
> mark@atrophy:/dev/sound$ ls -l
> total 0
> crw-rw----    1 root     audio     14,   4 Dec 31  1969 audio
> crw-rw----    1 root     audio     14,  10 Dec 31  1969 dmfm
> crw-rw----    1 root     audio     14,   9 Dec 31  1969 dmmidi
> crw-rw----    1 root     audio     14,   3 Dec 31  1969 dsp
> crw-rw----    1 root     audio     14,   2 Dec 31  1969 midi00
> crw-rw----    1 root     audio     14,   0 Dec 31  1969 mixer
> crw-rw----    1 root     audio     14,  18 Dec 31  1969 unknown

ah, i forgot that you're using devfs.  you need more config entry in
devfs.conf for /dev/dsp1...  sorry, i cannot tell more, since i'm not
a devfs expert.

> 
> Which I guess makes sense since oss only supports one audio card and I
> have two . . .
> 
> > if the above doesn't work (or even if it works), please send me the
> > output of /proc/asound/card1/stream* files. 
> > this contains the supported formats on the device.
> 
> There is only a stream0:
> 
> atrophy:/proc/asound/card1# more stream0 
> Creative Technology Ltd. : USB Audio

it looks that the driver couldn't get the product name.
needs check lsusb output...

> 
> Playback:
>   Status: Stop
>   Interface 1
>     Altset 1
>     Format: S16_LE
>     Channels: 2
>     Endpoint: 2 OUT (ASYNC)
>     Rates: 48000

also, only one format was parsed..  it's strange, too.

> 
> > to the MIDI devices, you can access either via ALSA sequencer or via
> > a rawmidi device file /dev/snd/midiC1D0 (or midiC1D1), which is
> > equivalent with OSS midi device, /dev/midi01 (or amidi01).
> 
> The two devices popping up seem consistant with the hardware, it has a
> midi in/out (no thru).
> 
> > but please note that here MIDI means not the MIDI wavetable playback.
> > apparently Extigy has _no_ wavetable function at all.
> > they are connectors to external MIDI devices.
> 
> Thats ok, wavetables suck. I'd rather hook up my own bank via midi, heh.
> 
> > and finally, if the above works properly, please check what mixer
> > element corresponds to which role (as i asked in my original post).
> > you can see the mixer topology and each mixer element in
> > usbmixer_maps.c.
> 
> As shown above, amixer doesnt want to fire up. I hate to do it, but
> here is a wishlist:
> 
> - have the volume knob on the front of the device work 

usually such a "hardware" control is independent from the software
stuff, so it may work regardless what driver does.

> - be able to turn "cmss" on and off either via the controling client or on
>   the hardware itself. (its a dsp that fakes surround sound fairly well).

perhaps already implemented but just cannot be used :)
i guess that one of the unknown switch corresponds to this.

> - be able to control the output detection of the unit. apparently it
>   can tell if you have 2, 4, or 5 speakers hooked up into it. the option
>   can certainly be automatic (testable when i unplug the card and use it
>   as a standalone), and I'm fairly certain it can be forced as well (as
>   evidenced in windows). When I was able to use the hardware with the 
>   current rc release of alsa, it forced 2 speaker selection.

i think this is because only one format was parsed correctly.
the other formats with 6 channels were not parsed by some reason.

but i'm not sure how the driver can detect the plug status...

> - the remote ir stuff is cute but not really essential -- would be nice to 
>   have hooks to attach triggers to or something.

this will not be implemented on the usb audio driver.
it's a role of another input-driver, and a corresponding daemon.


ciao,

Takashi


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf