On 01/26/2010 01:32 AM, Takashi Iwai wrote:
> At Mon, 25 Jan 2010 19:19:31 -0500,
> Ken Prox wrote:
>    
>> On 01/24/2010 05:32 AM, Takashi Iwai wrote:
>>      
>>> At Tue, 12 Jan 2010 10:07:02 +0100,
>>> I wrote:
>>>
>>>        
>>>> At Fri, 8 Jan 2010 09:09:54 +0100 (CET),
>>>> Jaroslav Kysela wrote:
>>>>
>>>>          
>>>>> On Sun, 3 Jan 2010, knprox@mi.rr.com wrote:
>>>>>
>>>>>
>>>>>            
>>>>>> From: Ken Prox<kprox@users.sourceforge.net>
>>>>>>
>>>>>>
>>>>>> Signed-off-by: Ken Prox<kprox@users.sourceforge.net>
>>>>>>
>>>>>>              
>>>>> I've applied your patch to my "devel" git tree with space/tab
>>>>> replacements. Please, use same coding style matching other ALSA code and
>>>>> run "scripts/checkpatch.pl" on your patch before posting. Thanks.
>>>>>
>>>>>            
>>>> I pulled your branches now to sound git tree.
>>>>
>>>>          
>>> And, now I'm checking the code and your change again while I'm working
>>> on fixing / clean-up of cxt5051 code.  I don't understand why your patch
>>> works.  Does it really work as is?
>>>
>>> For example, the pin 0x17 has no multiple connections.  So, changing
>>> the amp for the input #1 doesn't make sense at all.  If it supports only
>>> the pin 0x1d, then the init verbs have to be different.
>>>
>>> Doesn't another model, e.g. model=hp-dv6736, work as is?
>>>
>>>        
>>      
>>> thanks,
>>>
>>> Takashi
>>>
>>>        
>> Hello Takashi,
>>
>> Yes, using model=hp-dv6736 works for device 103c:30ea, but the mixer
>> control does not control the microphone.  However, with the patch I
>> supplied, the microphone mixer control does work.
>>
>> Yes, the patch for the 103c:30ea does work as is.  If there is a change
>> to the init verbs that you would like me to try, I would be more than
>> happy to help.
>>      
> Interesting.  Could you give alsa-info.sh outputs (run with
> --no-upload option) at different mic values?
>
> The mysterious thing is that you change the pin 0x17 index 1.  This
> has no multiple sources, so the index shouldn't matter.
>
>
> thanks,
>
> Takashi
>    

I tried recompiling the driver changing the init verb from:

static struct hda_verb cxt5051_f700_init_verbs[] = {
         /* Line in, Mic */
         {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1) | 0x03},

to:

static struct hda_verb cxt5051_f700_init_verbs[] = {
         /* Line in, Mic */
         {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0) | 0x03},

The driver works fine.  I suspect that there was a persistent setting in 
my mixer that was effecting nid 0x17, and not the fact that UNMUTE was 
indexed to 0.

Attached are the output from alsa-info.sh --no-upload.

The first is with the microphone mixer set to 0%, second the differences 
with the microphone mixer set to 50%, and third the differences with the 
microphone mixer set to 100%.

Glad I could help.

-- 
"So often do the spirits of great events stride on before the events. And in today already walks tomorrow." - Coleridge

Ken

pgp     hkp://wwwkeys.eu.pgp.net
Skype   callto://ken.prox