Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[James Courtier-Dutton]

[-- Attachment #1: Type: text/plain, Size: 1195 bytes --]

Takashi Iwai wrote:

>Update of /cvsroot/alsa/alsa-kernel/pci/emu10k1
>In directory usw-pr-cvs1:/tmp/cvs-serv9192
>
>Modified Files:
>	emu10k1_main.c 
>Log Message:
>fixed default mask values
>(these change won't affect the actual behavior, though)
>
>
>
>
>Index: emu10k1_main.c
>===================================================================
>RCS file: /cvsroot/alsa/alsa-kernel/pci/emu10k1/emu10k1_main.c,v
>retrieving revision 1.12
>retrieving revision 1.13
>diff -u -r1.12 -r1.13
>--- emu10k1_main.c	13 Mar 2002 19:49:38 -0000	1.12
>+++ emu10k1_main.c	8 Apr 2002 12:20:36 -0000	1.13
>@@ -630,9 +630,9 @@
> 	
> 	emu->fx8010.fxbus_mask = 0x303f;
> 	if (extin_mask == 0)
>-		extin_mask = 0x1fcf;
>+		extin_mask = 0x3fcf;
> 	if (extout_mask == 0)
>-		extout_mask = 0x3fff;
>+		extout_mask = 0x1fff;
> 	emu->fx8010.extin_mask = extin_mask;
> 	emu->fx8010.extout_mask = extout_mask;
> 
>
This seems to have fixed AC3 Passthru mode on SB Live (rev 7) cards.

Just so you know, all SB Live cards worked before for AC3 Passthru 
except the rev 7 cards.

I just got a report on the xine-user list about it.
Ok, so you did not know you were fixing it, but well done anyway!!!! ;-)

Cheers
James

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Takashi Iwai]

At Thu, 11 Apr 2002 15:42:23 +0100,
James Courtier-Dutton wrote:
> 
> [1  <text/plain; us-ascii (7bit)>]
> Takashi Iwai wrote:
> 
> >Update of /cvsroot/alsa/alsa-kernel/pci/emu10k1
> >In directory usw-pr-cvs1:/tmp/cvs-serv9192
> >
> >Modified Files:
> >	emu10k1_main.c 
> >Log Message:
> >fixed default mask values
> >(these change won't affect the actual behavior, though)
> >
> >
> >
> >
> >Index: emu10k1_main.c
> >===================================================================
> >RCS file: /cvsroot/alsa/alsa-kernel/pci/emu10k1/emu10k1_main.c,v
> >retrieving revision 1.12
> >retrieving revision 1.13
> >diff -u -r1.12 -r1.13
> >--- emu10k1_main.c	13 Mar 2002 19:49:38 -0000	1.12
> >+++ emu10k1_main.c	8 Apr 2002 12:20:36 -0000	1.13
> >@@ -630,9 +630,9 @@
> > 	
> > 	emu->fx8010.fxbus_mask = 0x303f;
> > 	if (extin_mask == 0)
> >-		extin_mask = 0x1fcf;
> >+		extin_mask = 0x3fcf;
> > 	if (extout_mask == 0)
> >-		extout_mask = 0x3fff;
> >+		extout_mask = 0x1fff;
> > 	emu->fx8010.extin_mask = extin_mask;
> > 	emu->fx8010.extout_mask = extout_mask;
> > 
> >
> This seems to have fixed AC3 Passthru mode on SB Live (rev 7) cards.
> 
> Just so you know, all SB Live cards worked before for AC3 Passthru 
> except the rev 7 cards.
> 
> I just got a report on the xine-user list about it.
> Ok, so you did not know you were fixing it, but well done anyway!!!! ;-)

err, frankly speaking, i still don't figure out why with this change
the pass-through works.  i'm glad to to hear a good news, anyway ;)


Takashi

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

Hi Takashi!

* Takashi Iwai writes:
> At Thu, 11 Apr 2002 15:42:23 +0100,
> James Courtier-Dutton wrote:
>> 
>> [1  <text/plain; us-ascii (7bit)>]
>> Takashi Iwai wrote:
>> 
>> >Update of /cvsroot/alsa/alsa-kernel/pci/emu10k1
>> >In directory usw-pr-cvs1:/tmp/cvs-serv9192
>> >
>> >Modified Files:
>> >	emu10k1_main.c 
>> >Log Message:
>> >fixed default mask values
>> >(these change won't affect the actual behavior, though)

[snip]

>> This seems to have fixed AC3 Passthru mode on SB Live (rev 7) cards.
>> Just so you know, all SB Live cards worked before for AC3 Passthru 
>> except the rev 7 cards.
>> I just got a report on the xine-user list about it.
>> Ok, so you did not know you were fixing it, but well done anyway!!!! ;-)
> err, frankly speaking, i still don't figure out why with this change
> the pass-through works.  i'm glad to to hear a good news, anyway ;)

Just wanted to let you know that it wasn't the changes in emu10k1_main.c,
but the changes in emumixer.c that made my (I was the one reporting it on
the xine ML) rev07 card working.

Perhaps this helps
Regards
Markus

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Jaroslav Kysela]

On Mon, 15 Apr 2002, Markus Plail wrote:

> Hi Takashi!
> 
> * Takashi Iwai writes:
> > At Thu, 11 Apr 2002 15:42:23 +0100,
> > James Courtier-Dutton wrote:
> >> 
> >> [1  <text/plain; us-ascii (7bit)>]
> >> Takashi Iwai wrote:
> >> 
> >> >Update of /cvsroot/alsa/alsa-kernel/pci/emu10k1
> >> >In directory usw-pr-cvs1:/tmp/cvs-serv9192
> >> >
> >> >Modified Files:
> >> >	emu10k1_main.c 
> >> >Log Message:
> >> >fixed default mask values
> >> >(these change won't affect the actual behavior, though)
> 
> [snip]
> 
> >> This seems to have fixed AC3 Passthru mode on SB Live (rev 7) cards.
> >> Just so you know, all SB Live cards worked before for AC3 Passthru 
> >> except the rev 7 cards.
> >> I just got a report on the xine-user list about it.
> >> Ok, so you did not know you were fixing it, but well done anyway!!!! ;-)
> > err, frankly speaking, i still don't figure out why with this change
> > the pass-through works.  i'm glad to to hear a good news, anyway ;)
> 
> Just wanted to let you know that it wasn't the changes in emu10k1_main.c,
> but the changes in emumixer.c that made my (I was the one reporting it on
> the xine ML) rev07 card working.

The joke is, that this behaviour is caused by a mistake. Here is the right 
patch against current CVS:

Index: emumixer.c
===================================================================
RCS file: /cvsroot/alsa/alsa-kernel/pci/emu10k1/emumixer.c,v
retrieving revision 1.9
diff -u -r1.9 emumixer.c
--- emumixer.c  8 Apr 2002 12:21:45 -0000       1.9
+++ emumixer.c  16 Apr 2002 07:10:59 -0000
@@ -369,7 +369,7 @@
        int change;

        spin_lock_irqsave(&emu->reg_lock, flags);
-       if (emu->audigy) {
+       if (!emu->audigy) {
                reg = inl(emu->port + HCFG);
                val = ucontrol->value.integer.value[0] & 1 ? 0 : HCFG_GPOUT0;
                change = (reg & HCFG_GPOUT0) != val;


At least, we know that for rev07 cards we shouldn't modify GPOUT0. These 
cards have separated the digital output or shared with an analog output?


						Jaroslav

-----
Jaroslav Kysela <perex@suse.cz>
Linux Kernel Sound Maintainer
ALSA Project  http://www.alsa-project.org
SuSE Linux    http://www.suse.com

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Takashi Iwai]

At Tue, 16 Apr 2002 09:15:10 +0200 (CEST),
Jaroslav wrote:
> 
> On Mon, 15 Apr 2002, Markus Plail wrote:
> 
> > Hi Takashi!
> > 
> > * Takashi Iwai writes:
> > > At Thu, 11 Apr 2002 15:42:23 +0100,
> > > James Courtier-Dutton wrote:
> > >> 
> > >> [1  <text/plain; us-ascii (7bit)>]
> > >> Takashi Iwai wrote:
> > >> 
> > >> >Update of /cvsroot/alsa/alsa-kernel/pci/emu10k1
> > >> >In directory usw-pr-cvs1:/tmp/cvs-serv9192
> > >> >
> > >> >Modified Files:
> > >> >	emu10k1_main.c 
> > >> >Log Message:
> > >> >fixed default mask values
> > >> >(these change won't affect the actual behavior, though)
> > 
> > [snip]
> > 
> > >> This seems to have fixed AC3 Passthru mode on SB Live (rev 7) cards.
> > >> Just so you know, all SB Live cards worked before for AC3 Passthru 
> > >> except the rev 7 cards.
> > >> I just got a report on the xine-user list about it.
> > >> Ok, so you did not know you were fixing it, but well done anyway!!!! ;-)
> > > err, frankly speaking, i still don't figure out why with this change
> > > the pass-through works.  i'm glad to to hear a good news, anyway ;)
> > 
> > Just wanted to let you know that it wasn't the changes in emu10k1_main.c,
> > but the changes in emumixer.c that made my (I was the one reporting it on
> > the xine ML) rev07 card working.
> 
> The joke is, that this behaviour is caused by a mistake. Here is the right 
> patch against current CVS:

wow :)


> At least, we know that for rev07 cards we shouldn't modify GPOUT0. These 
> cards have separated the digital output or shared with an analog output?

Markus, could you try the latest cvs version, which fixes the "real"
behavior, whether digital out works?
if it doesn't work again, then definitely GPOUT0 is the spot.

looking at OSS source codes, i found GPOUT0 is necessary only for
5.1 cards (and seems for Audigy too - i need to fix again).

so i'm not sure whether we can determine it from revision number.


Takashi

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

Hi Takashi!

* Takashi Iwai writes:
> Markus, could you try the latest cvs version, which fixes the "real"
> behavior, whether digital out works?  if it doesn't work again, then
> definitely GPOUT0 is the spot.

It doesn't work with latest CVS.

> looking at OSS source codes, i found GPOUT0 is necessary only for 5.1
> cards (and seems for Audigy too - i need to fix again).
> so i'm not sure whether we can determine it from revision number.

Someone should test if the version which is doing fine here, is also OK
on other revisions of the card, so 'we' could just leave the 'else
statement' away. James has a rev06 to check it, for example.

Regards
Markus

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[James Courtier-Dutton]

Markus Plail wrote:

>Hi Takashi!
>
>* Takashi Iwai writes:
>  
>
>>Markus, could you try the latest cvs version, which fixes the "real"
>>behavior, whether digital out works?  if it doesn't work again, then
>>definitely GPOUT0 is the spot.
>>    
>>
>
>It doesn't work with latest CVS.
>
>  
>
>>looking at OSS source codes, i found GPOUT0 is necessary only for 5.1
>>cards (and seems for Audigy too - i need to fix again).
>>so i'm not sure whether we can determine it from revision number.
>>    
>>
>
>Someone should test if the version which is doing fine here, is also OK
>on other revisions of the card, so 'we' could just leave the 'else
>statement' away. James has a rev06 to check it, for example.
>
>Regards
>Markus
>  
>
There are several different versions of the SB Live.
My SB Live rev06 is not a SB Live 5.1, so it has 2 audio stereo output 
jacks, one for front, one for rear, and then a spdif jack. The spdif 
jack is not switchable between analogue and digital. It is fixed at digital.

Maybe the rev07 card is a SB Live 5.1, which has a switchable 
digital/analogue spdif jack, so it can output on 3 analogue stereo 
jacks, or with a switch (maybe GPOUT) it changes the 3rd analogue out 
jack into an spdif digital only jack.

Does this info help at all?

Cheers
James

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

* James Courtier-Dutton writes:

> There are several different versions of the SB Live.
> My SB Live rev06 is not a SB Live 5.1, so it has 2 audio stereo output
> jacks, one for front, one for rear, and then a spdif jack. The spdif
> jack is not switchable between analogue and digital. It is fixed at
> digital.

I haven't used it as analogue outout, but I guess that it can be
switched.. at least the digital output can be switched off (don't know
if it switches analogue output on, though).

> Maybe the rev07 card is a SB Live 5.1, which has a switchable
> digital/analogue spdif jack, so it can output on 3 analogue stereo
> jacks, or with a switch (maybe GPOUT) it changes the 3rd analogue out
> jack into an spdif digital only jack.

I bought it as 'Soundblaster Live Player 5.1'.

Regards
Markus

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[James Courtier-Dutton]

Markus Plail wrote:

>* James Courtier-Dutton writes:
>
>  
>
>>There are several different versions of the SB Live.
>>My SB Live rev06 is not a SB Live 5.1, so it has 2 audio stereo output
>>jacks, one for front, one for rear, and then a spdif jack. The spdif
>>jack is not switchable between analogue and digital. It is fixed at
>>digital.
>>    
>>
>
>I haven't used it as analogue outout, but I guess that it can be
>switched.. at least the digital output can be switched off (don't know
>if it switches analogue output on, though).
>
>  
>
>>Maybe the rev07 card is a SB Live 5.1, which has a switchable
>>digital/analogue spdif jack, so it can output on 3 analogue stereo
>>jacks, or with a switch (maybe GPOUT) it changes the 3rd analogue out
>>jack into an spdif digital only jack.
>>    
>>
>
>I bought it as 'Soundblaster Live Player 5.1'.
>
>Regards
>Markus
>
>
>  
>
This think the intention for the SB Live 5.1 with alsa, was that 
alsamixer would display an extra switch for analogue/digital out which 
would switch the digital output between digital spdif and analogue 
stereo. Did you ever look in alsamixer for such a switch ?

Cheers
James

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

Hi James!

* James Courtier-Dutton writes:
> Markus Plail wrote:
>> * James Courtier-Dutton writes:
>> 
>> 
>>> There are several different versions of the SB Live.
>>> My SB Live rev06 is not a SB Live 5.1, so it has 2 audio stereo output
>>> jacks, one for front, one for rear, and then a spdif jack. The spdif
>>> jack is not switchable between analogue and digital. It is fixed at
>>> digital.
>>> 
>> 
>> I haven't used it as analogue outout, but I guess that it can be
>> switched.. at least the digital output can be switched off (don't know
>> if it switches analogue output on, though).
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

> This think the intention for the SB Live 5.1 with alsa, was that
> alsamixer would display an extra switch for analogue/digital out which
> would switch the digital output between digital spdif and analogue
> stereo. Did you ever look in alsamixer for such a switch ?

Yes. See above. That was the very switch I was talking about. the switch
is working properly -> I can turn digital sound off. As mentioned before
I have never used its analog ability further than stereo.

Regards
Markus

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Takashi Iwai]

Hi James,

At Tue, 16 Apr 2002 14:27:07 +0100,
James Courtier-Dutton wrote:
> 
> Markus Plail wrote:
> 
> >Hi Takashi!
> >
> >* Takashi Iwai writes:
> >  
> >
> >>Markus, could you try the latest cvs version, which fixes the "real"
> >>behavior, whether digital out works?  if it doesn't work again, then
> >>definitely GPOUT0 is the spot.
> >>    
> >>
> >
> >It doesn't work with latest CVS.
> >
> >  
> >
> >>looking at OSS source codes, i found GPOUT0 is necessary only for 5.1
> >>cards (and seems for Audigy too - i need to fix again).
> >>so i'm not sure whether we can determine it from revision number.
> >>    
> >>
> >
> >Someone should test if the version which is doing fine here, is also OK
> >on other revisions of the card, so 'we' could just leave the 'else
> >statement' away. James has a rev06 to check it, for example.
> >
> >Regards
> >Markus
> >  
> >
> There are several different versions of the SB Live.
> My SB Live rev06 is not a SB Live 5.1, so it has 2 audio stereo output 
> jacks, one for front, one for rear, and then a spdif jack. The spdif 
> jack is not switchable between analogue and digital. It is fixed at digital.
 
does "Digital/Analog Jack" switch affects the behavior on your digital
jack?  i guess it doesn't.
if it doesn't affect, then we can switch this bit safely on all
cards.


Takashi

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

Hi Takashi!

* Takashi Iwai writes:
> Hi James,

> does "Digital/Analog Jack" switch affects the behavior on your digital
> jack?  i guess it doesn't.
> if it doesn't affect, then we can switch this bit safely on all
> cards.

So you thought that my problem was this not being enabled? I can assure
you then, that this wasn't the problem. I have played with that switch
all way through my alsa SB live experience. It worked properly to switch
digital sound off and the default value (i.e. after having asound.state
deleted) was/is digital sound ON.

Regards
Markus

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[James Courtier-Dutton]

Takashi Iwai wrote:

>>There are several different versions of the SB Live.
>>My SB Live rev06 is not a SB Live 5.1, so it has 2 audio stereo output 
>>jacks, one for front, one for rear, and then a spdif jack. The spdif 
>>jack is not switchable between analogue and digital. It is fixed at digital.
>>    
>>
> 
>does "Digital/Analog Jack" switch affects the behavior on your digital
>jack?  i guess it doesn't.
>if it doesn't affect, then we can switch this bit safely on all
>cards.
>
>
>Takashi
>
>  
>
My SB Live (non 5.1) does not have a Digital/Analog Jack, it just has a 
Digital Jack, so I would prefer that the Digital/Analogue Switch in 
alsamixer was removed for rev06 cards and below, but present for rev7 
cards and above.

Currently, toggling the alsamixer Digital/Analogue Switch has no effect 
what so ever.

I think this rev07 passthru issue might have been the user just not 
knowing about the Digital/Analogue switch in alsamixer.

Cheers
James

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

Hi James!

* James Courtier-Dutton writes:
> I think this rev07 passthru issue might have been the user just not
> knowing about the Digital/Analogue switch in alsamixer.

NO! I have been knowing about it all the time. And as I wrote to Takashi
this had/has nothing to do with my problem. The switch worked, but AC3
wasn't passed through, so this is surely not the readon. I wouldn't have
gotten any sound then, or better only digital enabled sound, because the
preset is digital sound 'ON'.

Regards
Markus Plail

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[James Courtier-Dutton]

Markus Plail wrote:

>Hi James!
>
>* James Courtier-Dutton writes:
>  
>
>>I think this rev07 passthru issue might have been the user just not
>>knowing about the Digital/Analogue switch in alsamixer.
>>    
>>
>
>NO! I have been knowing about it all the time. And as I wrote to Takashi
>this had/has nothing to do with my problem. The switch worked, but AC3
>wasn't passed through, so this is surely not the readon. I wouldn't have
>gotten any sound then, or better only digital enabled sound, because the
>preset is digital sound 'ON'.
>
>Regards
>Markus Plail
>
>  
>
Sorry, it was only an idle thought. Maybe the Digital/Analogue switch 
never worked until now.

Cheers
James

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

Hi James!

* James Courtier-Dutton writes:
> Markus Plail wrote:

> Sorry, it was only an idle thought. Maybe the Digital/Analogue switch
> never worked until now.

No offense intended/taken, but you also have to understand that I was
struggling for ages with that AC3 passthrough, really trying everything,
giving bug reports and offering my help and then someone says: "I think
he just didn't see that switch." ...  I hope you understand.
And about that switch. It worked perfectly well, i.e. switching
'normal' digital outout on/off, since I started using alsa.

Regards
Markus

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Takashi Iwai]

At Tue, 16 Apr 2002 14:47:35 +0200,
Markus Plail wrote:
> 
> Hi Takashi!
> 
> * Takashi Iwai writes:
> > Markus, could you try the latest cvs version, which fixes the "real"
> > behavior, whether digital out works?  if it doesn't work again, then
> > definitely GPOUT0 is the spot.
> 
> It doesn't work with latest CVS.

ok, now it's clear.
interesting, perhaps on alsa implemented in the inversed manner.
if my guess is correct, you'll get digital outs by changing
/usr/share/alsa/cards/EMU10K1.conf like the following:

--- /usr/share/alsa/cards/EMU10K1.conf	2002-03-22 16:05:56.000000000 +0100
+++ EMU10K1.conf	2002-04-16 16:23:28.000000000 +0200
@@ -240,7 +240,7 @@
 				name "SB Live Analog/Digital Output Jack"
 				lock true
 				preserve true
-				value 1
+				value 0
 			}
 		]
 	}


please check this?


ciao,

Takashi

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

Hi Takashi!

* Takashi Iwai writes:
> ok, now it's clear.

> interesting, perhaps on alsa implemented in the inversed manner.  if
> my guess is correct, you'll get digital outs by changing
> /usr/share/alsa/cards/EMU10K1.conf like the following:
> --- /usr/share/alsa/cards/EMU10K1.conf	2002-03-22 16:05:56.000000000 +0100
> +++ EMU10K1.conf	2002-04-16 16:23:28.000000000 +0200
> @@ -240,7 +240,7 @@
>  				name "SB Live Analog/Digital Output Jack"
>  				lock true
>  				preserve true
> -				value 1
> +				value 0
>  			}
>  		]
>  	}
> please check this?

Doesn't change anything really. I did an 'alsasound stop', edited
EMU10K1.conf and then 'alsasound start'. Was that right?

Regards
Markus

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Takashi Iwai]

At Tue, 16 Apr 2002 16:42:30 +0200,
Markus Plail wrote:
> 
> Hi Takashi!
> 
> * Takashi Iwai writes:
> > ok, now it's clear.
> 
> > interesting, perhaps on alsa implemented in the inversed manner.  if
> > my guess is correct, you'll get digital outs by changing
> > /usr/share/alsa/cards/EMU10K1.conf like the following:
> > --- /usr/share/alsa/cards/EMU10K1.conf	2002-03-22 16:05:56.000000000 +0100
> > +++ EMU10K1.conf	2002-04-16 16:23:28.000000000 +0200
> > @@ -240,7 +240,7 @@
> >  				name "SB Live Analog/Digital Output Jack"
> >  				lock true
> >  				preserve true
> > -				value 1
> > +				value 0
> >  			}
> >  		]
> >  	}
> > please check this?
> 
> Doesn't change anything really. I did an 'alsasound stop', edited
> EMU10K1.conf and then 'alsasound start'. Was that right?

yes.  now please check the state of this switch by alsamixer.
try to play raw data once, and check whether the switch is changed.
alsamixer will show the current status (and change the status if any
notified).

if this doesn't work..  hmm..  needs more consideration.


Takashi

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

Hi Takashi!

* Takashi Iwai writes:
> yes.  now please check the state of this switch by alsamixer.  try to
> play raw data once, and check whether the switch is changed.
> alsamixer will show the current status (and change the status if any
> notified).

This is the same as if I changed the switch in alsamixer by hand, right?
This does not work. I have done this before thousands of times. The
switch is working fine, but has nothing to do with my problem
(IMHO). Only if I change emumixer the way you did by accident I get AC3
passthrough.
If AC3 passthrough is enabled, i.e. I watch a DVD with xine, the
digital/analog switch doesn't do anything, i.e. I can't turn off the
audio. If I play for example a mp3 with xmms, the digital/analog switch
is working as expected.

Regards
Markus

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[James Courtier-Dutton]

Markus Plail wrote:

>Hi Takashi!
>
>* Takashi Iwai writes:
>  
>
>>yes.  now please check the state of this switch by alsamixer.  try to
>>play raw data once, and check whether the switch is changed.
>>alsamixer will show the current status (and change the status if any
>>notified).
>>    
>>
>
>This is the same as if I changed the switch in alsamixer by hand, right?
>This does not work. I have done this before thousands of times. The
>switch is working fine, but has nothing to do with my problem
>(IMHO). Only if I change emumixer the way you did by accident I get AC3
>passthrough.
>If AC3 passthrough is enabled, i.e. I watch a DVD with xine, the
>digital/analog switch doesn't do anything, i.e. I can't turn off the
>audio. If I play for example a mp3 with xmms, the digital/analog switch
>is working as expected.
>
>Regards
>Markus
>  
>
I think what Takashi is saying is that when xine attempts to do 
passthru, it should change the switch automatically. It will in fact 
override whatever you entered in alsamixer. So maybe before it was 
overriding it to "analogue" instead of overriding it to "digital".

I think that is what Takashi was trying to say.

Cheers
James

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

Hi James!

* James Courtier-Dutton writes:

> I think what Takashi is saying is that when xine attempts to do
> passthru, it should change the switch automatically. It will in fact
> override whatever you entered in alsamixer. So maybe before it was
> overriding it to "analogue" instead of overriding it to "digital".
> I think that is what Takashi was trying to say.

OK, but then again xine didn't change, but alsa, but then again I
propbably just do not have the knowledge to really judge what is going
on.

Regards
Markus

BTW: Does your rev06 card work with the 'wrong' patch from Takashi?

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Markus Plail]

Hi Takashi!

* Takashi Iwai writes:
> At Tue, 16 Apr 2002 16:42:30 +0200,
> Markus Plail wrote:

>> Doesn't change anything really. I did an 'alsasound stop', edited
>> EMU10K1.conf and then 'alsasound start'. Was that right?

I am very sorry, I must have missed something yesterday. It DOES indeed
work with the change made in EMU10K1.conf

> yes.  now please check the state of this switch by alsamixer.
> try to play raw data once, and check whether the switch is changed.
> alsamixer will show the current status (and change the status if any
> notified).

As I wrote in another mail: While playing in passthrough mode (xine) the
switch doesn't do anything, while it works properly with normal (xmms
mp3 for example) digital mode.

Sorry for the confusion
Regards
Markus

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Jaroslav Kysela]

On Wed, 17 Apr 2002, Markus Plail wrote:

> Hi Takashi!
> 
> * Takashi Iwai writes:
> > At Tue, 16 Apr 2002 16:42:30 +0200,
> > Markus Plail wrote:
> 
> >> Doesn't change anything really. I did an 'alsasound stop', edited
> >> EMU10K1.conf and then 'alsasound start'. Was that right?
> 
> I am very sorry, I must have missed something yesterday. It DOES indeed
> work with the change made in EMU10K1.conf
> 
> > yes.  now please check the state of this switch by alsamixer.
> > try to play raw data once, and check whether the switch is changed.
> > alsamixer will show the current status (and change the status if any
> > notified).
> 
> As I wrote in another mail: While playing in passthrough mode (xine) the
> switch doesn't do anything, while it works properly with normal (xmms
> mp3 for example) digital mode.

This behaviour is ok. When the raw S/PDIF device is used (passthough for 
EMU10K1), the switch is locked and can't be modified by another 
application.

So, we know, at least, that for your card this switch must be off. 
Perhaps, it's only inverted? Can somebody with shared s/pdif, line-out 
confirm this?

						Jaroslav

-----
Jaroslav Kysela <perex@suse.cz>
Linux Kernel Sound Maintainer
ALSA Project  http://www.alsa-project.org
SuSE Linux    http://www.suse.com

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[Takashi Iwai]

At Wed, 17 Apr 2002 10:33:59 +0200 (CEST),
Jaroslav wrote:
> 
> On Wed, 17 Apr 2002, Markus Plail wrote:
> 
> > Hi Takashi!
> > 
> > * Takashi Iwai writes:
> > > At Tue, 16 Apr 2002 16:42:30 +0200,
> > > Markus Plail wrote:
> > 
> > >> Doesn't change anything really. I did an 'alsasound stop', edited
> > >> EMU10K1.conf and then 'alsasound start'. Was that right?
> > 
> > I am very sorry, I must have missed something yesterday. It DOES indeed
> > work with the change made in EMU10K1.conf
> > 
> > > yes.  now please check the state of this switch by alsamixer.
> > > try to play raw data once, and check whether the switch is changed.
> > > alsamixer will show the current status (and change the status if any
> > > notified).
> > 
> > As I wrote in another mail: While playing in passthrough mode (xine) the
> > switch doesn't do anything, while it works properly with normal (xmms
> > mp3 for example) digital mode.
> 
> This behaviour is ok. When the raw S/PDIF device is used (passthough for 
> EMU10K1), the switch is locked and can't be modified by another 
> application.
> 
> So, we know, at least, that for your card this switch must be off. 
> Perhaps, it's only inverted?

i believe yes.  compared with oss code, i found the bit is flipped.
i'll check in the fix now.


Takashi

Re: Re: [alsa-cvslog] CVS: alsa-kernel/pci/emu10k1 emu10k1_main.c,1.12,1.13

[James Courtier-Dutton]

>
>
>>This behaviour is ok. When the raw S/PDIF device is used (passthough for 
>>EMU10K1), the switch is locked and can't be modified by another 
>>application.
>>
>>So, we know, at least, that for your card this switch must be off. 
>>Perhaps, it's only inverted?
>>    
>>
>
>i believe yes.  compared with oss code, i found the bit is flipped.
>i'll check in the fix now.
>
>
>Takashi
>
>  
>
Could all this be the long lost solution to those "Gateway" versions of 
the SB Live card, which have also never worked with alsa for passthru.
I don't have one of the "Gateway" cards, but I remember them being 
discussed on this list a long time ago.

Cheers
James