Re: potential emu10k1 bug/issue for Audigy2 ZS Platinum Pro

Re: potential emu10k1 bug/issue for Audigy2 ZS Platinum Pro

[Lee Revell]

[forwarding to alsa-devel]

On Thu, 2005-02-10 at 11:26 -0500, Rick Wright wrote:
> Hi Lee,
> 
> Sorry to keep bothering you re the emu10k1 driver, but I've finally had 
> some success with multichannel I/O and now think I have found some 
> bugs.  I really want to thank you for your efforts in maturing this driver.
> 
> I've never filed a bug report for anything in Linux, but I would like to 
> contribute to the community.  Could you give me an overview of the 
> process so that I can use that avenue for all future issues I find?
> 
> I've finally gotten the multichannel capture to show me 16 channels by 
> default like everyone assured me it should do.  Not sure how this 
> happened.  I was fooling with trying to use ALSA CVS and apply your v009 
> patch (per your suggestions), but couldn't get it to compile (newb 
> problem..).  Anyway, I finally reverted to ALSA 1.0.8 patched with your 
> v008 and suddenly it gave me 16 capture channels using the original 
> asound.state file that I previously had only 2 cpature channels with (I 
> had backed it up, then copied it back after reinstalling the patched 
> 1.0.8).  I've no idea what happened to make it work, but work it does!
> 
> The problem:
> The Left & Right PCM channel volume controls (gnome-control-volume) are 
> reversed for each of the L/R stereo output signals.
> 
> I remember you mentioned that you couldn't guarantee all the correct 
> channel mapping for every card out there, so here is my contribution.   
> I've done some thorough testing to map capture/playback ports to 
> physical channels on my Audigy2 Platinum Pro card w/external connection 
> box and here are my results (this is both FYI and relevant info to the 
> bug/issue I believe I've found):
> 
> Inputs to PCM Capture ports:            (with my 5.1 setup -- I tried to 
> keep the input routing is consistent with the 5.1 output routing)
> Line In #1 Left     =    capture_10      (LF)
> Line In #1 Right   =    capture_9        (RF)
> Line In #2 Left     =    capture_12      (LR)
> Line In #2 Right   =    capture_11      (RR)
> Line In #3 Left     =    capture_13      (Cen)
> Line In #3 Right   =    capture_14      (LFE)
> 
> Outputs to PCM Playback ports:                        (5.1 setup 
> described in Audigy2 product literature)
> Stereo Output #1 Left     =    playback_10         (LF)
> Stereo Output #1 Right   =    playback_9           (RF)
> Stereo Output #2 Left     =    playback_4           (LR)
> Stereo Output #2 Right   =    playback_3           (RR)
> Stereo Output #3 Left     =    playback_8           (Cen)
> Stereo Output #3 Right   =    playback_7           (LFE)
> 
> The first observation is that it's odd that all of the L/R channel pairs 
> are backwards from the typical Left-comes-first configuration (except 
> Line In #3)
> 
> The real problem is that the PCM output volume control sliders are 
> backwards (or perhaps the remapping the channels to L/R, rather than R/L 
> will fix this).  For example: PCM Front L slider controls the volume of 
> playback_9 which is the RIGHT Front signal, PCM Front R slider controls 
> playback_10 which is the LEFT Front signal.  This is true for all of the 
> Stereo Output channel pairs.  If you think about it, the PCM sliders are 
> actually controlling the correct physical L & R outputs on the card, but 
> the alsa_pcm:playback signals being routed to those physical channels 
> are reversed, make sense?
> 
> Is this a bug?  If so, what is the proper place/way to report it so that 
> I don't use any more of your time?  Has it already been fixed?  I know 
> you mentioned v008 still had problems...
> 
> Many thanks for both your work and your advice,
> -Rick
> 

ISTR that this is a hardware bug, and the kX driver works around it.  I
will look into it.

Are the channels reversed if you use JACK with the default stereo PCM,
hw:0,0?

Lee



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: potential emu10k1 bug/issue for Audigy2 ZS Platinum Pro

[Lee Revell]

On Thu, 2005-02-10 at 16:20 -0500, Rick Wright wrote:
> > > The real problem is that the PCM output volume control sliders are 
> > > backwards (or perhaps the remapping the channels to L/R, rather than R/L 
> > > will fix this).  For example: PCM Front L slider controls the volume of 
> > > playback_9 which is the RIGHT Front signal, PCM Front R slider controls 
> > > playback_10 which is the LEFT Front signal.  This is true for all of the 
> > > Stereo Output channel pairs.  If you think about it, the PCM sliders are 
> > > actually controlling the correct physical L & R outputs on the card, but 
> > > the alsa_pcm:playback signals being routed to those physical channels 
> > > are reversed, make sense?
> > > 
> > > Is this a bug?  If so, what is the proper place/way to report it so that 
> > > I don't use any more of your time?  Has it already been fixed?  I know 
> > > you mentioned v008 still had problems...
> > > 
> > >     
> > 
> > ISTR that this is a hardware bug, and the kX driver works around it.  I
> > will look into it.
> > 
> > Are the channels reversed if you use JACK with the default stereo PCM,
> > hw:0,0?
> > 
> >   
> Using hw:0,0 for BOTH the Input & Output devices in QJackCtl:
> (actually, input probably doesn't matter, huh?)
> playback_1 maps to Stereo Output #1 RIGHT and volume is controlled by
> the PCM LEFT slider
> playback_2 maps to Stereo Output #1 LEFT and volume is controlled by
> the PCM RIGHT slider
> 
> In this case, it seems that slider is controlling the correct
> playback_x stream, but the stream is mapped to the wrong stereo
> channel of the soundcard.

This does look like an upstream bug.  Can you verify that the channels
are still reversed without my patch?

Lee



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: potential emu10k1 bug/issue for Audigy2 ZS Platinum Pro

[Lee Revell]

On Thu, 2005-02-10 at 17:35 -0500, Rick Wright wrote:

> I will test without the patch tomorrow and report results then.
> 
> Just curious, what do you mean by upstream bug?
> 

Please make sure to cc: alsa-devel on all responses.

I mean that it's a bug in ALSA and has nothing to do with my patch.

Lee





-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: potential emu10k1 bug/issue for Audigy2 ZS Platinum Pro

[Rick Wright]

[-- Attachment #1: Type: text/plain, Size: 2701 bytes --]

Lee Revell wrote:

>On Thu, 2005-02-10 at 16:20 -0500, Rick Wright wrote:
>  
>
>>>>The real problem is that the PCM output volume control sliders are 
>>>>backwards (or perhaps the remapping the channels to L/R, rather than R/L 
>>>>will fix this).  For example: PCM Front L slider controls the volume of 
>>>>playback_9 which is the RIGHT Front signal, PCM Front R slider controls 
>>>>playback_10 which is the LEFT Front signal.  This is true for all of the 
>>>>Stereo Output channel pairs.  If you think about it, the PCM sliders are 
>>>>actually controlling the correct physical L & R outputs on the card, but 
>>>>the alsa_pcm:playback signals being routed to those physical channels 
>>>>are reversed, make sense?
>>>>
>>>>Is this a bug?  If so, what is the proper place/way to report it so that 
>>>>I don't use any more of your time?  Has it already been fixed?  I know 
>>>>you mentioned v008 still had problems...
>>>>
>>>>    
>>>>        
>>>>
>>>ISTR that this is a hardware bug, and the kX driver works around it.  I
>>>will look into it.
>>>
>>>Are the channels reversed if you use JACK with the default stereo PCM,
>>>hw:0,0?
>>>
>>>  
>>>      
>>>
>>Using hw:0,0 for BOTH the Input & Output devices in QJackCtl:
>>(actually, input probably doesn't matter, huh?)
>>playback_1 maps to Stereo Output #1 RIGHT and volume is controlled by
>>the PCM LEFT slider
>>playback_2 maps to Stereo Output #1 LEFT and volume is controlled by
>>the PCM RIGHT slider
>>
>>In this case, it seems that slider is controlling the correct
>>playback_x stream, but the stream is mapped to the wrong stereo
>>channel of the soundcard.
>>    
>>
>
>This does look like an upstream bug.  Can you verify that the channels
>are still reversed without my patch?
>
>Lee
>
>
>  
>
Using a stock (unpatched) alsa-driver-1.0.8 (verified that hw:0,3 is not 
available), here are the 2ch stereo output (using output device hw:0,0 
in qjackctl) results:

playback_1 maps to Stereo Output #1 RIGHT and volume is controlled by 
Front (NOT PCM Front) LEFT slider
playback_2 maps to Stereo Output #2 LEFT and volume is controlled by 
Front (NOT PCM Front) RIGHT slider

PCM Front sliders have no effect.  However, a slider named PCM (there 
are two of these, I'm talking about the one on the right, next to PCM 
Center....I believe the one on the left controls feedthrough?) behaves 
exactly like the Front sliders described above, but seems to supercede 
the Front sliders.

Again in this unpatched case, the sliders seem to control the correct 
playback_x channel (assuming L-comes-first), but the playback_x channels 
are being routed to the sound card's physical output channels 
incorrectly (L/R backwards).

-Rick

[-- Attachment #2: Type: text/html, Size: 3179 bytes --]

Re: potential emu10k1 bug/issue for Audigy2 ZS Platinum Pro

[Rick Wright]

[-- Attachment #1: Type: text/plain, Size: 3050 bytes --]

Rick Wright wrote:

> Lee Revell wrote:
>
>>On Thu, 2005-02-10 at 16:20 -0500, Rick Wright wrote:
>>  
>>
>>>>>The real problem is that the PCM output volume control sliders are 
>>>>>backwards (or perhaps the remapping the channels to L/R, rather than R/L 
>>>>>will fix this).  For example: PCM Front L slider controls the volume of 
>>>>>playback_9 which is the RIGHT Front signal, PCM Front R slider controls 
>>>>>playback_10 which is the LEFT Front signal.  This is true for all of the 
>>>>>Stereo Output channel pairs.  If you think about it, the PCM sliders are 
>>>>>actually controlling the correct physical L & R outputs on the card, but 
>>>>>the alsa_pcm:playback signals being routed to those physical channels 
>>>>>are reversed, make sense?
>>>>>
>>>>>Is this a bug?  If so, what is the proper place/way to report it so that 
>>>>>I don't use any more of your time?  Has it already been fixed?  I know 
>>>>>you mentioned v008 still had problems...
>>>>>
>>>>>    
>>>>>        
>>>>>
>>>>ISTR that this is a hardware bug, and the kX driver works around it.  I
>>>>will look into it.
>>>>
>>>>Are the channels reversed if you use JACK with the default stereo PCM,
>>>>hw:0,0?
>>>>
>>>>  
>>>>      
>>>>
>>>Using hw:0,0 for BOTH the Input & Output devices in QJackCtl:
>>>(actually, input probably doesn't matter, huh?)
>>>playback_1 maps to Stereo Output #1 RIGHT and volume is controlled by
>>>the PCM LEFT slider
>>>playback_2 maps to Stereo Output #1 LEFT and volume is controlled by
>>>the PCM RIGHT slider
>>>
>>>In this case, it seems that slider is controlling the correct
>>>playback_x stream, but the stream is mapped to the wrong stereo
>>>channel of the soundcard.
>>>    
>>>
>>
>>This does look like an upstream bug.  Can you verify that the channels
>>are still reversed without my patch?
>>
>>Lee
>>
>>
>>  
>>
> Using a stock (unpatched) alsa-driver-1.0.8 (verified that hw:0,3 is 
> not available), here are the 2ch stereo output (using output device 
> hw:0,0 in qjackctl) results:
>
> playback_1 maps to Stereo Output #1 RIGHT and volume is controlled by 
> Front (NOT PCM Front) LEFT slider
> playback_2 maps to Stereo Output #2 LEFT and volume is controlled by 
> Front (NOT PCM Front) RIGHT slider
>
> PCM Front sliders have no effect.  However, a slider named PCM (there 
> are two of these, I'm talking about the one on the right, next to PCM 
> Center....I believe the one on the left controls feedthrough?) behaves 
> exactly like the Front sliders described above, but seems to supercede 
> the Front sliders.
>
> Again in this unpatched case, the sliders seem to control the correct 
> playback_x channel (assuming L-comes-first), but the playback_x 
> channels are being routed to the sound card's physical output channels 
> incorrectly (L/R backwards).
>
> -Rick


Just went back and checked.  The behavior described above is exactly the 
same when I use cvs alsa-driver from 2/10/05 patched with Lee's v009.

Looks like this is an upstream bug and Lee's patch has no effect on this 
behavior.

-Rick

[-- Attachment #2: Type: text/html, Size: 3643 bytes --]

Re: potential emu10k1 bug/issue for Audigy2 ZS Platinum Pro

[Rick Wright]

At 05:12 PM 2/10/2005 2/10/2005, Lee Revell wrote:
>On Thu, 2005-02-10 at 16:20 -0500, Rick Wright wrote:
> > > > The real problem is that the PCM output volume control sliders are
> > > > backwards (or perhaps the remapping the channels to L/R, rather 
> than R/L
> > > > will fix this).  For example: PCM Front L slider controls the 
> volume of
> > > > playback_9 which is the RIGHT Front signal, PCM Front R slider 
> controls
> > > > playback_10 which is the LEFT Front signal.  This is true for all 
> of the
> > > > Stereo Output channel pairs.  If you think about it, the PCM 
> sliders are
> > > > actually controlling the correct physical L & R outputs on the 
> card, but
> > > > the alsa_pcm:playback signals being routed to those physical channels
> > > > are reversed, make sense?
> > > >
> > > > Is this a bug?  If so, what is the proper place/way to report it so 
> that
> > > > I don't use any more of your time?  Has it already been fixed?  I know
> > > > you mentioned v008 still had problems...
> > > >
> > > >
> > >
> > > ISTR that this is a hardware bug, and the kX driver works around it.  I
> > > will look into it.
> > >
> > > Are the channels reversed if you use JACK with the default stereo PCM,
> > > hw:0,0?
> > >
> > >
> > Using hw:0,0 for BOTH the Input & Output devices in QJackCtl:
> > (actually, input probably doesn't matter, huh?)
> > playback_1 maps to Stereo Output #1 RIGHT and volume is controlled by
> > the PCM LEFT slider
> > playback_2 maps to Stereo Output #1 LEFT and volume is controlled by
> > the PCM RIGHT slider
> >
> > In this case, it seems that slider is controlling the correct
> > playback_x stream, but the stream is mapped to the wrong stereo
> > channel of the soundcard.
>
>This does look like an upstream bug.  Can you verify that the channels
>are still reversed without my patch?
>
>Lee


Lee,

Should I report this issue on the ALSA bugtracking system or is it already 
being worked on?

Rick



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: potential emu10k1 bug/issue for Audigy2 ZS Platinum Pro

[Lee Revell]

On Sat, 2005-02-12 at 10:12 -0500, Rick Wright wrote:
> At 05:12 PM 2/10/2005 2/10/2005, Lee Revell wrote:
> >On Thu, 2005-02-10 at 16:20 -0500, Rick Wright wrote:
> > > > > The real problem is that the PCM output volume control sliders are
> > > > > backwards (or perhaps the remapping the channels to L/R, rather 
> > than R/L
> > > > > will fix this).  For example: PCM Front L slider controls the 
> > volume of
> > > > > playback_9 which is the RIGHT Front signal, PCM Front R slider 
> > controls
> > > > > playback_10 which is the LEFT Front signal.  This is true for all 
> > of the
> > > > > Stereo Output channel pairs.  If you think about it, the PCM 
> > sliders are
> > > > > actually controlling the correct physical L & R outputs on the 
> > card, but
> > > > > the alsa_pcm:playback signals being routed to those physical channels
> > > > > are reversed, make sense?
> > > > >
> > > > > Is this a bug?  If so, what is the proper place/way to report it so 
> > that
> > > > > I don't use any more of your time?  Has it already been fixed?  I know
> > > > > you mentioned v008 still had problems...
> > > > >
> > > > >
> > > >
> > > > ISTR that this is a hardware bug, and the kX driver works around it.  I
> > > > will look into it.
> > > >
> > > > Are the channels reversed if you use JACK with the default stereo PCM,
> > > > hw:0,0?
> > > >
> > > >
> > > Using hw:0,0 for BOTH the Input & Output devices in QJackCtl:
> > > (actually, input probably doesn't matter, huh?)
> > > playback_1 maps to Stereo Output #1 RIGHT and volume is controlled by
> > > the PCM LEFT slider
> > > playback_2 maps to Stereo Output #1 LEFT and volume is controlled by
> > > the PCM RIGHT slider
> > >
> > > In this case, it seems that slider is controlling the correct
> > > playback_x stream, but the stream is mapped to the wrong stereo
> > > channel of the soundcard.
> >
> >This does look like an upstream bug.  Can you verify that the channels
> >are still reversed without my patch?
> >
> >Lee
> 
> 
> Lee,
> 
> Should I report this issue on the ALSA bugtracking system or is it already 
> being worked on?
> 

Please file a bug report.  We will have to work around this behavior in
the driver.

The next step is to figure out exactly which devices are affected.

Lee



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click