Re: [Alsa-user] aloop questions and a possible suggestion

[Clemens Ladisch]

Xover wrote:
> Clemens Ladisch  googlemail.com> writes:
>> Xover wrote:
>>> Clemens Ladisch  googlemail.com> writes:
 It is intended to be used when there is no actual hardware device.
>>>
>>> By way of some background, what is the snd-loop rate shift facility intended
>>> to be used for?
>>
>> To synchronize with some other clock.
>>
>> (Which is what you were doing; but a plugin doesn't require
>> synchronization to begin with.)
>
> The thing is, I see people using the snd-aloop driver as a means to tee off
> audio between an application and a sound card, or even as a means to
> connect one audio source to two sound cards, all over the place. Sometimes
> they set the rate shift manually in an attempt to prevent the inevitable
> drift problems, but of course without 'closing the loop' this will always
> result in disaster eventually.
>
> In the simple requirement of needing to record the output of an audio
> player app while at the same time listening to it on a real (playback-only)
> sound card, what should they do?

Use PulseAudio.  ;-)

Or the file plugin.  (Which isn't easy to use.  It uses either a fixed file
name, or an external script which you have to write yourself.)


Regards,
Clemens

--
___
Alsa-user mailing list
Alsa-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-user

Re: [Alsa-user] RME HDSP9652 Routing problem @ 96khz

[Markus Seeber]

On 10/30/2015 12:29 PM, Klaus Löhmann wrote:
> Thanks for your answer.
> I use alsa 1.0.25
> 
> In the link you see that a channelmap exists for the hdsp9632 but not
> for the hdsp9652.
> The missing part is:
> 
> char channel_map_h9632_ds
> [12]
> = {
>   0, 1, 2, 3, 8, 9, 10, 11, 12, 13, 14, 15};
> 

OK i think I know what you mean, there are destination maps for the 9652
but no channel maps.

> As i said, i don't know if that is the problem.
> 
> Imagine you have a 12-track project with someone speaking "one one
> one..." on track 1, "two two two..." on track 2 and so on. Imagine also
> that your output-routing is straight. Track 1 > Out 1, Track 2 > Out 2,
> .
> When you open hdspmixer while jack is working at double speed (88,2 -
> 96), what happens is that:
> Track 1 is not routed to any out, Track 2 is routed to Out 1, Track 3
> not routed, Track 4 routed to Out 2,
> Track 5 is not routed, Track 6 is routed to Out 5, Track 7 is not
> routed, Track 8 is routed to Out 6, and so on...
> 
> I actually lose my (physical) outs 3,4,7,8,11,12.
> There is no way to get them adressed.

That seems strange... I can't answer that now, maybe the maintainers of
hdspmixer can help out here.





--
___
Alsa-user mailing list
Alsa-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-user

Re: [Alsa-user] RME HDSP9652 Routing problem @ 96khz

[Klaus Löhmann]

Thanks for your answer.
I use alsa 1.0.25

In the link you see that a channelmap exists for the hdsp9632 but not 
for the hdsp9652.

The missing part is:

char channel_map_h9632_ds 
[12] 
= {

  0, 1, 2, 3, 8, 9, 10, 11, 12, 13, 14, 15};

As i said, i don't know if that is the problem.

Imagine you have a 12-track project with someone speaking "one one 
one..." on track 1, "two two two..." on track 2 and so on. Imagine also 
that your output-routing is straight. Track 1 > Out 1, Track 2 > Out 2, 
.
When you open hdspmixer while jack is working at double speed (88,2 - 
96), what happens is that:
Track 1 is not routed to any out, Track 2 is routed to Out 1, Track 3 
not routed, Track 4 routed to Out 2,
Track 5 is not routed, Track 6 is routed to Out 5, Track 7 is not 
routed, Track 8 is routed to Out 6, and so on...


I actually lose my (physical) outs 3,4,7,8,11,12.
There is no way to get them adressed.

Thanks


Le 29/10/2015 21:55, Markus Seeber a écrit :

On 10/29/2015 02:16 PM, Klaus Löhmann wrote:

Hello,

first post here, so i don't know if i do everything like expected.
When i open hdspmixer @ 96k, the output routing is strange:
1st channel doesn't output, channel 2 outputs on out1,
CH 3 doesn't output, CH 4 outputs on out 2 ...and so on.

Yes this should be correct. ADAT optical does only support up to 48KHz
per channel, thus the card switches on 96KHz into double speed mode,
which means, that the data for a single channel is transmitted over two
channels. This means that the channel count is cut to half.


I opened a topic on the RME-Linux-Forum a while ago:
https://www.forum.rme-audio.de/viewtopic.php?id=17373

While regarding the sources, i realized (i am at no means a
coder/dev/whatever, so...)
that in the  hdspmixer the channelmap is missing for the HDSP9652, but
for the other cards (Digiface, Multiface, RPM...) Channelmpas are specified.

Maybe you are using an old version? I found a matching mapping right
here for alsa-tools version 1.0.29 :
https://fossies.org/dox/alsa-tools-1.0.29/channelmap_8cxx_source.html#l00099


--
___
Alsa-user mailing list
Alsa-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-user



--
___
Alsa-user mailing list
Alsa-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-user

Re: [Alsa-user] aloop questions and a possible suggestion

[Xover]

Clemens Ladisch  googlemail.com> writes:

> Use PulseAudio.  
> 
> Or the file plugin:
> http://www.alsa-project.org/alsa-doc/alsa-
lib/pcm_plugins.html#pcm_plugins_file
> (Which isn't easy to use.  It uses either a fixed file name, or an external
> script which you have to write yourself.)

Many thanks Clemens. From what you are saying, there might indeed be a use 
for an ALSA loopback driver that simply relays the 'drain count' to the 
playback side..? Or could PulseAudio be guaranteed to perform this function 
without resampling? Maybe most people aren't worried about the need to 
resample, but I see it as quite important even if only to save CPU power.



--
___
Alsa-user mailing list
Alsa-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-user