On Sat, 14 Jul 2007 22:01:36 +0400
Vladimir <kv11111@xxxxxxx> wrote:
> Hello,
>
> thanks for your response.
>
> stan wrote:
> > I think you are asking if the quality of the sound will be the same
> > if routed to the back channels of the card.
> Yes, I should formulate my questions better next time.
>
> > I just looked at
> > the .asoundrc at the link and it looks to me like the output will be
> > the same as long as you use this version that takes front
> > channels and routes them to the back channels at full volume.
> > # upmixing:
> > pcm.ch51dup {
> > type route
> > slave.pcm dmix6
> > slave.channels 6
> > ttable.0.0 1
> > ttable.1.1 1
> > ttable.0.2 1 <--- routing to rear
> > ttable.1.3 1 <--- at full volume
> > ttable.0.4 0.5
> > ttable.1.4 0.5
> > ttable.0.5 0.5
> > ttable.1.5 0.5
> > }
> >
> > It does some unnecessary mixing for your case, but you won't hear
> > the difference, it will only make the cpu work a little harder.
> >
> > There is one caveat: That the plugins don't alter the quality of
> > the sound as it passes through them.
> I'm not fully understand this sentence. Why this is a caveat ?
Well, the mixing might introduce distortion because it is done in
software instead of being directly from the hardware. It shouldn't
but might.
>
> > If you feel like experimenting you could change the ch51dup
> > to ch20dup as follows:
> > # upmixing:
> > pcm.ch20dup {
> > type route
> > slave.pcm dmix6
> > slave.channels 6
> > ttable.0.2 1
> > ttable.1.3 1
> > }
> >
> > Beyond that you could change the dmix6 to use only 4 channels, 0,
> > 1, 2, 3.
> I'll try this. Can I also just skip dmix6 and route the sound
> directly to hardware ?
>
I'm not sure, but I don't think so. I think the channels are being
defined by the dmix6 plugin, and so can't be used unless it is there.
I'm not familiar enough with the plugin system to know if there are
default channels defined, the 0-5 of the above ch51dup. Just drop it
and leave the channels slave there and it might be enough to try
# upmixing:
pcm.ch20dup {
type route
slave.channels 4
ttable.0.2 1
ttable.1.3 1
}
This will be true if the channels slave is defining the channel
numbers. It might be, and if this works it would definitely be more
robust and less cpu load than the first version. It should give you
the identical output that you would get on the front channels.
Predefined channels would only require
# upmixing:
pcm.ch20dup {
type route
ttable.0.2 1
ttable.1.3 1
}
Just try them, you won't hurt anything, and then you'll find out which
one is best for you. It would be good if you posted your results back
here so other people looking for a solution gain the benefit of your
experience. :-)
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Alsa-user mailing list
Alsa-user@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/alsa-user
