Re: sample rates

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: sample rates

Jussi Laako-2
Oswald Berthold wrote:
> is there anything speaking against using jack with absurdly high sample
> rates? i was playing around with the bt878 adc the other day at rates of
> 448000 and 896000 Sps, which the alsa driver supports but starting jack
>
> and the lower one respectively made my box just drop unconscious. thanks for
> any comments.

Possibly it would work, but probably requires big period size to be
used. Something like "-p 16384 -n 2" at least, otherwise interrupt rate
will be too high.


        - Jussi

-------------------------------------------------------------------------
SF.Net email is sponsored by: The Future of Linux Business White Paper
from Novell.  From the desktop to the data center, Linux is going
mainstream.  Let it simplify your IT future.
http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
_______________________________________________
Jackit-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/jackit-devel
Reply | Threaded
Open this post in threaded view
|

Re: sample rates

oswald berthold
On Tue, Dec 04, 2007 at 03:57:55PM -0500, Paul Davis wrote:

> On Tue, 2007-12-04 at 21:18 +0100, Oswald Berthold wrote:
> > hello
> >
> > is there anything speaking against using jack with absurdly high sample
> > rates? i was playing around with the bt878 adc the other day at rates of
> > 448000 and 896000 Sps, which the alsa driver supports but starting jack
> > with
> >
> > jackd -d alsa -d hw:1,1 -r 896000
> >
> > and the lower one respectively made my box just drop unconscious. thanks for
> > any comments.
>
> it will work if you pick a period size that doesn't translate into 1
> interrupt every 10 CPU cycles .... try -p 10240
On Wed, Dec 05, 2007 at 12:02:42AM +0200, Jussi Laako wrote:
>
> Possibly it would work, but probably requires big period size to be
> used. Something like "-p 16384 -n 2" at least, otherwise interrupt rate
> will be too high.

holla

thanks for quick reply.

the driver (the hardware?) doesn't allow setting of period size and jack
wouldn't start with any -p value other than 1024.

regardless, in my second attempt it all just worked, even up to and
including baudline happily crunching on the 448 kHz wide spectrogram.

the -n value doesnt affect the frequency of process() calls if i read the
manpage correctly(?). maybe it has to do with realtime priority 0/1 but i
need to test that again.

cheers, opt



--
>          <          D          V          .          O          r          g                    
gpg-key at http://xdv.org/~opt/60B2DA43.asc

-------------------------------------------------------------------------
SF.Net email is sponsored by: The Future of Linux Business White Paper
from Novell.  From the desktop to the data center, Linux is going
mainstream.  Let it simplify your IT future.
http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
_______________________________________________
Jackit-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/jackit-devel

attachment0 (196 bytes) Download Attachment