linuxaudio.org

On Apr 1, 2011, at 2:13 PM, Kjetil S. Matheussen wrote:

quoted text
>

> Peder Hedlund:

>>

>>> On 04/01/2011 04:34 AM, Robin Gareus wrote:

>>>>> Robin Gareus wrote:

>>>>>

>>>>>> Hya,

>>>>>>

>>>>>> I've just brought jack-stdout up to scratch.

>>>

>>> This should definitely be a part of the jack default toolset.

>>

>> Not that I don't appreciate the work behind jack-stdin/out (and I

>> wouldn't mind having them included in the jack toolset) but these

>> things can also be performed using for instance mplayer and

>> jack_capture :

>>

>> sox cdda.wav -t wav - tremolo 5 100 | mplayer -cache 26000 -ao jack -

>> and

>> jack_capture --channels 2 --port system:playback_1 --port system:playback_2 \

>> -ws | oggenc -R 44100 -B 16 -C 2 - -o file.ogg

>>

>

> This command does exactly the same:

>

> jack_capture -c 2 -ws | oggenc -R 44100 -B 16 -C 2 - -o file.ogg

>

> although this command is strongly preferable (no bits lost in piping):

>

> jack_capture -c 2 -f ogg file.ogg

> 

Sure, this is un*x. There's always at least two ways to get the job done :-)
> However, for piping jack-stdout just got options for 24 bits, while jack_capture still only supports 16 bits.
See the announce from today. jack-stdio supports 8/16/24/32 bit integer, both big and little endian, as well as 32 bit float (both native-endian as well as swapped endianess).
@Kjetil:

Does jack_capture buffer the [processed] data before I/O? jack-stdio does not.

jack-stdin/out reads/writes one sample at a time, and thus is relatively heavy on the CPU

 (3-5% on a 2GHz machine - quite a lot for a such a small tool).
> The reason why jack_capture lags a bit behind here is probably because jack_capture used code from jack-stdout to implement writing to stdout. :-)
Before you look into merging it: I just have spotted a bug with endianess for PPC/ big-endian architectures.

all the #ifdef __BIG_ENDIAN__ code can go away:  The compiler (at least gcc) does swap

constants 0x00ff <> 0xff00 and indices for uint8_t.
I'll have access to a PPC machine on Monday and will fix and test then.
Cheers!

robin
> _______________________________________________

quoted text
> Linux-audio-user mailing list

> Linux-audio-user@lists.linuxaudio.org

> http://lists.linuxaudio.org/listinfo/linux-audio-user

_______________________________________________

Linux-audio-user mailing list

Linux-audio-user@lists.linuxaudio.org

http://lists.linuxaudio.org/listinfo/linux-audio-user