This is all very interesting!

To get back to my initial problem, i managed to solve it by launching Pd with a longer audio buffer (-audiobuf 100). For some reason Pd came with a default buffer of 25ms.

Now, I can't tell you whether it sounds as "good" or worse than the same audio data played in aplay. I'll need to test this further.

Cheers,

Pierre.

2012/8/29 Tedb0t <lists@liminastudio.com>
Ah, wow, that's crazy.  Has anyone yet tried out any USB audio interfaces on the RPi yet?  I have one handy that I can try but not until I get these client projects out of the way ;)

I'm extremely curious if using a supported external audio interface solves all the sound problems or if there are also still alsa issues.  I'm also now wondering who will be the first to come out with an RPi audio "shield" ;)

This renews my interest in modifying the RPi design to make a Pd-friendly board that's intended to be headless with a built-in quality DAC.  But without RPi's supply chain hookups, it'd be quite hard to keep the price down to the RPi's level!

—t3db0t

On Aug 29, 2012, at 1:08 PM, Andy Farnell wrote:

>
>
> Yes I heard that too. It's single bit PWM modulated output.
> Basically there's no DAC, so audio output is a fun hack,
> and likely as good as it's worth trying to make it for
> simple applications.
>
> If you want good quality audio I/O there are things like
> Turtle Beach Amigo for about 15 quid
> http://www.turtlebeach.com/products/sound-cards/audio-advantage-amigo-ii.aspx
>
> You can take it out the board from its plastic housing. Its about
> as small as an SD card.
>
> a.
>
> On Wed, Aug 29, 2012 at 04:40:44PM +0200, Pierre Massat wrote:
>> There's a guy called "dom" on the RPi forum who seems to be involved in
>> audio development. I've seen his interventions in a couple of audio related
>> threads, and i got the information about the analog output being simple
>> filtered PWM off one of his posts.
>> Might be interesting to get in touch with him.
>>
>> Pierre.
>>
>> 2012/8/29 Tedb0t <lists@liminastudio.com>
>>
>>> This is what I'm curious about; who is doing this alsa development, where,
>>> and can we pitch in?  We are all interested  in getting the sound working
>>> better on the Pi...
>>>
>>> That said, it seems like the jury is out on if it's an alsa problem or a
>>> Pd problem.  I seem to have experienced both.
>>>
>>> —Tedb0t
>>>
>>> On Aug 29, 2012, at 6:46 AM, geoffroy wrote:
>>>
>>> With the PI the alsa is in alpha development on the arm platform and
>>> therefor won't be at it's best for a little while. The floating point
>>> integration in the latest system made the audio a lot better, as on the
>>> first gen OS of the PI the audio was just impossible.
>>>
>>>
>>>
>>> _______________________________________________
>>> Pd-list@iem.at mailing list
>>> UNSUBSCRIBE and account-management ->
>>> http://lists.puredata.info/listinfo/pd-list
>>>
>>>
>
>> _______________________________________________
>> Pd-list@iem.at mailing list
>> UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
>


_______________________________________________
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list