Yes, it should be.  If you look at the file (I think from memory) sound/oss/ac97_codec.c there are only a small handful of functions which are used to customize for specific hardware.  You can look at how I did the ucb1400 stuff for an example which seems to do more than any of the other chips there.


On May 20, 2005, at 12:35 PM, Jérôme Multrier wrote:

effectively, I tried our code with audiostix, and there is the same problem ...
But one thing different : madplay doesn't say that it needs to resample my little mp3
Another thing : a 400*.patch defines 5~6 ac97 chips ... perhaps adding our chip may be easy ...? ;)

2005/5/20, Craig Hughes <>:
Ah, ok -- didn't realize this wasn't and audiostix.  I haven't looked
at the 4546 driver code.  I know that natively the UCB1400 that
audiostix uses doesn't do anything but 48kHz, but bplay somehow seems
to be converting on the fly to 48kHz is you specify -b on the command


On May 20, 2005, at 7:21 AM, Cyril BAZIN wrote:

> Sorry Craig,
> I work with Jerome on the same project...
> We don't need to record sound, but to play sound...
> Our problem is the following:
> 1/ When we try to play a mp3 file sampled at 22050Hz with madplay,
> madplay says it is not possible to change output rate to 22050hz
> but he can resample the mp3 to play it correctly.
> 2/ When we try to play a wav file sampled at 22050Hz with bplay or
> our application, the rate couldn't be changed and they try to play
> the wav file at 48000Hz (that is not very audible... )
> The problem may be that we have build our own daughter card using
> the following AC'97 codec chip: LM4546A from National Semicondustor .

This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
gumstix-users mailing list