"Erik E. Fair" <fair@clock.org>  wrote:
 > Me, too, but my impression from reading this list was that the problem was
 > the arm32 port's interrupt structure, and resulting interrupt response
 > latency leading to an inability to feed the shark sound chip at 44.1 KHz,
 > not lack of FP performance. See Charles Hannum's messages from August...
 > 
 > This is why I inquired about that issue recently.

I've successfully played a pcm file directly from across the net on my 
NFS server and it played well... Its just mp3 files that I've had 
problems with...

'course, I only did this once and that was before the interrupt 
rototilling... I haven't tried mp3 since then.