Nik Clayton <nclayton@lehman.com>  wrote:
 > On Fri, Jul 09, 1999 at 10:59:58AM +0100, Steve Woodford wrote:
 > > as they depend on fast floating point. Since the StrongARM has no
 > > hardware floating point, performance will suck big time. 
 > 
 > Even on a 220MHz machine?

Well, I can tell you from experience that it doth verily suck on a 233Mhz
Shark... I bet a 220Mhz machine would suck about 8% more. :)

I have my shark plugged into my TV running gqmpeg with mpg123 in the
background and my mp3's NFS mounted from my sparc downstairs.  It's
basically just fine unless you do even a single _other_ thing on the
shark like bring up the playlist window and start browsing for files;
at which point you'll start to generate gaps in the music... Also, if
you cause gqmpeg to keep up to a lot of mouse rollovers on buttons, 
you'll drain the audio queue and get more gaps...

 > > There are several
 > > fixed point MP3 implementations knocking around, but I haven't seem
 > > them in source form. (I assume Empeg's MP3 algorithm is fixed point).

The mpg123 I have, I can't distribute however I'm told that another NetBSD
sociallite in .au has fixed-pointed mpg123 for the shark and made it
work... I'm not sure if he plans or is able to distribute source.

(sorry if this has come up before, I've not been following this thread).