On Fri, 12 May 2000 11:48:51 -0400, Jeff Sutherland wrote:
> You must be using floating point in your encoder, correct? 600/1 would
> be something like the speed difference between a 450MHz pentium with
> hardware fpu and the floating point emulation of the StrongARM. Caveat: if
> you need to do a lot of floating point, you picked the wrong part. If you
> can get away with integer math, a SA1110 running full tilt at 221MHz is
Or fixed point math. BTW: We made some minor modifications (read: added 3
lines) to the kernel that allow read-buffer access from userland. JDB is
currently using it to squeeze some more speed out of his (already fast)
FFT implementation. I'll send the read buffer stuff to Nico when I have a
more or less complete patch ready.
> comparable to a K6-233, in some tests faster than a K6. Nicolas Pitre wrote
> an MP3 _decoder_ that uses fixed point math. On my 275MHz Netwinder it uses
> about 1% of the system bandwidth while playing MP3's. BTW system clock
I'm not completely sure about that 1%. Rob Leslie pointed out that "time"
or "top" don't show the correct values with multi-threaded programs.
> speed is set by the boot loader, NOT by anything in the kernel. We don't
> use Angelboot so sorry, can't help you there...
> ----
>
> 3. Is there a driver for the UDA 1341 audio codec?
>
> ----
> Any ideas, folks? I'll be looking for one meeself soon...
Nico is working on one. His latest patch contains a framework, but the
comment says that it will crash your kernel.
Erik
--
J.A.K. (Erik) Mouw, Information and Communication Theory Group, Department
of Electrical Engineering, Faculty of Information Technology and Systems,
Delft University of Technology, PO BOX 5031, 2600 GA Delft, The Netherlands
Phone: +31-15-2783635 Fax: +31-15-2781843 Email: [EMAIL PROTECTED]
WWW: http://www-ict.its.tudelft.nl/~erik/
unsubscribe: body of `unsubscribe linux-arm' to [EMAIL PROTECTED]
++ Please use [EMAIL PROTECTED] for ++
++ kernel-related discussions. ++
