I did some measurements this afternoon after I got home from work.
Using a naive test loop, I was able to observe roughly 11.5M
multiply-accumulates per second.
That’s not enough for any serious wideband work, but it’ll do for simple
audio processing.
I’m going to muck about with my VLF SID receiver flowgraph and see if I
can make it run on
the Sheeva without too many aO and with it actually producing some
useful output.
Cheers
–
Marcus L.
Principal Investigator
Shirleys Bay Radio Astronomy Consortium