32-bit trig operations are likely faster on Cortex M4F, given the
instruction set. On X86_64 you are going to have to try to get them
compiled to one of the SSE instruction sets if you don't want operations
to happen in wider intermediary registers and get converted going in and
out. The applicable flags may be something like "-m32 -mfpmath=sse
-Ofast -flto -march=native".
Thanks
Bruce
------------------------------------------------------------------------------
Slashdot TV.
Video for Nerds. Stuff that matters.
http://tv.slashdot.org/
_______________________________________________
Freetel-codec2 mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
