>Hi all,
>
>maybe a stupid one, but how would you detect an overflow condition after
>_signed_ 16 bit addition and subtraction? I cannot come up with any elegant
>soulution. TIA.
>
>
>Josef

Use 24 bit signed math and check for overrange, ie if the
answer falls outside +-32767 then it's an overflow.


Ray Gardiner (DSP Systems) ray@dsp-systems.com http://www.dsp-systems.com
private email to:- ray@netspace.net.au