[linux-audio-dev] Mixing signals

Steve Harris S.W.Harris at ecs.soton.ac.uk
Mon May 23 09:24:24 UTC 2005


On Mon, May 23, 2005 at 11:03:30 +0200, Richard Spindler wrote:
> On 5/23/05, Viceic Predrag <viceic at net2000.ch> wrote:
> > Could someone please help with this apparently simple problem?
> 
> I'm not a "professional" either, but this is what I do:
> 
> unsigned int TLData::MixChannels(float *A, float *B, float* out,
> unsigned int   count)
>   /*Mix function for (-1)-(1) float audio*/
> {
>         unsigned int i;
>         float *p_output = out;
>         float *p_A = A;
>         float *p_B = B;
>         for ( i=count; i > 0;i--){
>                 if (*p_A<0 && *p_B<0) {
>                         *p_output =(*p_A+1)*(*p_B+1)-1;
>                 } else {
>                         *p_output =2*(*p_A+*p_B+2)-(*p_A+1)*(*p_B+1)-3;
>                 }
>                 p_output++;
>                 p_A++;
>                 p_B++;
> 
>         }
>         return count;
> }

I've no idea what your trying to do cos I missed the original post, but
just FYI, you will generally be better off doing x[i] rather than *x and
incrementing x. Modern C compilers (eg. gcc 4) can vectorise array
operations as long as you dont obfuscate them with pointer maths.

- Steve



More information about the Linux-audio-dev mailing list