i think it has something to do with the intra-processor 
representation of floating point numbers, maybe the intermediate 
results are rounded or truncated. BUT for sure, if you replace this 
line:
>       float x,y,z1,z2,i,j;
WITH THIS:
      double x,y,z1,z2,i,j;
i bet it will work as u can't imagine! Try it out and see.


Reply via email to