Collision detection: a comparison.

[Dar Scott]

On Monday, September 30, 2002, at 01:35 AM, malte brill wrote:

> thanks for the leg up. Indeed, I checked the whole data after 
> every move,
> silly me.

The more I think about this, try optimizing a simple pixel overlap 
in the overlapped region of the two images.  That may be enough.  A 
quick single pixel test might be to call them overlapped if both 
are after "z"; that avoids conversion and math.  Optimize the 
subrow comparison.

You might need to go to the subrect method we talked about, but 
maybe not.

Dar Scott