You could create invisible mask objects to handle the collisions separately like obj_weightcol_L and obj_weightcol_R and set their x and y accordingly in a Step event, perhaps.
The issue would be if you had more than one weight at a time on screen, though, and there's probably an easier way to handle this.
The issue would be if you had more than one weight at a time on screen, though, and there's probably an easier way to handle this.
Â