Schurmann Alfred

On the minimum error in addition processes of positive floating-point numbers