The decimal has a capacity for being more precise. It's "better" while taking up more memory as an expense for being so. There is also a type called a float which is the same sort of thing yet crummier (less memory, less impressive) than either the decimal or the double. For more see:
No comments:
Post a Comment